23219.compressor handbook principles and practice

Compressor Handbook:Principles and Practice

“This page left intentionally blank.”

Compressor Handbook:Principles and Practice

ByTony Giampaolo, MSME, PE

Library of Congress Cataloging-in-Publication Data

Giampaolo, Tony, 1939- Compressor handbook: principles and practice/by Tony Giampaolo. p. cm. Includes index. ISBN-10: 0-88173-615-5 (alk. paper) ISBN-10: 0-88173-616-3 (electronic) ISBN-13: 978-1-4398-1571-7 (Taylor & Francis : alk. paper) 1. Compressors--Handbooks, manuals, etc. I. Title.

TJ990.G53 2010 621.5’1--dc22 2010008818

Compressor handbook: principles and practice by Tony Giampaolo©2010 by The Fairmont Press. All rights reserved. No part of this publication may be re-produced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher.

Published by The Fairmont Press, Inc.700 Indian TrailLilburn, GA 30047tel: 770-925-9388; fax: 770-381-9865http://www.fairmontpress.com

Distributed by Taylor & Francis Ltd.6000 Broken Sound Parkway NW, Suite 300Boca Raton, FL 33487, USAE-mail: [email protected]

Distributed by Taylor & Francis Ltd.23-25 Blades CourtDeodar RoadLondon SW15 2NU, UKE-mail: [email protected]

Printed in the United States of America10 9 8 7 6 5 4 3 2 1

0-88173-615-5 (The Fairmont Press, Inc.)978-1-4398-1571-7 (Taylor & Francis Ltd.)

While every effort is made to provide dependable information, the publisher, authors, and editors cannot be held responsible for any errors or omissions.

Pages 221-232: Compressor specifications in Appendix XX From the 2009 Compressor Tech-nology Sourcing Supplement, courtesy COMPRESSORTechTwo magazine, published by Diesel & Gas Turbine Publications. Current compressor information can be found at www.compressortech2 or www.CTSSNet.net.

v

Dedication

This book is dedicated to my five grandchildren:

Amanda Rose Anna Josephine Carly Paige Riley James Nickolas Anthony

They are the future.

vii

Contents

Preface .............................................................................................................xiAcknowledgements ....................................................................................... xiii

Chapter 1—Introduction .................................................................................1 History .......................................................................................................1

Chapter 2—General Compressor Theory ....................................................7 Thermodynamics of Compression ........................................................7

Chapter 3—Compressor Types ....................................................................15 Dynamic Compressors ..........................................................................15 Axial Compressors..........................................................................15 Centrifugal Compressors ...............................................................22 Variations in Compressor Design ................................................25 Positive Displacement Compressors ..................................................26 Blowers .............................................................................................26 Reciprocating Compressors ...........................................................29 Preliminary Selection and Sizing .................................................53 Screw Compressors ........................................................................58 Screw Compressor Control .................................................... 59

Chapter 4—Effect of Operating Conditions ............................................ 63 Effects of Temperature & Pressure ..................................................... 63 Effects of Compression Ratio .............................................................. 64 Effects of Specific Heat Ratio .............................................................. 65

Chapter 5—Throughput Control ................................................................. 73 Speed Control ......................................................................................... 73 Suction Throttling Control ................................................................... 73 Discharge Valve Throttling Control ................................................... 74 Recycle Valve Control ........................................................................... 74 Variable Volume Pocket Control ......................................................... 76

Chapter 6—Description of Surge ............................................................... 81 Surge & Stall ........................................................................................... 81

viii

Chapter 7—Surge Control ............................................................................ 85 Minimum Flow Control ........................................................................ 85 Maximum Pressure Control ................................................................. 85 Ratio Control ........................................................................................... 87

Chapter 8—Vibration ..................................................................................... 97 Rotor Response ....................................................................................... 97 Sources of Vibration ..................................................................................

Chapter 9—Valve Requirements ............................................................... 105 Valve Types ................................................................................................. Valve Trim ...................................................................................................

Chapter 10—Instrument Requirements ....................................................111 Sensor Types ............................................................................................... Speed of Response .....................................................................................

Chapter 11—Detectable Problems ............................................................ 115 Mechanical Problems ................................................................................. Electrical Problems..................................................................................... Performance Problems ..............................................................................

Chapter 12—Controlling Reciprocatng and Centrifugal Compressors in Identical Processes ........................................ 131

Chapter 13—Optimization & Revitalization of Existing Reciprocating Compression Assets ......................................... 145

Chapter 14—Piston Rod Run-out is a Key Criterion for Recip Compressors ...................................................................... 183

Chapter 15—Effect of Pulsation Bottle Design on the Performance of a Modern Low-speed Gas Transmission Compressor Piston ............... 193

Chapter 16—Resolution of a Compressor Valve Failure: A Case Study ................................................................................ 211

ix

APPENDIX A1 Compressor Manufacturers ......................................................... 221 A2 Comparison of Three Types of Compressors ..........................233 B1 List of Symbols..............................................................................234 B2 Glossary of Terms .........................................................................237 B3 Conversion Factors .......................................................................267 C Gas Processers Suppliers Association Select Curves & Charts................................................................272 D Classification of Hazardous Atmospheres ...............................310 E Air/Oil Cooler Specifications Check List ................................311 F Cylinder Displacement Curves ..................................................314 G Compressor Cylinder Lubrication .............................................319 H Troubleshooting Chart .................................................................322 I Typical Starting, Operating and Maintenance Procedures for a Reciprocating Compressor ...........................324 J Basic Motor Formulas and Calculations ..................................340

INDEX ..........................................................................................................353

xi

Preface

Compressors have played a major role in setting our standard of living and they have contributed significantly to the industrial revolu-tion. Early compressors like the bellows (used to stoke a fire or the water organ use to make music) marked the beginning of a series of compres-sion tools. Without compression techniques we could not have efficiently stabilized crude oil (by removing its trapped gasses) or separated the various components of gas mixtures or transported large quantities of gas cross country via gas pipelines. Today, compressors are so much a part of our every day existence that many of us do not even recognize them for what they are. Compressors exist in almost every business and household as vacuum cleaners and heating & air conditioning blowers. Even those who have worked with compressors (usually only one or two types of compressor) have only a vague awareness of the variety of compressors in existence today. It is always interesting to see how the inventive process takes place, and how the development process progresses from inception to final design. Therefore, included in some sections of the book is histori-cal information on the development of various compressors. Due to the number of different types of compressors it was too time consuming to research the origins of each compressor type. For the roots blower and screw compressor the inventive process is clear as discussed in Chapters 1 and 3. However, the origin of the reciprocating compressor is somewhat obscured. No doubt the water organ devised by Ctesibius of Alexandria paved the way. Nevertheless, using water to compress air in a water organ is a far cry from a piston moving within a cylinder to compress gas. True there is significant similarity between reciprocat-ing engines and reciprocating compressors: Just as there is similarity between turbo compressors and turbine expanders. Many engineers/technicians/operators spend their entire careers in one product discipline (manufacturing, maintenance, test, sales, etc…). Sometimes they have had the opportunity to work in several disciplines. This book is intended to assist in the transition from an academic back-ground to a practical field, or from one field to another. It will assist the reader in his day-to-day duties as well as knowing where to look for ad-ditional information. Also people respond better when they understand

xii

why they are asked to perform certain functions, or to perform them in a certain order. My intention is to provide a basic understanding of the variety of compressors. The need for this book has grown out of the request for seminars and training sessions from utilities and oil & gas companies. Most often these companies hire new employees or relocate and retrain their current employees. The reader may have some experience in the operation or maintenance of some compression equipment from previ-ous assignments. This book provides a practical introduction to dynamic and posi-tive displacement compressors, including compressor performance, op-eration and problem awareness. In reading this book the reader will learn what is needed to select, operate and troubleshoot compressors and to communicate with peers, sales personnel and manufacturers in the field of dynamic and positive displacement compressor applications. In addition to the theoretical information, real life case histories are presented. The book demonstrates investigative techniques to iden-tify and isolate various contributing causes such as: design deficiencies, manufacturing defects, adverse environmental conditions, operating er-rors, and intentional or unintentional changes of the machinery process that precede the failure. Acquiring and perfecting these skills will enable readers to go back to their workplace and perform their job functions more effectively. In addition to the content of this book the engineer/technician/operator will find that the information provided in the appendix will become a useful reference for years to come.

Tony GiampaoloWellington, FL

January 2010

xiii

Acknowledgements

I would like to recognize and thank the following individuals for their support and assistance in obtaining photographs for use in this book:

Norm Shade, President, ACI Services, Inc. Danny L. Garcia, Project Manager, Sun Engineering Services, Inc. Roger Vaglia, Product Manager (Retired), Cooper Ind., White Superior Division John Lunn Engineering Manager (Retired) Rolls Royce USA Everette Johnson, Engineering Manager, Cameron Compressor Corpora-tion Ben Suurenbroek (Retired), Cooper Energy Services—Europe Dave Kasper, District Manager, Dresser Roots, Inc.

I also want to acknowledge and thank Peter Woinich, Design Engineer, Construction Supervisor and Associate (Retired) of William Ginsberger, Associates for his help in proofreading this manuscript. Also I wish to acknowledge and thank the following companies for their confidence and support by providing many of the photographs and charts that are in this book.

ACI Services, Inc. Baldor Electric Company Cameron Compressor Corporation COMPRESSORTechTwo magazine, published by Diesel & Gas Turbine Publications Dresser Roots, Inc Gas Processors Suppliers Association MAN Turbo AG Oil & Gas Journal Penn Engineering Petroleum Learning Programs Rolls Royce USA Sun Engineering Services, Inc. United Technologies Corp, Pratt & Whitney Canada

Chapter 1

Introduction

HISTORY

The history of compressors is as varied as are the different typesofcompressors.Therefore it isfitting thatwefirst identify thedifferenttypesofcompressors.AsshowninChart1-1,compressorsfall intotwoseparateanddistinct categories:dynamicandpositivedisplacement.

Chart 1-1

Somewhere in antiquity the bellows was developed to increaseflowintoa furnace inorder tostokeor increase furnaceheat.Thiswasnecessary to smelt ores of copper, tin, lead and iron. This led thewaytonumerousother inventionsof tools andweapons. One of the earliest recorded uses of compressed gas (air) datesbackto3rdcenturyB.C.Thisearlyuseofcompressedairwasthe“waterorgan.” The invention of the “water organ” is commonly credited toCtesibiusofAlexandria1.TheconceptwasfurtherimprovedbyHeroofAlexandria(alsonotedfordescribingtheprinciplesofexpandingsteamto convert steampower to shaftpower).

1

2 CompressorHandbook:PrinciplesandPractice

The water organ consisted of a water pump, a chamber partlyfilledwithairandwater,arowofpipesontop(organpipes)ofvariousdiametersand lengthsplus connecting tubingandvalves.Bypumpingwater into the water/air chamber the air becomes compressed. Thanbyopeningvalves to specificorganpipes thedesiredmusical sound iscreated. Ctesibius also developed the positive displacement cylinder andpiston tomovewater. Itwasnotuntilthelate19thcenturythat many of these ideas were turnedintoworkinghardware. In the 1850s, while trying to finda replacement for the water wheel attheirfamily’swoolenmill,PhilanderandFrancisRootsdevisedwhathascometobe known as the Roots blower3. Theirdesignconsistedofapairoffigure-eightimpellersrotatinginoppositedirections.While some Europeanswere simultane-ously experimenting with this design,the Roots brothers perfected the designandput it into large-scaleproduction. It is not surprising that othercompressor designs followed power-

Figure 1-1. Water Organ De-veloped By Ctesibius.2

Figure 1-2. Photo Courtesy of Frick by Johnson Controls.

Introduction 3

producingdesigns.Forexample,thereciprocatingengineconcepteasilytransfers to the reciprocating compressor. The integral-engine-compressor is a good example as its designutilizes one main shaft connected to both the power cylinders andthe compression cylinders. The form and function of the compressorcylinders are the samewhether it is configured as an integral engine-compressor or a separable-compressor driven by an electricmotor, gasengineor turbine. Other examples are the centrifugal compressor, (Figure 1-4) theturbo-expander, the axial compressor, and the axial turbine (Figure 1-5and1-6). In 1808 JohnDumball envisioned amulti-stage axial compressor.Unfortunatelyhisideaconsistedonlyofmovingbladeswithoutstation-aryairfoils to turn theflow intoeach succeeding stage.4,5,6 Notuntil1872didDr.FranzStolzecombinetheideasofJohnBar-

Figure 1-3. Cooper-Bessemer Z-330 Integral Engine Compressors in Krun-mhorn, Germany. Courtesy of Ben Suurenbroek (Retired Cooper Energy Services)


ber and JohnDumball to develop the first axial compressor driven byan axial turbine.Due to a lack of funds, he did not build hismachineuntil1900.Dr.Stolze’sdesignconsistedofamulti-stageaxialflowcom-pressor,asinglecombustionchamber,amulti-stageaxialturbine,andaregeneratorutilizingexhaustgasestoheatthecompressordischargegas.Thisunitwastestedbetween1900and1904,butneverransuccessfully. Operating conditions have a significant impact on compressor

Figure 1-4. Barrel Compressor Cour-tesy of Rolls-Royce USA (formerly Coo-per Industries En-ergy Services).

Figure 1-5. Five Stage Power Tur-bine Rotor From RT15 Turbine De-signed For 12,000 RPM Courtesy of Rolls-Royce USA (formerly Cooper Industries Ener-gy Services).

Introduction 5

Figure 1-6. Courtesy of United Technologies Corporation, Pratt & Whitney, Canada. The ST-18 is a 2 Megawatt Aeroderivative Combin-ing Centrifugal Com-pressor & Axial Expan-sion Turbine.

selectionandcompressorperformance.Theinfluencesofpressure, tem-perature,molecularweight,specificheatratio,compressionratio,speed,vaneposition,volumebottles,loadersandunloaders,etc.areaddressedinthisbook.Theseconditionsimpactcompressorcapacityandthereforethe compressor selection. They also impact the compressor efficiency.Flexibilityinselectionisstillpossibletosomeextentascompressorscanbeoperatedinparallelandseriesmodes.Forexample,toachievehigherpressuresmultiplecompressorscanbeconfiguredinserieswherebythedischargeofonecompressor feedsdirectly into thesuctionofa secondcompressor,etc.Likewise,toachievehigherflowsmultiplecompressorscan be configured in parallel whereby the suction of each compressoris manifolded together and the discharge of each compressor is alsomanifolded together. DifferentmethodsofthroughputcontrolareaddressedinChapter5, suchas,discharge throttling, suction throttling, guidevaneposition-ing,volumebottles,suctionvalveunloadersandspeedcontrol;andhoweachof these controlmethodseffects compressor life. This book discusses different compressors; how they operate andhow theyare controlled. Since the cost ofprocessdowntimeanddam-age to a compressor can range from thousands to millions of dollars;the types of failures that can occur and how to avoid these failures isalsoaddressed in thisbook. In view of the fact that themost destructive event in a dynamiccompressorissurge,compressorsurgewillbedefinedanddiscussedindetail.Alsodiscussedarethevarioustypesof instrumentation(control-lers, valves, pressure and temperature transmitters, etc..) available and


whicharemostsuitableincontrollingsurge.Destructivemodesofothercompressorsarealsoaddressed. A few algorithms are presented, primarily in Chapters 4 and 7,tohelpdemonstrate interactionsofpressure, temperature andquantifyresults, but their understanding is not essential to the selection of theproper control scheme and instrumentation. The reader should not beintimidatedby these algorithms as theirunderstandingwill openup abroaderappreciationofhow the compressorworks.

Footnotes1 A History of Mechanical Inventions,AbbottPaytonUsher.ThisDover

edition,firstpublishedin1988,isanunabridgedandunalteredre-publicationoftherevisededition(1954)oftheworkfirstpublishedbyHarvardUniversityPress,Cambridge,MA, in1929.

2 Multiple sourceswere found for this sketch, none ofwhich refer-enceda source.

3 Initiative In Energy, The Story of Dresser Industries, Darwin Payne,1979

4 Engines—The Search for Power, JohnDay, 19805 The Gas Turbine,NormanDavy, 19146 Modern Gas Turbines,ArthurW. Judge1950

Chapter 2

General Compressor Theory

Compressors aremechanical devices used to increase the pres-sureofair,gasorvaporandintheprocessmoveitfromonelocationto another. The inlet or suction pressure can range from low sub-atmospheric pressure levels to any pressure level compatible withpiping and vessel strength limits. The ratio of absolute dischargepressuretoabsolutesuctionpressureisthecompressorpressureratio(CR—seeAppendixB2,GlossaryofTerms).Stagecompressionislim-ited to themechanical capabilities of the compressor and, generally,approachesaCRof4.Toachievehighpressuresmultiplestagesmustbe employed. Compressiontheory isprimarilydefinedbythe IdealGasLawsand theFirst&SecondLawsofThermodynamics.Asoriginallycon-ceivedtheIdealGasLawisbasedonthebehaviorofpuresubstancesand takes the following form:

Pv=RT (2-1)

Where P =Absolute Pressure v = SpecificVolume R =GasConstant T =Absolute Temperature

ThisequationisbasedonthelawsofCharles,Boyle,Gay-LussacandAvogadro (seeAppendixB2Glossary of Terms). Note all properties should be defined in the same measuringsystem(forexampleeither theEnglishsystemor themetricsystem).Conversion factors listed in Appendix B3 can be used to assist inobtaining consistent units. Table 2-1 sumsup the two systems.

7


Table 2-1.———————————————————————————————————Parameter Symbol EnglishSystem MetricSystem———————————————————————————————————Pressure P Absolute Pascalsor pressure (psia) Kilopascals———————————————————————————————————Temperature T Absolute Degrees temperature (oR) Kelvin (oK)———————————————————————————————————SpecificVolume v Cubic inches Cubic centimeters perpound pergramor cubic metersperkilogram———————————————————————————————————UniversalGas R 1545 ft-lbf/ 8.3144kNm/ Constant lbmoR kmol oK———————————————————————————————————

The idealgas lawcanbemanipulated toobtain severaluseful re-lationships.Bymultiplyingbothsidesof theequationbythemass“m”of thegas the specificvolumebecomes totalvolume:

V=mv

PV=mRT (2-2)

Considering that the mass of any gas is defined as the numberofmoles times itsmolecularweight than (seeAvogadro’sLaw inAp-pendixB2):

m=n×mw

and PV=n×mw×RT (2-3)

and R =mw×R (2-4)

WhereR is theuniversalgas constant

P1V1 P2V2 —— = nR = mR = —— (2-5) T1 T2

Thespecificgasconstantmaybeobtainedusingtheuniversalgas

GeneralCompressorTheory 9

constantandequation2-4above.However,Table2-2listthespecificgasconstant for someof themore commongases.

Table 2-2———————————————————————————————— SpecificGas Gas Formula Molecular Constant Weight ft× lbf ———— × °R lbm————————————————————————————————Helium He 4.003 386.2CarbonMonoxide CO 28.01 55.18Hydrogen H2 2.016 766.6Nitrogen N2 28.02 55.16Oxygen O2 32.00 48.29CarbonDioxide CO2 44.01 35.12SulfurDioxide SO2 64.07 24.12WaterVapor H2O 18.02 85.78Methane CH4 16.04 96.35Ethane C2H6 30.07 51.40Iso-butane C4H10 58.12 26.59————————————————————————————————

Dividingbothsidesby“time”thetotalvolumebecomesvolumetricflowandthemassflowperunit timebecomesthemassflowrate“W.”

PQ=WRT (2-6)

Where Q =VolumetricFlowRate W =MassFlowRate

A pure substance is one that has a homogeneous and constantchemicalcomposition throughoutallphases (solid, liquidandgas).Formost compressor applications amixture of gasesmay be considered apuresubstanceas longas there isnochangeofphase.Thesignificanceof introducing this concept is that the state of a simple compressiblepure substance isdefinedby two independentproperties. Anadditional termmaybe consideredat this time to correct fordeviations from the ideal gas laws. This term is the compressibility


factor “Z.” Therefore, the idealgas equationbecomes

Pv=ZRT (2-7)

and

PQ=ZWRT (2-8)

Compressorperformanceisgenerallyshownaspressureratioplot-tedagainstflow. (Note: it ismoreaccurate tousehead insteadofpres-sureratio,becauseheadtakes intoaccountthecompressibilityfactorofthegas,molecularweight,temperature,andtheratioofspecificheatofthe gas—and corrected flow—all at constant speed). This is discussedinmoredetail later in this chapter. Other relationships that arealsouseful are:

Reduced Temperature and Pressure

T Tr = — (2-9) Tc

P Pr = — (2-10) Pc

Where Tr =ReducedTemperature Pr =ReducedPressure Tc =CriticalTemperature Pc =CriticalPressure T =ObservedTemperature P =ObservedPressure

Partial Pressure The totalpressure is equal to the sumof thepartialpressures

P=P1+P2+P3+ ... (2-11)


This relationship isdefinedbyDalton’sLaw (seeAppendixB2). If thetotalpressureofthemixtureisknownthanthepartialpres-sure canbe calculated from themole fraction.

P=P1+P2+P3+ ... (2-12)

Themole fraction“x” is

M1 M2 M3 x1 = —— ; x2 = —— ; x3 = —— ; (2-13) Mm Mm Mm

Thepartialpressuremay thenbe calculatedas follows

P1=x1×P;P2=x2×P;P3=x3×P (2-14)

x1+x2+x3+=1.0 (2-15)

FirstLawofThermodynamics Thefirst lawof thermodynamics states thatenergycannotbecre-atedordestroyedbut it canbe changed fromone form toanother.

Qh=Ww+∆E (2-16)

Where Qh =Heat into the system Ww = Workby the system ∆E =Change in systemenergy

Note: the symbolW is used to denote work, whereas the symbolWindicatesweightflowrate.

SecondLawofThermodynamics The second lawof thermodynamics states that the entropyof theuniversealways increases.This is the same law that indicates thatper-petualmotionmachinesarenotpossible.

∆s≥0 (2-17)


Horse Power Calculations Thebrakehorsepower(BHP)requiredtodrivethecompressorcanbedeterminedby calculating the gas horsepower (GHP) and then cor-recting formechanical losses.

Hd×Wg GHP= ——————— (2-18) 60×33,000×Ep

Where BHP =Brakehorsepower Hd =Head (adiabatic) – ft-lb/lb Wg =Weightflowof thegas– lbs/hr Ep =Adiabatic efficiency

and SCFD×MW Wg= ——————— (2-19) 24×379.5

If capacity is availableGHPcanbe calculateddirectly.

GHP = 229 # Ep^ h

Q1 # P1 # Z1Zavc m#

kk–1

CR kk–1 –1J

L

KKKK

N

P

OOOO

Z

[

\

]]

]]

_

`

a

bb

bb

(2-20)

Where Q1 = Inlet cubic feet/minute (ICFM) P1 =Suctionpressure—psia CR =Compression ratio K =Specificheat ratio Zav =Averagegas compressibility factor Z1 =Gas compressibility factorat compressor inlet

Thenbrakehorsepower is

BHP=GHPx (1+%MechanicalLosses) (2-21)


Each compressor typehas its ownunique characteristics thatwillbe covered in the section specific to that compressor type.

—————————Note: The physical and thermodynamic properties of many gases are provided in Appendix C, courtesy of the Gas Processors Suppliers Association (GPSA).

15

Chapter 3

Compressor Types

DYNAMIC COMPRESSORS

Two types of dynamic compressors are in use today—they are the axial compressor and the centrifugal compressor. The axial compressor is used primarily for medium and high horsepower applications, while the centrifugal compressor is utilized in low horsepower applications. Both the axial and centrifugal compressors are limited in their range of operation by what is commonly called stall (or surge) and stone wall. The stall phenomena occurs at certain conditions of flow, pressure ratio, and speed (rpm), which result in the individual compressor airfoils going into stall similar to that experienced by an airplane wing at a high angle of attack. The stall margin is the area between the steady state operating line and the compressor stall line. Surge or stall will be discussed in detail later in this chapter. Stone wall occurs at high flows and low pressure. While it is difficult to detect. Stone wall is manifested by increasing gas temperature.

Axial Compressors Gas flowing over the moving airfoil exerts lift and drag forces approximately perpendicular and parallel to the surface of the airfoil (Figure 3-1). The resultant of these forces can be resolved into two com-ponents:

1. the component parallel to the axis of the compressor represents an equal and opposite rearward force on the gas—causing an increase in pressure;

2. a component in the plane of rotation represents the torque required to drive the compressor.

16 Compressor Handbook: Principles and Practice

From the aerodynamic point of view there are two limiting factors to the successful operation of the compressor. They are the angle of at-tack of the airfoil and the speed of the airfoil relative to the approaching gas (Figure 3-2). If the angle of attack is too steep, the flow will not fol-low the concave surface of the airfoil. This will reduce lift and increase drag. If the angle of attack is too shallow, the flow will separate from the concave surface of the airfoil. This also results in increased drag.

Figure 3-1. Forces Acting on the Blades

Figure 3-2. Airfoil Angle of attack Relative to Approaching Air or Gas

Compressor Types 17

If the speed of the airfoil relative to the gas is too high, a shock will develop as the gas exceeds the speed of sound trying to accelerate as it passes around the airfoil. This shock will cause turbulent flow and result in an increase in drag. Depending on the length of the airfoil, this excessive speed could apply only to the tip of the compressor blade. Manufacturers have overcome this, in part, by decreasing the length of the airfoil and increasing the width (or chord). For single-stage operation, the angle of attack depends on the rela-tion of flow to speed. It can be shown that the velocity relative to the blade is composed of two components: the axial component depends on the flow velocity of the gas through the compressor, and the tan-gential component depends on the speed of rotation of the compressor (Figure 3-3). Therefore, if the flow for a given speed of rotation (rpm) is reduced, the direction of the gas approaching each blade is changed so as to increase the angle of attack. This results in more lift and pressure rise until the stall angle of attack is reached.

Figure 3-3. Velocity Component Relative to Airfoil

This effect can be seen on the compressor characteristic curve. The characteristic curve plots pressure against flow (Figure 3-4). The points on the curve mark the intersection of system resistance, pressure, and flow. (Note that opening the bleed valve reduces system resistance and moves the compressor operating point away from surge.) The top of each constant speed curve forms the loci for the compressor stall (surge) line. Therefore, the overall performance of the compressor is depicted on the compressor performance map, which includes a family of constant speed (rpm) lines (Figure 3-5). The efficiency islands are included to show the effects of operating on and off the design point. At the design speed and flow, the angle of attack relative to the blades is optimum


and the compressor operates at peek efficiency. If flow is reduced at a constant speed, the angle of attack increases until the compressor airfoil goes into stall. As flow is increased at a constant speed the compressor charac-teristic curve approaches an area referred to as “stone wall.” Stone wall does not have the dynamic impact that is prevalent with stall, but it is a very inefficient region. Furthermore, operation at or near stonewall will result in overtemperature conditions in the downstream process. From the mechanical point of view, blade stresses and blade vi-bration are limiting factors. The airfoil must be designed to handle the varying loads due to centrifugal forces, and the load of compressing

Figure 3-4. Compression System Curve

Figure 3-5. Compressor Performance Curve

Compressor Types 19

gas to higher and higher pressure ratios. These are conflicting require-ments. Thin, light blade designs result in low centrifugal forces, but are limited in their compression-load carrying ability; while thick, heavy designs have high compression-load carrying capability, but are limited in the centrifugal forces they can withstand. Blade vibration is just as complex. There are three categories of blade vibration: resonance, flutter, and rotating stall. They are explained here.

Resonance—As a cantilever beam, an airfoil has a natural frequency of vibration. A fluctuation in loading on the airfoil at a frequency that coincides with the natural frequency will result in fatigue failure of the airfoil.

Flutter—A self-excited vibration usually initiated by the airfoil ap-proaching stall.

Rotating Stall—As each blade row approaches its stall limit, it does not stall instantly or completely, but rather stalled cells are formed (see Rotating Stall Figure 6-1). Stall is discussed in more detail in Chapter 6.

The best way to illustrate flow through a compressor stage is by constructing velocity triangles (Figure 3-6). Gas leaves the stator vanes at an absolute velocity of C1 and direction θ1. The velocity of this gas relative to the rotating blade is W1 at the direction β1. Gas leaves the rotating stage with an absolute velocity C2 and direction θ2, and a rela-tive velocity W2 and direction β2. Gas leaving the second stator stage has the same velocity triangle as the gas leaving the first stator stage. The projection of the velocities in the axial direction are identified as Cx, and the tangential components are Cu. The flow velocity is represented by the length of the vector. Velocity triangles will differ at the blade hub, mid-span, and tip just as the tangential velocities differ. Pressure rise across each stage is a function of the gas density, ρ, and the change in velocity. The pressure rise per stage is determined from the velocity triangles: ρ ρ ∆P = —— (W1

2 – W22) + —— (C2

2 – C32) (3-1)

2gC 2gC

This expression can be further simplified by combining the differential pressure and density, and referring to feet of head.


µ ∆P Head = —— x ∆C = —— (3-2) gC ρ

where ∆P/ρ is the pressure rise across the stage and head is the pres-sure rise of the stage measured in feet head of the fluid flowing. The standard equation for compressor head is given below.

Head MW

Z T R 1ave s c=v-vc m

(3-3)

where Zave is the average compressibility factor of gas, and MW is the mole weight. Before proceeding further, we will define the elements of an airfoil (Figure 3-7).

Fixed Blade Row Turning Angle ≡ ε = β2’ - β1’xxMoving Blade Row Chord Length

Figure 3-6. Velocity Diagrams for an Axial Flow Compressor.

Compressor Types 21

Camber Line Leading EdgeCamber or Blade Angle ≡ β2 - β1 Trailing EdgeInlet Blade Angle ≡ β1 Maximum Thickness ≡ tExit Blade Angle ≡ β2 Width ≡ wInlet Flow Angle ≡ β1‘ Height ≡ hExit Flow Angle ≡ β2’ Aspect Ratio ≡ ratio of blade heightAngle of Deviation ≡ β2 - β2’ to blade chordStagger Angle ≡ γ Angle of Attack or angle ifPitch ≡ s incidence ≡ i = β1 - β1’

Figure 3-7. Elements of an Airfoil.


Centrifugal Compressors The centrifugal compressor, like the axial compressor, is a dynamic machine that achieves compression by applying inertial forces to the gas (acceleration, deceleration, turning) by means of rotating impellers. The centrifugal compressor is made up of one or more stages, each stage consisting of an impeller and a diffuser. The impeller is the rotating element and the diffuser is the stationary element. The impeller consist of a backing plate or disc with radial vanes attached to the disc from the hub to the outer rim. Impellers may be either open, semi-enclosed, or enclosed design. In the open impeller the radial vanes attach directly to the hub. In this type of design the vanes and hub may be machined from one solid forging, or the vanes can be machined separately and welded to the hub. In the enclosed design, the vanes are sandwiched between two discs. Obviously, the open design has to deal with gas leakage between the moving vanes and the non-moving diaphragm, whereas the enclosed design does not have this problem. However, the enclosed design is more difficult and costly to manufacture. Generally gas enters the compressor perpendicular to the axis and turns in the impeller inlet (eye) to flow through the impeller. The flow through the impeller than takes place in one or more planes perpen-dicular to the axis or shaft of the machine. This is easier to understand when viewing the velocity diagrams for a centrifugal compressor stage. Although the information presented is the same, Figure 3-8 demonstrates two methods of preparing velocity diagrams. Centrifugal force, applied in this way, is significant in the devel-opment of pressure. Upon exiting the impeller, the gas moves into the diffuser (flow decelerator). The deceleration of flow or “diffuser action”

Figure 3-8. Velocity Diagrams for a Centrifugal Compressor

Compressor Types 23

causes pressure build-up in the centrifugal compressor. The impeller is the only means of adding energy to the gas and all the work on the gas is done by these elements. The stationary components, such as guide vanes and diffusers, can only convert velocity energy into pressure en-ergy (and incur losses). Pressure from the impeller eye to the impeller outlet is represented by the following:

G GPm = —— [(C2

2–C12)+(U2

2–U12)+(W1

2–W22)] = —— (C2uU2–C1uU1) (3-4)

2gc 2gc

The term (C22 – C1

2)/2gc represents the increase in kinetic energy contributed to the gas by the impeller. The absolute velocity C1 (enter-ing the impeller) increases in magnitude to C2 (leaving the impeller). The increase in kinetic energy of the gas stream in the impeller does not contribute to the pressure increase in the impeller. However, the kinetic energy does convert to a pressure increase in the diffuser section. Depending on impeller design, pressure rise can occur in the impeller in relation to the terms (U2

2 – U12)/2gc and (W1

2 – W22)/2gc. The term

(U22 – U1

2)/2gc measures the pressure rise associated with the radial/centrifugal field, and the term (W1

2 – W22)/2gc is associated with the

relative velocity of the gas entering and exiting the impeller. The ideal head is defined by the following relationship:

1 Headideal = Pm = —— (C2

2 – C12)/2gc (3-5)

gc

Q Flow coefficient φ is defined as = —— (3-6) AU

where Q = Cubic feet per second (CFS) A = Ft2

U = Feet per sec

700Qrewriting this equation φ = ——— (3-7) D3N

where D is diameter, and N is rotor speed.


The flow coefficients are used in designing and sizing compres-sors and in estimating head and flow changes resulting as a function of tip speed (independent of compressor size or rpm). Considering a constant geometry compressor, operating at a constant rpm, tip speed is also constant. Therefore, any changes in either coefficient will be directly related to changes in head or flow. Changes in head or flow under these conditions result from dirty or damaged compressor im-pellers (or airfoils). This is one of the diagnostic tools used in defining machine health. The thermodynamic laws underlying the compression of gases are the same for all compressors—axial and centrifugal. However, each type exhibits different operating characteristics (Figure 3-9). Specifi-cally, the constant speed characteristic curve for compressor pressure ratio relative to flow is flatter for centrifugal compressors than for axial compressors. Therefore, when the flow volume is decreased (from the design point) in a centrifugal compressor, a greater reduction in flow is possible before the surge line is reached. Also, the centrifugal com-pressor is stable over a greater flow range than the axial compressor, and compressor efficiency changes are smaller at off design points. For the same compressor radius and rotational speed the pressure rise per stage is less in an axial compressor than in a centrifugal compres-sor. But, when operating within their normal design range, the efficiency of an axial compressor is greater than a centrifugal compressor.

Figure 3-9. Comparing Characteristic Curves of Axial and Centrifugal Com-pressors.

Compressor Types 25

Variations in Compressor DesignVariable Guide Vane Compressor Variable guide vanes (VGVs) are used to optimize compressor performance by varying the geometry of the compressor. This changes the compressor characteristic curve and the shape and location of the surge line. In this way the compressor envelope can cover a much wider range of pressure and flow. In centrifugal compressors the vari-able guide vane (that is, the variable inlet vane) changes the angle of the gas flow into the eye of the first impeller. In the axial compressor up to half of the axial compressor stages may incorporate variable guide vanes. In this way the angle of attack of the gas leaving each rotating stage is optimized for the rotor speed and gas flow. Variable guide vane technology enables the designers to apply the best design features in the compressor for maximum pressure ratio & flow. By applying VGV techniques the designer can change the com-pressor characteristics at starting, low-to-intermediate and maximum flow conditions. Thereby maintaining surge margin throughout the operating range. Thus creating the best of all worlds. The compressor map in Figure 3-10 demonstrates how the surge line changes with changes in vane angle.

Figure 3-10. Composite Surge Limit Line Resulting From Variations In Vane Position


Variable guide vanes have an effect on axial and centrifugal com-pressors similar to speed: the characteristic VGV curves of the axial compressor being steeper than the characteristic VGV curves of the centrifugal compressor. The steeper the characteristic curve the easier it is to control against surge whereas the flatter characteristic curve the easier it is to maintain constant pressure control (or pressure ratio control).

POSITIVE DISPLACEMENT COMPRESSORS

There are many types of positive displacement compressors, but only the three major types will be discussed in this book. These are the blower, reciprocating compressor, and screw compressor. Blowers and screw compressors share some common features and a common history, while reciprocating compressors are noticeably different.

Blowers The blower is a positive displacement compressor that was in-vented and patented in 1860 by Philander H. Roots and Francis M. Roots two brothers from Connersville, Indiana. Initially it was intended as a gas pump for use in blast furnaces. The blower design consists of two figure-eight elements or lobes. These lobes are geared to drive in opposite directions (see Figure 3-11).

Since the rotary lobes need to maintain clearance between each of the lobes a single stage blower can achieve only limited pressure ratio differential—typically 2.0. However, it is capable of compressing large volumes of gas at efficiencies up to 70%. Therefore, blowers are

Figure 3-11. Rotary Blower Lobes Courtesy of Dresser

Roots Inc.

Compressor Types 27

frequently used as boosters in various compression applications. A typical two-lobe blower operating sequence is shown in Figure 3-12. Note that the upper rotor or lobe is turning clockwise while the lower lobe is turning counterclockwise. Position #1: gas enters the lower lobe cavity from the left as com-pressed gas is being discharged from this cavity to the right and simul-taneously gas is being compressed in the upper lobe cavity. Position #2: the upper lobe cavity is about to discharge it’s com-pressed gas into the discharge line. Position #3: gas enters the upper lobe cavity from the left as com-pressed gas is being discharged from this cavity to the right and simul-taneously gas is being compressed in the lower lobe cavity. Position #4: the lower lobe cavity is about to discharge it’s com-pressed gas into the discharge line.

Figure 3-13 shows a cross-section of a two-lobe blower. The blower discharges gas in discrete pulses, similar to the reciprocating compressor (pulsation control is discussed in more detail in the section on reciprocat-ing compressors). These pulses may be transmitted to the downstream equipment or process and must be treated accordingly. For example, if the downstream piece of equipment is a reciprocating compressor, the scrubber and suction pulsation bottle will dampen the pulses from the blower. Some blower designs incorporate pulsation chambers in the body of the blower casing to help minimize pulses. Blowers are used to boost low pressure gas (at or near atmospheric pressure) to the inlet of larger compressors. The various applications for blowers are listed in Table 3-1.

Figure 3-12. Typical Blower Operating Sequence


Figure 3-13. Roots Blower Courtesy of Dresser-Roots, Inc.

Table 3-1. Blower Applications———————————————————————

Gas Boosting Landfill Extraction Digester Gas Process Vacuum Gas Recycling Direct Reduced Iron Gas Blowing

———————————————————————

Blowers are usually belt driven with an electric motor as the driver. Chain driven applications are also used where the driver is close-coupled to the blower as in reciprocating engine/turbocharger applications. An exploded view of a two-lobe blower is shown in Figure 3-14. Table 3-2 list the general causes of problems encountered in blower operation.

Compressor Types 29

Figure 3-14. Exploded View of the Roots Universal RAI Blower Cour-tesy of Dresser Roots, Inc.

Most problems can be avoided by first checking that the:• Driver operates properly before connecting it to the blower;• Blower turns freely before connecting it to the driver and process

piping• Oil type is correct and• Oil reservoir is filled to the proper level

Blowers have been developed in both two-lobe and three-lobe configurations. A three-lobe rotor is shown in Figure 3-15.

Reciprocating Compressors The reciprocating, or piston compressor, is a positive displacement compressor that uses the movement of a piston within a cylinder to move gas from one pressure level to another (higher) pressure level. The simplest example of this is the bicycle pump used to inflate a bicycle tire. In the bicycle pump example we have the piston attached to a long rod, a pump cylinder (a tube or pipe closed on each end), two ball-type check valves (one inlet and one outlet), a flex hose and a connection de-


Table 3-2. Troubleshooting Chart————————————————————————————————TROUBLE POSSIBLE CAUSE————————————————————————————————Will Not Start Check Driver Uncoupled Impeller Stuck Dirty blower elements————————————————————————————————No flow Wrong rotation Obstruction in inlet or discharge piping———————————————————————————————— Speed too low Belt tension too looseCapacity Low Compressor ratio too high Obstruction in inlet or discharge piping Dirty blower elements————————————————————————————————Excessive Power Draw Compressor ratio too high(Motor Amp High) Obstruction in inlet or discharge piping Dirty blower elements———————————————————————————————— Misalignment Impellers unbalanced Impellers contactingVibration Driver or blower casings loose Foundation not level Worn bearings/gears Piping pulsations Dirty blower elements————————————————————————————————Bearing/Gear Damage Coupling/belt misaligned Belt tension excessive————————————————————————————————Breather Blow-by Broken seals or O-ringExcessive Oil level too high Oil type incorrect————————————————————————————————

vice. The piston is as simple as a leather or rubber disk with two smaller diameter metal disks on either side to give the piston strength and stiff-ness. The cylinder is the tube that the leather-sandwiched piston moves within. The inlet or suction spring-loaded ball valves allow ambient air to flow into the cylinder as the piston is partially withdrawn from the cylin-

Compressor Types 31

der (creating a partial vacuum) and closes while the piston compresses the air within the cylinder on the compression stroke; while the discharge ball valve (which is closed on the suction cycle) opens to let the compressed air flow out of the cylinder and into the receiver or tire. Reciprocating compressors are used in many different industries:• Oil & gas

— Oil refineries— Gas gathering— Gas processing

• Transportation— Gas pipelines

• Chemical plants• Refrigeration plants (large and small)

Reciprocating compressors types include the:— Single-cylinder compressor— Multi-cylinder balanced opposed compressor, and the— Integral-engine compressor.

Figure 3-15. Three-lobe Blower Rotor Courtesy of Dresser Roots, Inc.


Single-cylinder Reciprocating Compressor The single-cylinder reciprocating compressor (Figure 3-16) contains all the elements of any of the reciprocating compressors types. Major elements of this compressor are the:

— Frame— Piston— Cylinder— Crankshaft— Connecting Rod— Crosshead— Piston Rod

The frame, crankshaft, connecting rod and piston rod define the strength or horsepower capability of the compressor. This overall strength or horsepower capability is combined into one measurement and that is rod load. Rod load is quoted in the English measurement system as pounds and in the metric system as Newtons.

Figure 3-16. Worthington HB GG 24 x 13 Single Stage Compressor before overhaul. Courtesy of Sun Engineering Services, Inc. and Power & Compres-sion Systems Co.

Compressor Types 33

Rod load is the algebraic sum of the inertia forces acting on the compressor frame, crankshaft, connecting rod, crosshead, piston and piston rod and gas loads. Therefore, rod load must take into account in-ertia forces of the reciprocating components and the gas load calculated at crank angle increments (usually 10 degree increments). Rod loads are defined as combined or overall loads, compression load and tension load.

Compression Rod Load = 0.785 x [DP2 x (PD – PS) + DR

2 x PS] (3-8)

Tension Rod Load = 0.785 x [DP2 x (PD – PS) – DR

2 x PS] (3-9)

Where: DP = Piston diameter – inches DR = Rod diameter – inches PD = Discharge pressure – psia PS = Suction pressure – psia

A change in sign of the combined or overall loads indicates a reversal in the vector loading on the bearing. This is discussed later in the section on bearings.

Frame The frame or crankcase is generally made from cast iron although in some instances it is fabricated from steel plate. The crankcase forms the base upon which the crankshaft is supported. The crankcase also serves as the oil reservoir.

Crankshaft The crankshaft, typically made of forged steel, consists of crankpins and bearing journals (Figure 3-17). While small crankshafts have been machined from a single forging, most crankshafts are fabricated from three parts: bearing journals, crankpins and interconnection pieces. The crankshaft converts rotating motion to reciprocating linear mo-tion via the connecting rod. The crankshaft journal connects to the journal bearing on the crank-end of the connecting rod and the crankpin connects to the journal bearing on the crosshead-end of the connecting rod.

Crosshead The crosshead (Figure 3-18) rides in the crosshead guide moving linearly in alternate directions with each rotation of the crankshaft. The


piston rod connects the cross-head to the piston. Therefore, with each rotation of the crankshaft the piston moves linearly in alternating direc-tions. The crosshead sits in the crosshead guide with the crankshaft on one end and the piston rod on the other. The crosshead guide may be an integral part of the “distance piece” or a separate section. The guide consists of two bearing surfaces: one on top of the guide and one on the bottom. As the crosshead moves back and forth within the guide

Figure 3-18. Crosshead Assembly (Piston Rod Not Shown) Courtesy of ACI Services, Inc.

Figure 3-17. Superior MH/HW Compressor Crankshaft Made From Either Forged Or Billet Heat Treated SAE 4140 Steel Courtesy of Cooper Cameron Corporation.

Compressor Types 35

the piston rod exerts a vertical motion on it. As a result the crosshead is alternately in contact with the bottom bearing surface and the top bearing surface. Care must be taken that piston rod droop is within ac-ceptable limits so as not to result in the crosshead cocking excessively.

Note: See Chapter 12 Case History for the article by Gordon Ruoff on piston rod run-out.

Distance Piece The distance piece is used to provide sufficient space to facilitate separation of toxic vapors from the oil system and the atmosphere. The “distance piece” mounts to the compressor frame. Besides housing the crosshead guide the distance piece also houses the oil wiper packing (Figure 3-19). The piston rod traverses through the packing as it moves from the head-end to the crank-end thereby acting to seal the process gas in the cylinder from the lubricating environment. Due to the wear created on the piston rod resulting from contact with the packing the pis-ton is often hardened in this contact area. The rod packing is originally assembled when the piston and rod assembly is installed. However, after assembly the sealing rings can be removed and replaced as needed.

Piston Rod The piston rod is threaded on both ends and may even have a collar on the end that connects to the piston (Figure 3-20). Connecting the piston rod to the piston is an elaborate procedure. The piston must be secured to the rod so that the alternating inertial loading created

Figure 3-19. Rod Packing Assembly Courtesy of Cooper Cameron Corporation


by the accelerating, decelerating and stopping of the piston with each revolution of the crankshaft does not put undue stresses on the rod or the piston ends. This is accomplished by stretching the rod between the crank-end collar and the head-end piston rod nut while simultaneously compressing the piston ends. The sketch below, though grossly out of proportion, illustrates the process (Figure 3-21). Compressor manufacturers expend considerable amount of engi-neering and test man-hours to arrive at the proper lubricant and torque

Figure 3-20. Piston, Piston Rod, Rings and Wear Band Courtesy of ACI Services, Inc.

Figure 3-21. Exaggerated Representation of Piston Deformation and Rod Stretch.

Compressor Types 37

Figure 3-22. Three Piece Cast Iron Pistons With TFE Riders And Flanged Rods Courtesy of ACI Services, Inc.

to stretch the rod and compress the piston while staying within the elastic limits of the materials of both parts. The crank-end of the piston rod is threaded into the crosshead and a lock nut is used to guard against the rod backing out of the crosshead. The locknut also functions as a counterweight to compensate for loading on opposite sides of the crankshaft.

Piston Pistons may be single-acting or double-acting. Single-acting pistons compress the process gas as the pistons moves towards the head-end of the cylinder only (see Figure 3-24). Double-acting pistons compress the process gas as the piston moves in both directions (see Figure 3-23). Also pistons may be cast from one piece or assembled from several. Although piston balance has not been addressed by most manufacturers there have been cases where piston imbalance has caused rod failure.

Cylinders There are two types of compressor cylinder designs: “Valves In Bore” and “Valves Out of Bore.” The valves in bore design has the compressor valves located radially around the cylinder bore within the length of the cylinder bore. These cylinders have the highest percentage of clearance due to the need for scallop cuts at the head-end and crank-


Figure 3-24. Single Acting Piston Operation

Figure 3-23. Piston and Cylinder Assembly Courtesy of Cooper Cameron Corporation

Compressor Types 39

end of the cylinder bore to allow entry and discharge for the process gas. The valves out of bore design consists of compressor valves at each end of the cylinder. While this design provides a lower percent clearance it is more maintenance intense. The unusually large outer head requires a large number of bolts that must be removed for routine service of the piston, wear band and compression rings. Sealing gas pressure on the crank-end is also problematic. A compressor cylinder houses the piston, suction & discharge valves, cooling water passages (or cooling fins), lubricating oil supply fittings and various unloading devices. Medium to large compressor cylinders also include a cylinder liner. The volume change during each stroke of the piston is the pis-ton displacement (see Appendix F Compressor Cylinder Displacement Curves. The minimum cylinder volume is the clearance volume. There must be some clearance volume in a cylinder to avoid contact with the head-end or crank-end of the piston. See “Sizing & Selection” at the end of the Reciprocating Compression Section for details on calculating volumetric clearance and brake horsepower. While a cylinder’s capacity is physically fixed, several methods can be employed to reduce capacity and/or pressure differential. Volume pockets (fixed and variable), volume bottles, suction valve unloaders and speed can be used to vary the compressor throughput. These will be discussed in more detail in the section on controls.

Valves As discussed in the bicycle pump example, valves are used to con-trol the gas flow into and out of the cylinder. There are suction valves and discharge valves. The suction valves open at the start of the suction stroke to allow gas to be drawn into the cylinder and close at the start of the compres-sion stroke. The discharge valves open when sufficient pressure builds up in the cylinder to overcome the combination of valve plate(s) spring load and downstream pressure. At the end of the compression stroke (just prior to the piston reversing direction) the pressure across the valve equalizes and valve plate(s) spring load closes the discharge valve. Valve plate (or poppets) should be smooth and should not bounce as these shorten valve life. Figure 3-25 shows the correct valve operation. Figure


Figure 3-26. Incorrect Valve Op-eration—Valve Overspringing Courtesy of Roger Vaglia, Prod-uct Manager (Retired), White Superior Compressors 1962-2000

Figure 3-27 Compressor Pop-pet Valve Courtesy of Cooper Cameron Corporation

3-26 indicates an incorrect valve operation, in this case resulting from higher than necessary spring force. Valve designs may employ either plates or poppets and there are many variations on these designs. The pros and cons in the use of each are dependent on the com-pression ratio and the gas characteristics. Figures 3-27, 3-28 and 3-29 provide a general overview of the two valve types.

Figure 3-25. Correct Valve Operation Courtesy of Roger Vaglia, Product Man-ager (Retired), White Superior Compressors 1962-2000

Compressor Types 41

Figure 3-28 Compressor Plate Valve Courtesy of Cooper Cam-eron Corporation

Figure 3-29. Poppet Valve Pop-pets Courtesy of ACI Services, Inc.

Suction valves are installed in the cylinder valve cavity above the cylinder centerline and discharge valves are installed in the cylinder valve cavity below the cylinder centerline. API Standard 618 “Reciprocating Compressors For Petroleum, Chemical, And Gas Industry Services” stipulates that the suction and discharge valves be designed and manufactured such that they cannot be inadvertently interchanged. Suction valve unloaders (plug or finger type—see Figures 3-30, through 3-33 and volume pockets Figure 3-34) can be used to regulate compressor throughput. The suction valve unloaders prevent gas pres-sure from building up in the cylinder thereby reducing that cylinder’s output. A drawback of this approach is that cylinder gas temperature will rise if the same cylinder end is unloaded indefinitely. To prevent this from happening the control system must monitor cylinder gas tempera-tures and cycle through various cylinder loading configurations (this is addressed in more detail in Chapter 5, Throughput Control). An alternate approach is the Head-End Bypass offered by ACI Services, Inc. Utilizing SimPlex™, the configuration the head-end is un-loaded by porting the head-end cylinder compressed gas directly back to suction. The DuPlex™ adds a pocket unloader to increase the head-


Figure 3-30. Plug-Type Suction Valve Unloader Courtesy of ACI Services, Inc.

end clearance for a partial reduction in throughput and the head-end unloader for maximum head-end unloading. An advantage of this approach is that the head-end may be un-loaded indefinitely. As with the suction valve unloader approach addi-tional control logic must be incorporated into the compressor controller to activate the unloader. The plug type unloader (Figure 3-30) uses a plug or piston through the center of the valve. When the plug is retracted gas pressure cannot build up in the cylinder. The finger type unloader uses a set of “fingers” on each valve plate to hold the valve plates open thus preventing gas pressure build up in the cylinder (see Figure 3-32 and 3-33). There are some drawbacks with each type of unloader:

• The fingers, in the finger-type unloader, are subject to fatigue due to repeated use and could break off and be ingested into the cylinder.

• Liquids and contaminates in the gas could cause the plug in the plug-type unloaded to stick or seize.

Compressor Types 43

Figure 3-31. and Head-end Unloaders Courtesy of ACI Services, Inc.

Figure 3-32. Finger-type Un-loader Courtesy of Cooper Cameron Corporation


Volume pockets increase the clearance volume (see the discussion on clearance volume and volumetric efficiency later in this chapter). This is another way to vary compressor throughput. As seen in Figure 3-34 retracting the plug increases the cylinder volume by the amount of the bottle capacity. The only limitation to the size or number of volume bottles that can be added to a cylinder is the physical space available.

Lubrication Lubrication is required for the main bearings, the crankshaft bear-ings, the crankpin bearings, the crosshead pin and the crosshead guide. Lubrication is also required for the piston. Several methods are available to provide lubrication to the bearings and crosshead:

• Forced-feed method• Passive gravity-feed method

The forced-feed oil lubrication uses a pump to move the oil through the compressor oil feed passages. A forced feed lube oil system is de-picted in Figure 3-35. The passive gravity method, as shown in Figure 3-36, uses a scoop or oil collector to move oil from the reservoir to a high point in the compressor. From that location the oil flows due to gravity to the bear-ings throughout the compressor.

Figure 3-33. Finger Type Unloader in the Open and Closed Posi-tions Courtesy of Coo-per Cameron Corpora-tion

Compressor Types 45

Figure 3-35. Compressor Frame Lubrication Oil System Courtesy of Petroleum Learning Programs, 305 Wells Fargo Dr., Houston, Texas 77090.

Figure 3-34. Volume Bottle Unloader Assembly in the Unloaded Position Courtesy of ACI Services, Inc.


The force-feed lubrication system employs a shaft driven pump as the primary oil pump and an electric motor driven pump as the backup pump. Also included are dual lubrication filters. The force-feed lubrica-tion method is usually employed in higher speed compressors (over 600rpm). In cooler (or cold) climates a lubrication oil heater is installed in the reservoir. This oil heater is used to heat oil to a temperature at which oil will flow freely. For mineral oil this is approximately 35°F. A separate lubrication system is employed to lubricate the piston within the cylinder. Since this method adds oil, an incompressible fluid, into the cylinder care must be taken not to add too much oil. Excessive oil supply could accumulate in the cylinder resulting in damage to the rod, piston or cylinder. The manufacturer will normally provide instruc-tions to set the correct amount of oil supplied to the cylinder. Similar information is provided in the charts in Appendix G—Compressor Cyl-inder Lubrication.

Bearings Hydrodynamic bearings are used almost exclusively in reciprocat-ing compressors. The older, heavy-duty compressors use the split bear-

Figure 3-36. Oil Scoop can be seen directly below the bearing cap nut. Cour-tesy of Power & Compression systems

Compressor Types 47

ing design. The newer, lighter compressors use the tilt-pad designs. Due to the way the oil distribution paths are machined into the bearing holders and supports the crankshaft may only be rotated in one direction to maintain proper oil supply. Bearing life is affected by the condition and temperature of the lubricating oil, rod load, and rod reversal. The rod load should change from compression to tension and back to compression with each revolu-tion. That is with each revolution the load vector on the bearing must change to allow continuous lubrication to all bearing surfaces. If the load vector does not change the flow of lube oil becomes inadequate eventu-ally leading to bearing failure. Rod reversal is defined as the minimum distance, measured in degrees of crank revolution, between each change in sign of the force in the combined rod-load curve (Figure 3-38). Rod reversal less than 15 degrees crank angle is cause for concern.

Pulsation Bottles While strictly speaking pulsations bottles are not an integral part of the reciprocating compressor, they are a necessary part of the recip-rocating compressor package. Pulsation bottles are placed at the inlet to the compressor and at the discharge from each compression stage. The pulsation bottle’s stated purpose is to dampen pressure waves created by each compression pulse. The suction pulsation bottle dampens the pressure pulse to the suction piping. Similarly the discharge pulsation bottles dampen the pressure pulses to the discharge piping. These pulses, if transmitted to

Figure 3-37. Connect-ing Rod and Bearings Courtesy of ACI Ser-vices. Inc.


either the suction piping or discharge piping could be very detrimental to the operation and safety of the compressor. Suction bottles are mounted on top of the compressor and dis-charge bottles are mounted beneath the compressor cylinders (Figure 3-39). Sizing suction and discharge pulsation bottles is a complicated pro-cess. Although there are software programs that help with the calcula-tions they should only be used by engineering personnel with experience in the seizing process and the software program. The American Petro-leum Institute (API) suggests three design approaches to be followed when seizing pulsation bottles. The basic API guidelines are repeated here for the reader’s convenience. However, it is strongly suggested that the entire specification be read and applied.

Reference Paragraph 3.9.2.1 “Design Approach 1: Pulsation control through the use of pulsation suppression devices designed using proprietary and/or empirical techniques to meet pulsation levels required in 3.9.2.5 based on the normal operating condition.” “Design Approach 2: Pulsation control through the use of pulsation suppression devices and proven acoustical simulation techniques in conjunc-

Figure 3-38. Red Reversal is the Shortest Distance, in Crank Angle Degrees, from Point to Point.

Compressor Types 49

tion with mechanical analysis of pipe runs and anchoring systems (clamp design and spacing) to achieve control of vibrational response.” “Design Approach 3: The same as Design Approach 2, but also employ-ing a mechanical analysis of the compressor cylinder, compressor manifold, and associated piping systems including interaction between acoustical and mechanical system responses as specified in 3.9.2.6”*

In order to size the suction and discharged bottles accurately de-tailed, information about the compressor and the suction and discharge piping must be available. This study is referred to as the Harmonic Resonance Analysis. Failure to perform the harmonic resonance analysis could result in pipe vibrations severe enough to rupture the gas piping. The isometric piping drawing (see Figure 3-40) describes the pip-ing layout, changes in elevation, size and changes in size. This infor-mation is critical to the safe design and operation of the compressor package.

Figure 3-39. Dresser-Rand 7 HOSS-4 Compressors with Suction & Discharge Pulsation Bottles Courtesy of Power & Compression Systems

*API Standard 618 “Reciprocating Compressors For Petroleum, Chemical, And Gas Indus-try Services.” Latest Edition.

50

Com

pressor Handbook: Principles and Practice

Figure 3-40. Piping Isomet-ric (One Line) Drawing Courtesy of Sun Engineer-ing Services, Inc.

Compressor Types 51

Figure 3-41 is the overhauled and upgraded compressor shown in Figure 3-16. These two pictures demonstrate the longevity and useful-ness of compression equipment.

Balanced Opposed Reciprocating Compressor The balanced opposed reciprocating compressor as shown in Fig-ure 3-42 makes use of multiple cylinders (as well as a crankshaft with multiple crankpin throws, connecting rods, crossheads, pistons and pis-ton rods). Everything that was discussed in the section on single-cylinder compressors applies to the balanced opposed compressor. The only limiting factor to the number of cylinders installed on a balanced opposed compressor is the load capability of the frame, the ability to maintain main bearing alignment and speed. Balanced opposed compressors up to 12 cylinders are not unusual. Generally speaking compressors with two to six cylinders operate at speeds up to 1200 rpm. Compressors with eight or more cylinders run at speeds of 300 rpm.

Figure 3-41. Worthington HB GG 24 x 13 Single-stage Compressor after over-haul. Compliments of Sun Engineering Services, Inc., and Power & Compres-sion Systems Co.


The separable balanced opposed compressor may be driven by a gas engine, electric motor and, in rare cases, a steam turbine or a gas turbine. Whatever driver is selected a torsional analysis should be per-formed on the entire drive train. Torsional critical speed and response analysis will be addressed in more detail in Chapter 8, Vibration.

Integral Engine-compressor The integral engine-compressor is not a new innovation, but in fact, an old one. Although specific dates and manufacturers are not available the integral engine-compressor was derived from the steam engine in the late 1800s. In the mid-1800s a compressor cylinder was fitted to a gas engine. The cross-section of a typical integral gas engine-compressor is shown in Figure 3-43. The concept proved successful for a number of reasons:

1 Cylinder (head-end), 2 Cylinder (crank-end), 3 Piston, 4 Piston Rings, 5 Piston Rod (Induction hardened in packing area), 6 Compressor Packing, 7 Non-reversible, noninter-changeable valves, 8 Valve Retainer, 9 Valve Cap, 10 Crosshead Guide, 11 Crosshead, 12 Replaceable Tri Metal Shoe, 13 Connection Rod, 14 Main Bearing Cap, 15 Crankshaft, 16 Tie Bar, 17 Breather Cap, 18 Top Cover, 19 Oil Filter, 20 Force Feed Lubricator, 21 Crosshead Guide Support, 22 Suction Nozzle Flange, 23 Discharge Nozzle Flange

Figure 3-42. Superior WH-64 Separable Balanced Opposed Reciprocating Compressor Courtesy of Cooper Cameron Corporation

Compressor Types 53

• Manufacturing technology for crankshafts, connecting rods, pistons and bearings was well developed.

• Both engine and compressor shared the same crankshaft and the same lubricating system.

• The need for a coupling was eliminated.• The base plate was only slightly wider than the engine-only base

plate.• The engine and compressor cylinders could be shipped separately

and the compressor cylinders mounted in the field. Thus reducing shipping width.

Preliminary Selection & Sizing Compressor selection is dependent on target throughput and spe-cific knowledge of the compressor or compressors to be considered. This

Figure 3-43. Integral Engine-compressor Cross-section with Power Pistons in the Vertical Plane and Compressor Pistons in the Horizontal Plane, Courtesy of Cooper Cameron Corporation


process starts with determining the specific site and design condition. Site Conditions Altitude = Sea level (barometric pressure) Ambient temperature Design conditions Gas suction pressure = Ps Gas suction temperature = Ts Intercooler temperature = Tintercooler = (target 130°F maximum) Gas discharge pressure =Pd Required flow capacity = MMSCFD (by definition standard cubic feet per day is gas flow at 14.7 psia and 60°F).

Gas Properties Specific gravity (SpG) Ratio of specific heats = k = (see Appendix C— physical properties of gases).

In addition to the calculations discussed in Chapter 2, “General Theory,” the following calculations are specific to reciprocating compres-sors. Determine the number of compression stages by first dividing the final discharge pressure by the initial suction pressure (all pressures must be absolute).

PD Overall compression ratio = CR = —— (3-8) Ps

Where Pd = Discharge pressure (psia) Ps = Suction pressure (psia)

The potential stages are determined by taking the root (square root, cubed root, etc) of the CR. The stage CR higher than 4.0 should not be considered as this would exceed material strengths of most compressors. In most cases an acceptable CR will be between 1.3 and 3.9. The higher the CR the fewer stages required and the lower the compressor cost. Next the adiabatic discharge temperature for each stage is calcu-lated using the following:

Compressor Types 55

Td = Ts # CRstage kk–1

(3-9)

Where Td = Discharge temperature °R Ts = Suction temperature °R CRstage = Compression ratio per stage k = Specific heat ratio = Cp/Cv.

The maximum allowable gas discharge temperature for general service is 350°F. If calculated temperatures exceed this than use more stages. A good rule of thumb is to select approximately the same CRs per stage.

Note: In determining the number of stages the interstage pressure drop is neglected.

Estimating the required HP is the next step in the sizing & selec-tion process. With this information in hand one or more drivers can be considered. The required horsepower for a compressor depends on the net amount of work done on the gas during the complete compression cycle. The following equation may be used to determine the stage HP or the total HP depending upon the stage information selected.

MMCFDTheoretical HP = 46.9 # k–1

k# CR k

k–1–1` j8 B# 2 # Z 1

Z 1 + Z 2; E (3-10)

Where CR = Compression ratio overall or per stage Z = Compressibility factor overall or per stage

Multiplying the theoretical HP/MMCFD by the throughput re-quired (MMCFD) provides the total Theoretical HP. Figure 3-44 may be used to approximate HP and temperature rise. Determining the cylinder size, or sizes for multi-stage compressors, is an iterative process that requires knowledge of the cylinder diameter, stroke and speed. These calculations are easily and accurately performed in the manufacturer’s computer sizing program. However, the reader can approximate these calculations by utilizing the nomographs in


Appendix F—Cylinder Displacement Curves. By selecting the cylinder diameter, stroke and compressor speed the cylinder displacement, in thousand cubic feet per day (MCFD), is easily determined. With this information in hand the reader can approach any com-pressor and driver manufacturer and determine if they have the type & size of units required. Clearance Volume: the volume remaining in the compressor cylin-der at the end of the discharge stroke. This volume includes the space between the end of the piston and the cylinder head, the volume in the valve ports, the suction valve guards and the discharge valve seats. This volume is expressed as a percent of piston displacement.

Clearance Volume Percent Clearance (%Cl) = ————————— x 100 (3-11) Piston Displacement

Care should be taken that the units used are consistent. The effect that clearance volume has on volumetric efficiency (Ev) depends on the compression ratio and the characteristics of the gas (that is the “k” value of the gas).

%Ev = 100–CR– Z2

Z1 #CR k1

–1; E# %Cl6 @ (3-12)

Figure 3-44. Estimated Horse Power and Temperature Rise Based on Compres-sion Ratio

Compressor Types 57

SCREW COMPRESSORS

One of the earliest, if not the earliest, screw compressors was de-veloped in Germany in 1878 by Heinreich Krigar. The original Krigar rotor configuration resembles the Roots Blower rotor design, which was exhibited in Europe in 1867, with the exception that the Krigar rotors twist through an angle of 180 degrees along the rotor length. In 1935, Alf Lysholm of Sweden improved the screw compressor with asymmetric 5 female—4 male lobe rotor profile that is still in use today. The screw compressor is classified as a positive displacement device because a volume of gas becomes trapped in an enclosed space and than that volume is reduced. Within the compressor casing there are two screws with mating profiles, screw “A” and screw “B,” with “A” having concave inlets and “B” having convex inlets. These screws rotate in opposite directions with screw “A” receiving power from the outside source and transmitting this power to screw “B” through a set of synchronization gears. As the screws rotate the process gas is drawn into the inlet or suc-tion port. Gas is compressed by rotary motion of the two intermeshing screws. The gas travels around the outside of the screws starting at the

Figure 3-45. Gas Screw Compressor courtesy of MAN Turbo AG


top and traveling to the bottom while it is being moved axially from the suction port to the discharge port. The location of the discharge port determines when compression is complete. A slide valve over the discharge ports is used to control or vary the discharge pressure. The effectiveness of this arrangement is dependent on close fitting clearances between the two screws and sealing the suction and discharge ports. There is no contact between the screws. To improve compressor efficiency oil is injected into the inlet cavity to aid sealing and to provide cooling. Upon discharge the oil is separated from the gas stream to be recycled after filtering and cooling (as necessary). Separation of the oil from the gas can be achieved down to a level of 0.1 parts per million by weight (ppm wt) with a mesh pad and coalescing elements. Borocilicate microfiber material is typically used in coalescing elements. Depending on the application the gas may be further cooled and scrubbed to remove all traces of oil. The oil-flooded compressor is capable of flowing over 90,000 stan-dard cubic feet per hour (SCFH) at 200 psig. Injected oil quantities are approximately 10-20 gal/min per 100 HP. Where oil contamination must be avoided compression takes place entirely through the action of the screws. These oil-free compressors usu-ally have lower pressure and throughput capability. Higher throughputs can be achieved by employing multiple screw compressors in series. In this way throughputs of 120,000 SCFH at 150 psig can be achieved. In reciprocating compressors a small amount of gas (clearance vol-

Figure 3-46. Typical Screw Compressor Rotors Courtesy of Frick By Johnson Controls

Compressor Types 59

ume) is left at the end of the stroke. This gas expands on the next suction stroke thereby reducing the amount of suction gas that could have been drawn into the cylinder. At the end of the discharge process of the screw compressor no clearance volume remains as all the compressed gas is pushed out the discharge port. This is significant as it helps the screw compressor achieve much higher compression ratios than reciprocating compressors. To isolate the screw elements from the bearings, seals are furnished next to the rotor lobe at each end of the machine. Journal bearings are positioned outside the seal area. These are typically hydrodynamic sleeve-type bearings. Tilt-pad thrust bearings are positioned outside of the journal bearings. Both mineral-based and synthetic oils are used. The oil selected is based on compatibility with the gas being compressed.

Screw Compressor Control Three of the four control methods used in controlling screw compressors are also used on other compressor types. They are speed control, suction valve throttling and discharge valve throttling. The one control method unique to screw compressors is the slide valve.

Slide Valve Unloading is achieved by moving the slide valve towards the discharge port (see Figure 3-47). This effectively shortens the working

Figure 3-47. Slide Valve Operation


length of the rotor. Since less gas is compressed the required horsepower is reduced. This control technique is used with a constant speed drive. Flow variation ranges from 100% to 15% with respective power levels down to 40%.

Variable Speed Drive (VSD) This control technique varies the speed to match the required throughput. As in other rotating element machines attention must be paid to the critical speed of the individual rotating elements.

Suction Valve Throttling This control technique is utilized in a constant speed application. This technique controls throughput to the desired set point by modulat-ing gas flow into the inlet. While it is the least expensive control method is has it limitations. The two typical control modes are:

• Full Load: The suction valve is open and the unit is making 100% of its rated flow.

• Part Load: the suction valve is partially closed (or open) and re-stricting gas flow into the suction port.

• No Load: The unit is still turning but the suction valve is closed and there is no flow. As a result the unit is still consuming energy (15-35% of full load) but not producing any flow.

Discharge Valve Throttling This type of control system is not recommended and is seldom used. If the discharge is closed the compressor will overheat and casing pressure could exceed safe operating limits.

Figure 3-48 summarizes the various control methods vs. power consumed. Figure 3-49 demonstrates that screw compressors can be very large.

Compressor Types 61

Figure 3-48. Capacity Control vs. Power

Figure 3-49. Screw Compressor Courtesy of MAN Turbo AG

63

Chapter 4

Effect of Operating Conditions

GENERAL

Acompressor increases thegaspressure from thepoint “A” levelto thepoint“B” level in theprocessofmoving thegas frompoint“A”to point “B.” This applies to both dynamic and positive displacementtype compressors. However, the gas is acted upon by its surroundings. Gas com-position (molecular weight and specific heat), temperature and pres-

sure influence howa gas behaves. Alsoeffecting compressionis the system resis-tance (the pressurethat the compressormust overcome) andthe compressor com-pression ratio (CR).The compressor char-acteristic curve is atool in representinghowtheseparameterseffect compressor op-eration.As shown inFigure 4-1 there aredifferentcharacteristiccurves for differentpressures and tem-peratures.This isalsotrue when molecularweight and specificheat are variables.AsFigure 4-1. Effects of Temperature & Pressure


longasthedeviationsaresmallthecompressorwillbeabletoadapttothese changes through judicious control techniques. Considering the effect of gas temperature and pressure from theequation for GHP it is evident that increases in gas inlet temperature,pressureorcompressionratiowill increasetherequiredGHPnecessaryto compress thegas.This effect isdemonstrated inFigure4-2. Also an increase in molecular weight or specific heat ratio willincrease thehorsepower requirement at constant temperature,pressureandcompression ratioas show inFigure4-3.

DYNAMICCOMPRESSORS

Specificfactorsthatimpactthedynamiccompressoraresurgeandthedifferent aspectsof surgewithvarying compression stages. As design compression ratios are increased the surge line rotatesclockwiseabout the“Zero”pointwith increasingvolumetricflow.(Fig-ures4-4, 4-5, and4-6). Each additional compressor stage changes the shape and locationof the surge line therebygeneratinga composite surge line. Speed affects axial and centrifugal compressors differently; thecharacteristic speed curves of the axial compressor being steeper thanthecharacteristicspeedcurvesofthecentrifugalcompressor.Thesteeper

Figure 4-2 Compression Ratio Effects on Gas Temperature & BHP/Flow

EffectofOperatingConditions 65

thecharacteristiccurvetheeasierit istocontrolagainstsurge.Whereasthe flatter the characteristic curves the easier it is tomaintain constantpressure control (orpressure ratio control). As inlet conditions change the operating point moves along thesystem resistance curve (assume that the system resistance curve isconstant) as follows (seeFigure4-7):

• The operating pointmoves up along the system resistance curvetoPointBas— Ambientpressureincreases— Ambienttemperaturedecreases,andthe— Surgelinemovestotherightandclosertotheoperatingpoint

• TheoperatingpointmovesdownalongthesystemresistancecurvetoPointAas

Figure 4-3. Effect of Specific Heat Ratio on BHP/Flow


— Ambientpressuredecreases— Ambienttemperatureincreases,andthe— Surgelinemovestotheleftandfurtherawayfromtheoperating

point.

Figure 4-4. Single-Stage Compressor Map

Figure 4-5. Two-Stage Compressor Map


Whereasmolecularweight,gas compressibility, inletpressureandtemperaturehaveasignificantimpactoncontrollingcompressoropera-tion,theirimpactcanbeminimized.Thisisaccomplishedbyrecognizingthe similarities in theheadandflowrelationships.

Figure 4-6. Four-Stage Compressor Map

Figure 4-7. Effects of Changing Inlet Conditions


Todemonstrate theequationsforheadandflowcanbesimplifiedby considering head versus flow2. In so doing the equations for headandflow2 canbe reducedas follows:

(4-1)

(4-2)

since (4-3)

than (4-4)

and

(4-5)

Where Hp ≡ Polytropichead—feet Zave ≡ Compressibility factor—average Zs ≡ Compressibility factor—suction conditions Ts ≡ Suction temperature—0R MW ≡ Molecularweight CR ≡ Compression ratio k–1 s ≡ ——Ratioof specificheats andηp≡polytropic efficiency k*ηp ∆Pos ≡ Differentialpressureacrossfloworifice in suction line Ps ≡ Suctionpressure—psia

Thereducedcoordinatesdefineaperformancemapwhichdoesnotvarywithvaryinginletconditions;hasonesurgelimitpointforagivenrotational speed and compressor geometry and permits or facilitates


calculation of the operating point without obtainingmolecular weightandcompressibilitymeasurements. Foragivenrotationalspeedandcompressorgeometry,theoperat-ingpointcanbedefinedbyalinefromtheorigintotheoperatingpoint.This linecanbedefinedby its slope=Hp/Qs

2.Eachcoordinatecanbereducedbya common factorwithout changing the slope. Thustheoperatingpointcanbepreciselydefinedby:(Hp/A)/(Qs

2/A)orHp’/Qs

2’

CENTRIFUGALCOMPRESSORSAFETYLIMITS

A. Performance limits (usuallyoverspeed)arenotuncommon.B. Surge isaserious threat toalldynamiccompressors.Virtuallyev-

erycompressor isoutfittedwith somesortofantisurgeprotection(bleedvalves,VGVs).The consequencesare too severe to ignore.1. If the surge limit is crossed, there will be severe high speed

oscillations accompanied by vibration and rising gastemperatures

2. Surgeisahighspeedphenomenon—thedropfromfullflowto

Figure 4-8. Simplified Compressor Map Using Pressures Only


reverseflowcanoccurin0.04seconds.3. Inadditiontothesurgelimitseveralotherperformancelimits

(pressure,speed,load,choke)havetobeobservedaswell.4. Usuallychokewillnotbeaproblem.

POSITIVEDISPLACEMENTCOMPRESSORS

With positive displacement compressors variations in gas com-position (molecular weight, specific heat ratio, compressibility), gaspressureandgas temperatureare lesscritical than theyare indynamiccompressors.Whilesmallchangeshaveaminimaleffectoncompressoroperation,largechangescouldexceedtheavailabledriverpower,failtomeet the required throughput or fail to achieve the requireddischargepressure. In such a case the compressor could continue to operate atreducedthroughputandpower.However,failuretoachievetherequireddischargepressurewould shutdown theoperation. Reciprocating compressors aredesigned tooperate across a rangeof pressures or system resistances.As shown in Figure 4-9 the charac-teristiccurvebetweenthehighandlowsystemresistancelinesisalmostastraightvertical line.Atveryhighdifferentialpressuresflowstarts todecrease.Operationinthisrangeusuallyresults infailure(seals,pistonrods,bearings). Oneof themajor causesof failure in reciprocating compressors isliquids or hydrates in the gas. Only very small amounts of liquid canbeexpelledthroughthedischargevalve.However,prolongedoperationevenwithasmallamountofliquidwillleadtovalvefailure.Ifthevol-umeoftheliquidexceedstheclearancevolumeeithertherodwillbendor the pistonwill fail. Therefore, it is imperative that the gas enteringthe compressor isdryand free from liquidsorhydrates.

RECIPROCATINGCOMPRESSORSAFETYLIMITS

A. Maximumspeedof thedriveror the compressor.B. Performance limits—pressure, temperature, gas composition (mo-

lecularweight, specificheat ratio)have tobeobservedaswell.C. Liquidsandhydrates in thegasmustbeavoided


Figure 4-9. Reciprocating Compressor Characteristic Curve

73

Chapter 5

Throughput Control

Throughputcontrol(capacitycontrolorprocesscontrol)isachievedbycontrolling theenergy input to thecompressor inorder toreach thecontrolobjective(processsetpoint).Forallcompressor types this isac-complished by using speed control, suction throttling, discharge throt-tlingorrecyclecontrol. Inaddition to theabove, fordynamiccompres-sors, throughput control canbe achievedbyusingguidevanepositionwhereapplicable.Throughputcontrolforreciprocatingcompressorscanbeachievedbyusingfixedorvariablevolumepocketsandsuctionvalveunloaders.And for screw compressors throughput control is achievedby using a slide valve.A discussion of each of these control methodsfollows.

SPEEDCONTROL

Increasing speedmoves the characteristic curve to the right (seeFigure5-1)resultinginincreasingflowataconstantpressureorincreas-ingpressureatconstantflow.Speedcontrol isthemostefficientcontrolmethodand it canbecombinedwithothercontrolmethods forcontrolfine-tuning.

SUCTIONVALVETHROTTLING

This control method restricts the gas flow and pressure into thecompressor inlet. Suctionvalve throttling is accomplishedby installinga controlvalve (either linearorequalpercentage seeChapter9) imme-diatelyupstreamof thecompressorandcontrollingthevalve’spositionas a function of either discharge pressure or flow.As shown in Figure5-2 suctionvalve throttlingmoves the characteristic curve to the left.


Figure 5-2. Suction Valve Throttling Control

Figure 5-1. Speed Control

DISCHARGEVALVETHROTTLING

This controlmethod restricts thepressure from the compressorto match the required process pressure at constant flow (Figure 5-3).Becausethecompressorisworkingharderthantheprocessrequiresthiscontrolschemeisextremelyinefficient.Asaresultthiscontroltechniqueis rarelyused.

VARIABLEGUIDEVANECONTROL

Variableguidevane(VGV)controlappliestodynamiccompressortypesonly.Byopening theVGVs thesurgepointand thecharacteristic

ThroughputControl 75

curvemove to the right. In effect this creates the largest turndown ofallthecontrolmethods.ItalsoincreasesthesurgemarginastheVGV’sclose. While this control method is more efficient than suction valvethrottling it isalsomoreexpensive.Notonly is thefirstcosthigher fora compressorwithVGVs,butalso themaintenance cost ishigher.

Figure 5-3. Discharge Valve Throttling Control

Figure 5-4. Variable Guide Vane Control


RECYCLEVALVECONTROL

The recycle valve returns compressor discharge flowback to suc-tion (Figure 5-5). To optimize recycle valve operation a gas cooler isusually installed in the recycle line. In this way the hot discharge gasdoesnot increase the temperature of the suctiongas into the compres-sor. The recycle valvemay bemodulated from full open to full closedposition to provide a 100% range of control.Also a recycle valvemaybeusedwithsuctionvalveunloaderstosmooththeloading/unloadingsteps.

SUCTIONVALVEUNLOADERS

Suction valve unloaders apply to reciprocating compressors only.Suctionvalveunloaderspreventgaspressure frombuildingup ineachcylinder end. It accomplishes this by holding the valve plates open(finger type unloader) or retracting a plug (plug type unloader) fromthe center of thevalve (suctionvalveunloaders arediscussed indetailin Chapter 3). Suction valve unloaders can be activated to unload onecylinderendata timeorcombinationsofhead-endsandcrank-endsasa function of the total throughput required and gas temperature. Onsome compressor models the gas cycling in and out of the unloaded

Figure 5-5. Recycle Valve Control


cylinderendcanoverheat.Toaddress this issueavalvesequencemustbecreatedtoswitchunloadedends inordertomaintainacceptablegastemperatures.Consideringafourcylindersinglestagebalancedopposedseparable compressor, each of the eight ends (fourhead-ends and fourcrankends) canbe sequencedasdemonstrated inChart5-1.Endsmaybe switched, not only to achieve the required throughput, but also tomaintaingas temperature. Forexample,thefour-cylindersingle-stagebalancedopposedsepa-rablecompressor inFigure5-6 isusedasafuelgasboostercompressorin parallel with a centrifugal compressor to supply fuel gas to a gasturbine at constant pressure and varying throughput. Tomeet the gasturbine flow requirements suction valve unloaders where installed onallhead-endandcrank-endsuctionvalves.Thesevalvesaresequencedas shown inChart 5-1.Toachieve smoothflowandpressure transitionbetween each load/unload step the dedicated control program alsomodulated the recyclevalvewitheach step. Figure 5-7 shows the interaction of flow, horsepower and suctionpressure at each unloading step with changes in suction pressure.At30MMSCFDand190psig thecompressorcanoperatewith4endsun-loadedand the recyclevalve slightlyopen.

Figure 5-6. Four-cylinder Single-stage Balanced Opposed Separable Compres-sor


Chart 5-1. Compressor Loader/Unloader Sequence———————————————————————————————— SpecificEndsToBeLoaded Sequence ——————————————————————— NoEnds within 1H 2H 3H 4H 1C 2C 3C 4C Loaded EachStep 0 N/A 0 0 0 0 0 0 0 0

1 a X 0 0 0 0 0 0 0 b 0 X 0 0 0 0 0 0 c 0 0 X 0 0 0 0 0 d 0 0 0 X 0 0 0 0 e 0 0 0 0 X 0 0 0 f 0 0 0 0 0 X 0 0 g 0 0 0 0 0 0 X 0 h 0 0 0 0 0 0 0 X

2 a X X 0 0 0 0 0 0 b 0 0 X X 0 0 0 0 c 0 0 0 0 X X 0 0 d 0 0 0 0 0 0 X X

3 a X X X 0 0 0 0 0 b 0 0 0 X X X 0 0 c X 0 0 0 0 0 X X d 0 X X X 0 0 0 0 e 0 0 0 0 X X X 0 f X X 0 0 0 0 0 X g 0 0 X X X 0 0 0 h 0 0 0 0 0 X X X

4 a X X X X 0 0 0 0 b 0 0 0 0 X X X X

5 a X X X X X 0 0 0 b X X 0 0 0 X X X c 0 0 X X X X X 0 d X X X X 0 0 0 X e X 0 0 0 X X X X f 0 X X X X X 0 0 g X X X 0 0 0 X X h 0 0 0 X X X X X

6 a X X X X X X 0 0 b X X X X 0 0 X X c X X 0 0 X X X X d 0 0 X X X X X X

7 a X X X X X X X 0 b X X X X X X 0 X c X X X X X 0 X X d X X X X 0 X X X e X X X 0 X X X X f X X 0 X X X X X g X 0 X X X X X X h 0 X X X X X X X

8 a X X X X X X X X

EachstepistobesequencedasafunctionofTIMEandGASTEMPERA-TUREat the compressor end selected.


VARIABLEORFIXEDVOLUMEPOCKETS

Variable andfixedvolumepocket control applies to reciprocatingcompressors only. Volume pockets vary throughput and efficiency byaddingclearanceto thepistonarea.Volumepocketscanbeeitherfixedorvariable (manualorautomatic). Volumepocketsaremosteasilyaddedtothehead-endofthecom-pressorcylinder.Whilevolumepocketscouldbeaddedtothecrank-endthe amount of space is often limited. The p-v plot in Figure 5-8 dem-onstrates how the operation of the volumepockets effects volume andthroughput.

Figure 5-7. Variation in Flow & Horse Power as a Function of Suction Pressure


Figure 5-8. Affect of Opening/Closing The Head-End Variable Volume Pocket

Chapter 6

Centrifugal/Axial CompressorsDescription of Surge

SURGE & STALL

Surge is an operating instability that can occur in dynamic (axial and centrifugal) compressors and blowers (but not in other positive displacement compressors). Surge and stall are often considered to be one and the same. In fact stall is a precursor to surge. As the compressor operating point ap-proaches the surge line a point is reached where flow starts to become unstable. The stall phenomena occurs at certain conditions of airflow, pres-sure ratio, and speed, which result in the individual compressor airfoils going into stall similar to that experienced by an airplane wing at a high

Figure 6-1. Rotating Stall

81


angle of attack. The stall margin is the area between the steady state operating point and the compressor stall line. As each blade row approaches its stall limit, it does not stall in-stantly or completely, but rather stalled cells are formed (Figure 6-1). In axial compressors these stall cells can extend from a few blades up to 180o around the annulus of the compressor. Also these cells tend to rotate around the flow annulus at about half the rotor speed while the average flow across each stage remains positive. Operation in this region is relatively short and usually progresses into complete stall. Rotating stall can excite the natural frequency of the blades. Stall often leads to more severe instability—this is called surge. When a compressor surges flow reversal occurs in as little as 50 mil-liseconds and the cycle can repeat at a rate of 1/2 to 2 hertz (cycles per second). Although complete flow reversal may not occur in every surge cycle, the fact that flow does reverse has been documented in numerous tests. The most impressive, of which, is when the axial compressor on a jet engine goes into surge (see Figure 6-2). From the point of view of passengers on board the only evidence of surge is the repeated, loud “popping” sound that accompanies surge (unless seated by a window, forward of the engines, and looking directly out at the engine inlet). In industrial applications flow reversal is observed as fluctuations

Figure 6-2. Jet Engine Axial Compressor Surge

Centrifugal/Axial Compressors Description of Surge 83

in measured flow simultaneous with an increase in vibrations. If close to a surging compressor a “whooshing” sound may also be heard. Surge can cause the rotor to load and unload the active and passive sides of the thrust bearings in rapid succession. When severe the rotor may contact the stators resulting in physical damage. The compressor map is the single most useful tool for describing surge. The compressor map shows the characteristic curves and the locus of points defining the surge line. The “X” axis is volumetric flow and the “Y” axis may be discharge pressure (Pd), differential pressure (ΔP) or compressor pressure ratio (CR or Rc). In general, surge can be defined by the symptoms detailed in Table 6-1. Surge happens so quickly that conventional instruments and hu-man operators may fail to recognize it. This is especially true if the vibration does not trigger the vibration alarm setting. In cases where there was not an “observed incident,” the evidence of a surge may be verified by a loss in compressor efficiency. In industrial compressor applications the severity of surge is de-

Figure 6-3. Compressor Map


pendent on the volume of the downstream equipment including process vessels, piping, etc. The larger this volume, the greater the damage caused by surge, should it occur. Some factors leading to the unset of surge are listed in Table 6-2.

Table 6-1. Surge Description——————————————————————————————1 Flow reverses in 20 to 50 milliseconds2 Flow reverses at a rate if ½ to 2 Hertz3 Compressor vibration (mostly axial but vertical & horizontal may also

be affected)4 Gas temperature rises5 Surge is accompanied with a “popping” or “whooshing sound6 Trips or shutdowns may occur——————————————————————————————

Table 6-2. Factors Leading to Surge——————————————————————————————

FACTORS LEADING TO SURGE——————————————————————————————

At Full Operation Trips Power Loss Jammed Valves (VGVs) Process Upsets Molecular Weight Changes Intercooler Failure Rapid Load Changes Compressor Fouling Dirty Compressor

—————————————————————————————— At Reduced Operation Start/Stop Cycles Load Changes

——————————————————————————————

Some of the consequences of surge are listed below:• Unstable flow and pressure• Damage to seals, bearings, impellers, stators and shaft• Changes in clearance• Reduction in compressor throughput• Reduction in efficiency• Shortened compressor life

85

Chapter 7

Surge ControlCentrifugal/Axial Compressors

InChapter5throughputcontrolwasdiscussedas itappliedtoallcompressors—dynamic andpositivedisplacement. In this chapter, con-trolwillbediscussed,asitappliestodynamiccompressorsonly.Specifi-callythismeanscontrollingthedynamiccompressortoavoidoperatingat or near surge. To accomplish this several control techniqueswill beemployed. They are: a) minimum flow control; b) maximum pressurecontrol; c)dualvariable control; andd) ratio control. Beforeproceedingitisnecessarytodeterminewherethecompres-sorisoperatingrelativetothesurgeregion.Foragivensetofconstraints(such as, compressor geometry, available control mechanisms and theprocess), the compressor operating location can be defined by a linefromtheorigin to thecompressoroperatingpoint (aswasdiscussed inChapter4—DynamicCompressors).

MINIMUMFLOWCONTROL

Minimumflowcontrolispossiblythesimplestandleastexpensivecontrolmethod.AsindicatedinFigure7-1,thesurgemarginisnormallytunedtopreventsurgeathighpressures.Butwhenthecompressoroper-ates at lower pressure levels, the surgemargin is needlessly increased.This is avery inefficientway to control against surge.

MAXIMUMPRESSURECONTROL

Maximum pressure control is a simple and inexpensive (flowmeasurement instrumentation is not required) control technique mostfrequentlyfoundinconstantspeedcompressors.Thiscontrolmethodisnotveryefficientandoffersonly,about,a10%turndown.AsFigure7-2


shows, themaximum controller set point does not protect against theoperatingpointcrossing intosurgeat lowpressures.Andif theoperat-ingpointgoesabovethecontrollersetpointtherecyclevalvewillopeneven though the operating pointmay be far to the right of the actualsurge region.

DUALVARIABLECONTROL

Dual variable control (Figure 7-3) is employed to control onepri-mary variable while at the same time constraining one or more sepa-ratevariables. For example, aprocessmay require that the compressormaintainaconstantdischargepressure,butatthesametimeensurethatother parameters (such asmaximummotor current, minimum suctionpressureortemperature)notbeexceeded.Exceedinganyoftheselimits

Figure 7-1. Minimum Flow Control

SurgeControlCentrifugal/AxialCompressors 87

couldresult inseriousdamagetothecompressor,damagetothedriverordamage to theprocess equipment (upstreamordownstream). Both flow control andpressure control techniques leave the com-pressorvulnerabletosurgeasdiscussedabove.Amorerealisticcontrolapproachwould be to establish a control line (or surge limit line) thatmimicstheactualsurgelineandissetupatareasonabledistance(mar-gin)fromtheactualsurgeline.Thisistheapproachthathasprovenmostsuccessfulinsituationswheretheabovecontrolmethodsareinadequate.Thismethod is referred toas ratio control.

Ratio Control Recall the equations for polytropic head and flow squared fromChapter 4 (equation 4-1 & 4-2). The same relationships are rewrittenbelow inequation7-1.

s

Hp = MWZave •Ts • σ

Rc–1 and Q = MWZs•Tsc m• Ps

∆Posc m (7-1)

Figure 7-2. Maximum Pressure Control


Using the format fromFigure 7-4, the surge line is renamed thesurgelimitline(SLL)andthesurgelimitsetpointisrenamedthesurgecontrol line(SCL).Toenhancetheabilitytocontrolagainstsurgeovera wide range of conditions additional control ratio limits are added.TheseareshowninFigure7-5SurgeControlMap,andlistedinChart7-1. The surge control map in Figure 7-5 shows the relationship be-tween the actual surge line, the operating point and various controllines developed to control the compressor throughout its operatingrange. These lines aredefinedas follows: Using the definitions in Chart 7-1, the operating point can bedefinedbytheslopeofthelinefromtheorigintotheoperatingpoint.This is referred to as theoperatingpoint lineorOPL.

Hp Hp1,redSlopeof theoperatingpoint line=OPL= — = ——— (7-2) Qs

2 qs1,red

Figure 7-3. Dual Variable Control


Figure 7-4. Ratio Control

Figure 7-5. Surge Control Map


Similarly the surge limit line can be defined by the slope of thesurge limit line and is based on the slope of the surge line. The surgelimit line isdefinedas follows:

Hp Hp1,redSlopeof the surge limit line=SLL= — = ——— (7-3) Qs

2 qs1,red

Furthermore,usingtheratioofthetwoslopes,OPLandSLLanewtermisdefinedwhichisbasedoftherelationshipbetweentheseslopes.This term isGs.

OPLThenGs= —— . (7-4) SLL

And thedistancebetween theoperatingpointand thesurge limitpoint isdefinedas“d”where

Chart 7-1. Surge Control Factors———————————————————————————————— SSL Slopeof theSurgeStopLine———————————————————————————————— SLL Slopeof theSurgeLimitLine———————————————————————————————— SAL Slopeof theSurgeAvoidanceLine———————————————————————————————— SCL Slopeof theSurgeControlLine———————————————————————————————— OPL Slopeof theOperatingPointLine@DesignConditions———————————————————————————————— borb1 DistanceBetweenSCLandSLL———————————————————————————————— SA DistanceBetweenSCLandSAL———————————————————————————————— SS DistanceBetweenSSLandSLL———————————————————————————————— d DistanceBetweenOperatingPoint andSLL———————————————————————————————— dSA DistanceBetweenOperatingPoint andSAL———————————————————————————————— Td DerivativeActionTimeConstant———————————————————————————————— TSA Time IntervalBetweenStepOutputs———————————————————————————————— DEV Distance fromOperatingPoint toSCL————————————————————————————————


d=1–Gs (7-5)

When the slopesof the SLLand theOPLare equal,Gs= 1 andd=0.which is as it shouldbe since thedistancebetween the lines is 0. AlsoDeviationisdefinedasthedistancefromtheoperatingpointto the surge control line (SCL).Therefore,

DEV=d–b• f (∆Po,s) (7-6)

where f(∆Po,s) = function characterizing the shape of the SCL and b isthemarginof safety.

1 f(∆Po,s)maybeeither ——— or1. (7-7) ∆Po,s

1 If f(∆Po,s)= ——— thentheSCLisparalleltotheSLLandbisthe ∆Po,s distancebetween them.

If f(∆Po,s)=1 thenSCL&SLL interceptat 0 and

SCL b=1– —— (7-8) SLL

Using the 0 intercept approach for all control lines the distancebetweensurgecontrol line (SCL)and thesurgeavoidance line (SAL) isrepresentedby

SA output C C d T dt

G• • s= +1 SAo dc m

(7-9)

andb=b1+nb2wheren is thenumberof surge cycles.

The above control techniques can now be used to implement al-gorithms that will avoid a surge and stop surge, should it occur. Thesurge avoidance algorithm and the stop surge algorithm are describedbelow.


SURGEAVOIDANCEALGORITHM

Thedistance fromtheoperatingpoint line to thesurgeavoidanceline isdefinedbydSA.

Where dSA=Gs+b1+nb2–SA–1, (7-10)

SAL–SCLandSA= —————. (7-11) SLL

When the operating point (i.e. OPL) is to the right of the SAL,dSAwill be <0. Only when dSA goes positive is the step output to thebypassvalveimplemented.AlsothestepoutputisrepeatedinTSAtimeintervals until the bypass valve is full open.Donot use the derivativeif the signal isnoisy.

SURGESTOPALGORITHM

SS is the surge stop linedistanceand isdefinedas

OPL SS= —— –1 (7-12) SLL

When theOPL is to the right of the SSL, SS is < 0 andno actionis taken.When the OPL is to the left of the SSL, then SS > 0 and then value for the number if surge cycles is increased a set amount (n=1,n=2,n=3, etc each time theoperatingpoint crosses the surge stop line. Anexampleintheuseoftheabovesurgecontroltechniquefollows:

CALCULATETHESURGELINESLOPE

Using the following relationship:

∆Pos = N2C2d4Y2

1– d/D^ h46 @t # q2

(7-13)

whered=6.373andd=FadmeasandD=10.02andD=FaDmeas,andFa is theThermalExpansionFactor.


Fa=1@70°F (7-14)

Based on a compressormap the surge point is at the following condi-tions: q=1220acfm,flow inactualcubic feetperminute DischargePressure (Pd)=338psia SuctionPressure (Ps)=118.7psia

Alsogasparametersareas follows: MolecularWeight (MW)=17.811 SpecificHeatRatio (k)=1.286 CompressibilityFactor (Z)=0.98 SuctionTemperature (Ts)=70°F=530°R

MW•PsTherefore,density (ρ)=———— =0.37929 (7-15) Z•RsTsand

Velocity of Approach Factor (E)=

(E)

— /, d/D the beta ratio.

d Dwhere

11

4b /= =

^ h (7-16)

Rearrangingand squaring theabove term

1/E2 = [1–β4]= 0.83635 (7-17)

Gas Volumetric Flow Factor (N) ≡ 5.982114 from Table 9.16 pp 9.35*whenflow ismeasured inacfm.

Discharge Coefficient (C)=C∞+b/RDn (7-18)

and C∞ is the discharge coefficient at the infinite Reynolds Number(table9/1pp 9.141*).

C3 = 0.5959 + 0.0312 (β) 2.1 –0.184 (β) 8 +D [1– (β) 4]0.09 (β) 4

– D0.0337 (β) 3

= 0.57874

(7-19)


b=91.70(β)2.5=29.58420andn=0.75 and fromTable9.20equation ipp 9.39* (7-20)

RD = 2266.970 µDNρ; E q andµ=0.0106@1atmosphereand70°F

(7-21)

fromCraneHandbook.

MWgas 17.811SpecificGravity (SG)= ——— = ——— =0.61489 (7-22) MWair 28.966

FromGPSAcurvesreprintedinAppendixCthecriticaltemperatureandpressureof thegas isTc=362.5°FandPc=675psia

70+460Tr= ———— =1.46207 (7-23) 362.5

118.7Pr= ——— =0.17585 (7-24) 675

Thenµ=1.0176•0.0106=0.0108 (7-25)

RD=1,620,432.1 (7-26)

C=0.57939 (7-27)

GasExpansionFactor (Y)=

Y=1– [0.41+0.35(β)4]• kXi fromTable9.26pp 9.48* (7-28)

P1 –P2 cpWhereXi= ———fromeq9.35, andk=— P1 cv

=1.2863 (iterateasnecessary) (7-29)

——————————*ValuesarefoundinFlow Measurement Engineering Handbook,3rdEdition,byRichardMiller


∆PosAssumesurgepoint at 25 inchesofwater column then ——— Ps =0.919/118.7=0.00774 (7-30)

Y=0.99722 (7-31)

∆Pos = N2C2d4Y2

1– d/D^ h46 @ρ # q2=23.95920≈ 24 inches

water columnat surgepoint. (7-32)

and q = 1.61•10–5∆Pos (7-33)

Calculate the surge limit line (SLL)

k–1 cpWheres= —— andk= ——.Thens=0.0.22258 (7-34) k cv

SLL = ∆Pos/Psv

PsPdc m

v

–1; E= 5.837 . 5.8 (7-35)

Similarly the SurgeControlLine (SCL) =4.4 (7-36) SurgeAvoidanceLine (SAL) =4.8 (7-37) StopSurgeLine (SSL) =5.9 (7-38)

And“b” thedistancebetween theSLLand theSCL=

SCL 1– —— =0.24386 0.24 (7-39) SLL

97

Chapter 8

Vibration

Everypieceofequipment,regardlessofsizeorconfiguration,hasanaturalorresonantfrequency.Compressorsarenoexceptionand,infact,numerous vibration analysis techniques are employed to predict com-pressorvibrationduring thedesignphaseand to identify thesourceofhighvibrationintheoperationphase.Iftheresonantfrequency(criticalspeed)isbelowtheoperatingfrequencyorspeed,theunitisconsideredto have a flexible shaft. If the resonant frequency is above the operat-ing speed, theunit is said tohavea stiff shaft (seeFigure 8-1).Almostallaxialcompressorsareconsideredtohaveaflexibleshaft; that is, the

Figure 8-1. Rotor Response Curve Depicting Operation from Critically Damped to Critically Undamped


normal operating speed is above the resonant (or undamped criticalspeed) frequency. The resonant frequency of centrifugal compressorsfalls on either side of the undamped critical speed line depending onthenumberof impellers involvedand the shaft size.Blowers, integral-engine-reciprocating compressors and screw compressors generally fallinto the stiff shaft category. Separable reciprocating compressors trains(driver&compressor)mayfallintoeitherflexibleorstiffshaftcategorydependingon thenumberof cylinders and the typeofdriver selected.Vibrationisamajorconsiderationinthedesignofcompressorrotoras-semblies. Rotor designs vary widely among compressor types and, if fact,they vary even between compressors of similar frame size. Physicalfactors such as number, size, arrangement of impellers or pistons, andbearing spans, bearing housing design, torque requirements, couplingselection and system parameters such as volume flow, casing pressurerating,operatingspeedrangesallaffectthenaturalfrequencyandthere-fore rotordesign. Each element of the drive system, as well as the complete drivesystem and all mechanically connected elements, must be subjectedto detailed vibration analyses. Regardless of the complexity of the ro-tor system both lateral and torsional vibration analysis should be ad-dressed.Furthermore, each compressor rotor and support systemmustbesubjectedtoa thoroughdynamicrotorstabilityanalysis.Thereareanumberofexcellentcompanies,largeandsmall,thathavethecapabilitytoperform thenecessary lateral and torsional analysis. Dynamic rotor stability analysis looks for possible causes of frac-tional frequency whirl or non-synchronous precession such as a largevibrationamplitudecomponentatornearthefirstlateralbendingmodethatcanleadtoexcessiverotorvibration.Whenencounteredthisvibra-tioncanreachmagnitudes largeenoughtomovethecompressoron itsfoundationandevendestroy the compressor. Figure8-2isarepresentationofhydrodynamicbearingwhirlorbits.Using this technique the rotating shaft journal may be observed as itand theoilwedgemoveswithin thebearing. TheAmericanPetroleum Institute Standard 618 recommends thattorsionalnaturalfrequenciesofthedriver-compressorsystem(includingcouplingsandanygearunit)shallbeavoidedwithin10percentofanyoperatingshaftspeedandwithin5percentofanyothermultipleofop-eratingshaftspeedintherotatingsystemuptoandincludingthetenth

Vibration 99

multiple. For motor-driven compressors, torsional natural frequenciesshall be separated from the first and secondmultiples of the electricalpower frequencyby the sameseparationmargins.* The compressor rotor system (as shown in Figure 8-3 and 8-4),consistingof shaft, impellers,balancedrum,etc., togetherwith its sup-portsystem(that is,bearingtype,casing,supports,etc.), issubjectedtoanalysisofitsnaturallateralfrequencyresponseduringthedesignstage.Theoperating speed range is themost critical element in this analysis.Therefore thespeedrangeshouldbe limited to thepredictedoperatingpointswithouta large safetymargin. Identifying the lateral natural frequencies can be a slow processdue to the large number of factors to be considered. Experience is amajorassetinthisprocess.Forexample,acentrifugalbarrelcompressorusually operates between the first and second lateral bendingmodes.

Figure 8-2. Hydrodynamic Bearing Whirl Orbits

—————————*API Standard 618, 4th Edition June 1995 “Reciprocating Compressors for Petroleum,Chemical, andGas IndustryServices,”Section2.5.2


Whenfrequencychangesarenecessarytheycanbemadebymodifyingthe compressor configurationor stiffening the rotor shaft. Frequency isdirectly proportional to the square root of the element spring constantand inversely proportional to the square root of itsmass as shown inequation8-1.

K ω2= — (8-1) m

Where ω =natural circular frequency in radiansper second, k = spring constant or spring modulus or elastic modulus in

lb/in, and m =mass in lb sec2/in

Vibrationmeasurementsusuallyconsistofamplitude (inches),ve-locity(inchespersecond)andacceleration(inchespersecondpersecondor “g”). These measurements are taken with probes, such as seismicprobesthatmeasureamplitude,proximityprobesthatmeasurevelocity,andaccelerationprobesoraccelerometersthatmeasureacceleration(seeFigure 8-4). By the use of integrating or differentiating circuits, ampli-

Figure 8-3. Centrifugal Compressor Cross-section Compliments of Dresser Roots, Inc.

Vibration 101

tude,velocityoraccelerationreadoutsarepossiblefromanyofthethreeprobetypes.However,moreaccurateinformationcanbeobtainedfromdirectmeasurementsusing theproximityprobeandaccelerometer. Typicalsourcesofvibrationthatareusuallyfoundinacompressorpackagearedetailed inTable8-1.

Table 8-1. Typical Sources of Vibration————————————————————————————————VIBRATION TYPE OBSERVATION————————————————————————————————0.42xROTORSPEED OILWHIRL VIOLENTBUTSPO-RADIC0.5xROTORSPEED BASEPLATE VERTICAL RESONANCE MORESENSITIVE1xROTORSPEED ROTOR IMBALANCE2xROTORSPEED MISALIGNMENT AXIALHIGH2xROTORSPEED LOOSENESS AXIALLOWRPMxGEARTEETH GEARNOISE————————————————————————————————

Note that rotor imbalance is seen predominantly at the runningspeed, whereas misalignment can be seen at two times the runningspeedandisseenonreadoutsfromboththehorizontalorverticalprobeand the axialprobe.Loosenessdue to excessive clearances in thebear-

Figure 8-4. Simplified Rotor Shaft System


ingorbearingsupport isalsoseenat twotime’srunningspeed,exceptthe axial probe does not show any excursions. Hydrodynamic journalbearingoilwhirl is seenat 0.42 running speedand isbothviolent andsporadic.Veryclosetothat,at0.5runningspeed,isbaseplateresonance.Notethatbaseplateresonanceispickeduppredominantlyontheverticalprobe. Gear noise is seen at running speed times the number of gearteeth.Amplitudeisnotthebestcriteriatojudgeacceptableorexcessivevibration since a high amplitude for a low rpmmachinemight be ac-ceptablewhereas that same amplitude for a high rpmmachinewouldnotbeacceptable. A typical spectrographic vibration plot is shown in Figure 8-5below. This plot is a “must have” in determining the severity of thevibrationand the sourcesofvibration. AnotherusefultoolisthevibrationmapshowninFigure8-6.Thissemi-logplotprovidesRPMandHertz (cyclespersecond)on thehori-zontal scale and vibration inMils (0.001 inches) displacement, velocity(inches per second) and acceleration (Gs) on the vertical scales. Usingthischartitiseasytoconvertfromonesystemtoanother.However,foraccuracythefollowingequationisprovidedtoconvertMils tovelocity.

Figure 8-5. Spectrographic Vibration Plot

Vibration 103

Figure 8-6. Vibration Map Showing Relationship between Displacement, Ve-locity and Acceleration

Mils×π× rpm V= ——————— (8-2) 60,0000Where V =Velocity in inchesper second Π =Pi=Constant=3.14 RPM =Revolutionsperminute

Thischart isalsouseful indeterminingwhenvibration isnormal,smoothor rough.As shownon the chart amachine is running smoothwhen velocity is below 0.03 inches per second (ips), and it is alwaysrunning roughwhenvelocity is above0.2 ips.

105

Chapter 9

Valve Requirements

Valves have numerous uses in every compressor application. Ingeneral these includeprocess control, isolationand safety.

• The process control valve is used to regulate or control the flow,pressure and temperature of the system. The valves that bestperformthis functionare theballvalve,butterflyvalveandglobevalve.

• The isolationvalve isused to“shut in”acompressor.This isnec-essary in order to performmaintenance on the compressor. Thevalves that best perform this function are the gate valve, knifevalveandplugvalve.

• Thesafetyvalveisusedtoprotectthecompressorandpersonnelinthearea.Itisalwayseitherautomaticallyactivatedoriscontrolledtoactivate in response toa critical condition.Thevalves thatbestperform this functionarepressure reliefvalvesandcheckvalves.

Table 9-1 list the primary functions of the most frequently usedvalves Valveshavedifferentflow-through characteristics as a functionoftheir trim design (see Table 9-2). Valve trims are selected to meet thespecificcontrolapplicationneeds. Inaway this isa throw-back to thedaysofpneumaticandelectriccontrolwhichwerelimitedintheirabil-ity tomanipulate valves.While today’s computer control capability issufficiently sophisticated, valve trims are still selected tominimize theprogrammers work. The most common valve trim characteristics arelinear, equalpercentageandquick-opening. Figure9-1graphicallydemonstratesvalvetrimflowcharacteristicsrelative tovalve travel.Variationsof these threeconfigurationsarealsoavailable.


Table 9-1. Valve Types————————————————————————————————Ballvalve Provideson/off controlwithoutpressuredrop.————————————————————————————————Butterflyvalve Providesflowregulation in largepipediameters.————————————————————————————————Chokevalve Used forhighpressuredrops found inoil andgaswellheads.————————————————————————————————Checkvalve Anon-returnvalveallowsthefluidtopassinonedirectiononly.————————————————————————————————Gatevalve Used foron/off controlwith lowpressuredrop.————————————————————————————————Globevalve Providesgood regulatingflow.————————————————————————————————Knifevalve Providespositiveon/off control alsoused for slurriesorpowders.————————————————————————————————Needlevalve Providesaccurateflowcontrol.————————————————————————————————Pistonvalve Used for regulatingfluids that carry solids in suspension.————————————————————————————————Pinchvalve Used for slurryflowregulation.————————————————————————————————Plugvalve Provideson/off controlbutwith somepressuredrop.————————————————————————————————

Table 9-2. Valve Trim————————————————————————————————Linear Flowcapacity increases linearlywithvalve travel.————————————————————————————————EqualPercentage Equal incrementsofvalve travelproduceequal percentage changes in theexistingCv.————————————————————————————————Quickopening Largechanges inflowareprovided forverysmall changes in lift.Due to itshighvalvegain it isnotused for modulating controlbut it is ideal foron/off service.————————————————————————————————

PROCESSCONTROL

Not all processes are the same and different processes require dif-ferent valve trim and valve response. Throughput control may be ac-complishedwithoneormorevalvesinsuction,dischargeand/orrecycle(throughputcontrolwasdiscussed indetail inChapter5).Thefollowingrecommendationsapply tovalveselection forall compressor types:

ValveRequirements 107

• Valve full stroke tobe less than2 seconds.• Select linear or equal percentage valve characteristics for all but

“shut in”applications.• Minimize the lengthofpneumatic tubingbetween the I/P (pneu-

matic to current) converter and thevalvepositioner.

For surge control and reciprocating throughput control (for ex-ample,usingsuctionvalveunloaders) thevalverequirementsaremorestringent:

• Valve full stroke tobe1 secondor less.• Select linearvalve characteristics.• Installationof lineboostermaybe consideredasnecessary.

Figure 9-1. Valve Trim Flow Characteristics


Recycle (or blow-off in the case of air compressors) valves are al-ways usedwith dynamic compressors, but they can also provide flowcontrol in other compressor applications. The following recommenda-tions shouldbe considered forall recyclevalveapplications:

• Position the recycle valve as close to the discharge port of thecompressoraspossible.

• Minimizethevolumebetweenthecompressordischarge,thecheckvalve in thedischarge line, and the recyclevalve.

• Position the check valve downstream of the “tee” to the recyclevalve. The check valve should not be between the compressordischargeand the recyclevalve.

• Ifrecyclegascoolingisrequired,agascoolershouldbepositioneddownstreamoftherecyclevalvebetweentherecyclevalveandthesuction line.

• Acheckvalve is an important componentof the system toassuresafetyof the compressorunderdifficult conditions.

Valves are typically arrangedarounda compressor asdepicted inFigure9-2.Althoughbothsuctionanddischargevalvesareshown,onlyonevalvewouldbeneeded.

ValveRequirements

109

Figure 9-2. Typical Valve Arrangement

111

Chapter 1o

Instrument Requirements

Instruments aredevices thatmeasure or control variables such asflow, temperature, liquid level, orpressure. Instruments include:

• Sensors—tomeasure— flow— temperature— pressure— speed— vibration— liquidlevel

• Converters to transform the signal from— digitaltoanalogor— analogtodigital,

• Transmitters (to forward themeasured signal to local and remotelocations),

• Analyzers (combines multiple signals to provide a corrected orrelatedoutput).

AccordingtotheInstrumentationandSystemsAutomationSociety(ISA), formerly known as Instrument Society ofAmerica, the officialdefinition of Instrumentation is “a collection of Instruments and theirapplicationforthepurposeofObservation,MeasurementandControl.”* Table10-A lists someof themost commonlyused sensors. Attimesmorethanonsensorisneededtoproduceanappropriatemeasurement.Forexample,anaccurateflowmeasurementrequires the

———————*Reference: ISAstd.S51.1—(InstrumentSocietyofAmerica)


differentialpressureacrossanorifice,thetemperatureofthegasandthecompositionof thegas.

Theaccuracyofflowmeasurementisespeciallyimportantinappli-cationssuchasgassales.Inotherapplications,suchasindynamiccom-pressoranti-surgecontrolsystemsandpositivedisplacementthroughputcontrol, the main criteria for flowmeasurement are repeatability andsufficientsignal-to-noiseratio.Thepreferred locationofaflowmeasur-ingdevice is in thecompressorsuction lineasclose to the inletportaspossible.A less preferable location is in the compressor discharge lineas closeaspossible to the compressordischargeport.

Selectionof the locationshouldbebasedon the followingfactors:

• Simplificationof thealgorithm for throughputor surge control• Simplificationof surgeprotection fordynamic compressors• Installed costof theflowmeasuringdevice• Costofoperationofthesystemasafunctionoftheselecteddevice.

Speedof response is alsoan important consideration.

• Delayinthecontrolsystemduetotheflowmeasuringdevicemust

Table 10-1. Sensor Types————————————————————————————————Parameter Sensor————————————————————————————————Flow Orificeplate, annubar, rotometer, turbinemeter,venturime-

ter————————————————————————————————Temperature Resistance temperaturedetectors (RTDs)and thermocouples————————————————————————————————Pressure Manometers,Pitot tubes,piezometersand transducers————————————————————————————————Speed Proximitorprobesormagneticpickups.Proximitorsare less

likely tobedamagedbyvibration.Proximitorprobesarede-signed tobe intrinsically safe foruse in electricallyhazard-ousareas.

————————————————————————————————Vibration Seismic,proximitor, accelerometer————————————————————————————————Level Floats————————————————————————————————

InstrumentRequirements 113

beabsolutelyminimal (headmetersorgasvelocity).

The flowmeasuring device and its transmitter should be sizedbasedon themaximumflowof the compressor.

• Selectthetypeandlocationoftheflowmeasuringdevicesuchthatthepressuredifferentialcorrespondingtothemaximumflowofthecompressorwouldbe10”water columnormore.

• Minimize the length of the tubing between the flowmeasuringdeviceand the transmitter.

Select the transmitter typeandbrandbasedon the following:

• Reliability• Speedof response (suggest 80millisecondsor less)• Thermocouples (in thermowells) are too slow for controlor surge

detection• Thermocouples (exposed junction) are preferred for performance

trendingand surge control• Resistance temperature detectors (RTD) are accurate in low tem-

peratureapplications<500°F

ThermocouplesandRTDsaremadeofdifferentmaterials suitablefor different temperature ranges. Thermocouples are classified as typeJ,K,T,E,R, andS.RTDsare either 10Ω copperor 120ΩnickelRTD.Table10-2 lists thevarious temperature ranges for each type.

Pressuremeasurementsmaybeininchesofwater(“WCor“H2O),inches ofmercury (“Hg), poundsper square in gage (psig), or poundsper square inch absolute (psia). Transducers are the most commoninstrument used in pressure measurement. Both standard and smarttransducersare in commonuse today.Andbothhave their advantagesand disadvantages. Standard transducers provide near-instantaneousreadingswhicharecritical forsurgecontrol.Smart transducersaveragethe pressure measurements usually over a 1/4 of a second (250 mil-liseconds). Smart transducers shouldnotbeused for surge control.


Table 10-2. Thermocouple & RTD Temperature Ranges———————————————————————————————— TYPE RANGE———————————————————————————————— J –328° to1382°F, –200° - 750°C———————————————————————————————— K –328° to2498°F, –200° to1370°C———————————————————————————————— T –330° to760°F, –200° to404°C———————————————————————————————— E –328° to1832°F, –200° to1000°C———————————————————————————————— R 32° to3213°F, 0° to1767°C———————————————————————————————— S –40° to3214°F, –40° to1768°C———————————————————————————————— 100ΩPtRTD –328.0° to1382.0°F, –200.0° to750.0°C———————————————————————————————— 10ΩCuRTD –328° to500°F, –200° to260°C———————————————————————————————— 120ΩNiRTD –112° to608°F, –80° to320°C————————————————————————————————

115

Chapter 11

Detectable Problems

Compressors,eitherdynamicorpositivedisplacement,canexperi-enceavarietyofproblems,but therearerelatively fewdiagnostic toolsto identify them. Except in cases where vibration is the issue, a thermodynamicanalysisof a compressor is themost straightforwardway todetermineits health. However, this requires that specific instrumentation be in-stalledontheunit.Thermodynamicanalysisalgorithmswerediscussedindetail inChapter2 -CompressorTheory. Vibration was addressed in Chapter 8. In addition to the itemsdiscussed there, vibration cause and effect is detailed inAppendix H“Troubleshooting.” Fortunately many problems can be avoided before they becomecostly.Mostoftheseproblemscanbeavoidedorminimizedwithproperand timelypreventivemaintenance.Obviouslynotallproblemscanbeavoided.Someproblemsare inherent in thedesign.These fall squarelyon the shouldersof themanufacturer. Asaruleofthumbaproblemthatbecomesevidentduringthefirsttwelvemonths of operationwill be covered under themanufacturer’swarranty.Asusedhere “manufacturer” also applies to the “packager.”The packager is a “third-party” company that installs the compressor,driver(electricmotor,gasengine,gasturbineorsteamturbine),variousvessels,valves,piping,lubricationsystemandcontrolsonoradjacenttothemainskid.Thepackager,oftenreferredtoastheoriginalequipmentmanufacturer (OEM),hasapre-arrangementwith themaincomponent(compressor, driver) manufacturers to package its equipment. Usuallythisarrangementincludesanexclusivityagreementandpricediscounts.In almost all cases themajor equipmentwarranty is a “pass through”from the specificequipmentmanufacturer (compressor,driver, etc.). There are exceptions but once the unit has operated successfullyfor approximately twelvemonths any problem that surfaces is usuallynot covered under themanufacturer’s warranty.When equipment has


not operated successfully during the first twelvemonths of operation,itisimperativethattheowner/operatorandtheOEMworktogethertoidentifyand resolve theproblemorproblems. Whenaproblembecomes evident, regardlessof the compressor’sage or run time, the following basic questions should be addressed assoonaspossible: Is theproblemmechanical, electricalorperformance?

1. Mechanicalproblemsareassociatedwithvibration,sound,orleaks(air,gas,water,oil).Assuchmechanicalproblemsareself-evident.Although the causemaynotbeobvious the symptomsare.

2. Electricalproblemsareassociatedwithhighvoltagetotheelectriccomponents(motors,motorstartersandswitchgear)orlowvoltagetotheinstrumentsandcontrolsystems.Diagnosingelectricalprob-lems isnotaseasyasdiagnosingmechanicalproblems.However,once theproblem is identified thefix is relatively inexpensive.

Figure 11-1. Separable Balanced Opposed 6—Throw Engine Driven Compres-sor in Pipeline Service Complete with Off-skid Mounted Control System.

DetectableProblems 117

3. Performanceproblemsusually result frommechanicaldeficiencieswithin the compressor (engine or turbine driver and electricalproblemswithamotordriver).Theseproblemsmayormaynotbeimmediatelyevidentduringoperation,buttheywillleadtoequip-ment failure.This typeof failure isusually costlywith regards toreplacement parts, labor and downtime. Sometimes downtime isthemost expensive resultof the failure.

Regardlessofthetypeoffailureinmanycasescatastrophicfailurecanbeavoidedbyapplying the followingguidelines:

1. Upon installation run theunit todesign speed.a. Recordtheunit’svibrationsignatureatseveralpointsthroughout

theexpectedoperatingrange.b. Perform a gas path analysis at several points throughout the

expectedoperatingrange.c. Record temperatures on the compressor not monitored by

control instrumentation (such as suction and discharge valvecovertemperatures,coolingwatertemperatures,etc.)

d. Makenoteofthedifferentsoundsemanatingfromthecompressorduringstartup,shutdownandatnormaloperatingspeed.

2. After the unit has run successfully at a stable temperature for atleast four hours shut the unit down and inspect the unit withminimal disassembly (do not remove any close tolerance or closefit components).

Inaxialcompressorssurgeisthesinglemostsevereandexpensivefailureresultingindamagetothebearingsandsealsandsometimestothe impellers, diaphragms, blades and vanes as shown in Figure 11-2and 11-3. Signs of surge are vibration, gas flow instability, gas pres-sure and temperature fluctuation and audible “huffing” or “whoosh-ing” sounds. Surge occurs very rapidly and it can only be avoided ifproper control techniques are applied. Surge was discussed in detailinChapters 6 and7.

Compressor FoulingIndications and Corrective Action Acompressorefficiencydropof2percentisindicativeofcompres-sor fouling.This canbe calculatedas shown.


Figure 11-3. Axial Compressor Rotor Surge Damage

Figure 11-2. Axial Compressor Stator Surge Damage


ηc =

TinCDTc m k

k–1–1

PinCDPc m k

k–1–1

(11-1)

Forthefieldengineeroroperator,compressorfoulingisbest indi-catedbya2percentdrop incompressordischargepressureatconstantspeed and load or throughput.Another indication of fouling is a 3-5percentreductioninloadorthroughputcapacityatconstantcompressorinlet temperatureorambientair temperature.

Corrective Action• Dynamic compressor equipmentmanufacturersnormally specify

cleaning agents. Typically, a liquid-wash is specified. In extremecases complete disassemblymay be necessary to clean the com-pressor.

Note: Impact damage to dynamic compressor blades or impellers canmimiccompressor fouling. Ifperformancedoesnotrecoverafterwash-ing than theunit shouldbe inspected for internaldamage.Manyunitsare equippedwith boroscope ports which facilitate internal inspectionwithminimumdowntimeanddisassembly.

Figure 11-4. Axial Compressor with Contamination Build-up On Blades.


Figure 11-5. Axial Compressor with Contamination Build-up on Leading Edge of Stator Airfoils.

Figure 11-6. Reciprocating Compressor Distance Piece with Piston Rod.


Inaddition reciprocating compressor fouling (due toeitherdebrisor liquids in thegas)mayresult indamage to thesuctionordischargevalves and thepiston rod.Damage to thevalveswillmanifest itself asanincreaseintemperatureofthevalvecovers.Roddamagemaybede-tectedduringoperationasnon-linearmovementof the rodas itpassesthrough the gland at the crosshead-cylinder interface.Also the wearpatternwill becomemore severe as the rodwears against the packing(seeFigure11-7).

Corrective Action• Disassemblevalves, cleanand replacevalveplatesasnecessary• Removepistonand rodassemblyand replace rod.

Loose Piston Theinstallationofthepistonrodontothepistonisadelicatepro-cedurerequiringaspecial lubricant,special techniqueandoftenspecialtools.Ifnotproperlyinstalledthepistonmaybecomelooseonthepistonrod. Thiswill result in the piston hitting the endwall at the head-endand crank-end of its travel and eventually will result in failure of thepistonendwall.

Figure 11-7. Reciprocating Compressor Piston Rod Showing Wear at Location Where Rod Passes through Packing Area.


Corrective Action• Pull the head-end cover and head-end and crank-end valves to

inspect thepiston• Ifnodamage isevident, re-torque thepistonrodnut, re-assemble

the compressorand test.• Replacedamagedcomponentsasnecessary

Bearing Clearance (Hydrodynamic Bearings) Excess bearing clearance can result from improper bearing instal-lationor from journal andbearing shoewearover aperiodof time. Ineither case the effect of bearing looseness can be detected during op-eration by checking the bearingwhirl orbit. Bearingwhirl orbitswerediscussed inChapter8,Vibration. Hydrodynamicbearingwhirl orbits shouldbe checkedat the fac-toryaspartoftheunitruntest(wherethenecessaryvibrationequipmentismost likely to be available).At the factory this problem is relativelyeasy tofix. ReferringtoFigure11-10,theorbitontheleftisastableorbitwhilethe orbit on the right is an unstable orbit. The unstable orbitwill alsodemonstratehighvibration.However, thevibrationmayormaynotbeabovemaximumallowable limits.Continued runningwithanunstable

Figure 11-8, Compressor Piston on Inspection Table.


Figure 11-9. Typical Hydrodynamic Tilt-pad Bearing. This Bearing Pivots on Pins Specifically Sized to Obtain the Proper Bearing-to-journal Clearance.

Figure 11-10. Stable And Unstable Hydrodynamic Bearing Whirl Orbits


orbitwill lead to increases in vibration thatwill eventually exceed ac-ceptable limits and, therefore, shouldnotbe ignored.

Reciprocating Compressor Valves Due to the stresses on the valve plates during normal operationsuction and discharge valves are the highmaintenance items in recip-rocating compressors. Valve plate failure will have a negative impacton compressor performance. Gas path analysis coupled with directmeasurementofthevalvecovertemperatureistheeasiestwaytodetectvalveplate failure.

Corrective Action• Periodic gas path analysis tomonitor compressor performance is

thefirst step indetectingaproblem.• When degraded performance is detected a check of valve cover

temperature can isolate theproblemvalveorvalves.• Replaced thedefectivevalvesas soonaspossible.

Entrained Liquids Liquids in thegaspresentamajorproblemforreciprocatingcom-pressors.Asaminimumtheywillresultinearlyandfrequentfailureof

Figure 11-11. Reciprocating Compressor Discharge Valve Showing Carbon Build-up.


thecompressorvalvesandwhenextremecanresultinoneormorebentrods.Liquidsdonotpresentasbigaproblemfordynamiccompressors,screwcompressorsorblowersasthesecompressorsaremoretoleranttoliquids in thegas. In fact thescrewcompressor isso tolerantof liquidsthat sealing oil is intentionally injected into the gas stream (as in oilflooded screw compressors) prior to compression and extracted aftercompression.Extremely largeamountsof liquidscanresult indynamiccompressor surge.

Corrective Action• Thefirst lineofdefenseagainst liquids isascrubberorknock-out

vessel. This vessel is simply a large areawith a coalescingmediainstalledupstreamofthecompressorwheregasvelocityisreducedto allow the liquid to separate from the gas. Scrubbers are alsoinstalledbetween stages inmultistage compressors.

• Where liquid carryover is severe a cyclone separatormay be em-ployed.A cyclone separator utilizes the gas velocity and internalpassages to createvortex separation.

• Inadditiontoascrubber,linetracingorheatingmaybeemployedto raise thegas temperature tovaporize theentrained liquids.

Figure 11-12. Reciprocating Compressor Showing Piston Rod Installed in Crank End of Piston.


Variable Volume Pocket Plug Thevariablevolumepocketplugispositionedviaascrewmecha-nism.Failureof thescrewmechanism(seeFigure11-16)couldresult inthe inability to reposition the plug. Screw failure could also result inthe plug being sucked into the cylinder where it can becomewedgedresulting in catastrophic failure (seeFigures11-14and11-15).

Corrective Action• Whether the screw is turned manually or is automated via an

electric or hydraulic motor this mechanism should be inspectedperiodicallyandmaintainedper themanufacturer’s recommenda-tions.

Lubrication Oil Lubrication oil should be checked periodically for lubricity, con-taminants and water. For small compressors the unit should be shutdown and the oil replaced. For compressors with large volume reser-voirs the lube oil canbe reconditionedon sitewhile the compressor isoperating.Lubricationoilshouldbestoredcorrectlyinclosedcontainerswith the containers in a rain protected shelter. Figures 11-17 and 11-18show near-new and failed bearing shoes. The failurewas the result of

Figure 11-13. Reciprocating Compressor Showing Failed Piston Rod at Mating Surface with Crank End of Piston.


Figure 11-14. Reciprocating Compressor VVP Failure at Screw.

Figure 11-15. Reciprocating Compressor Showing Failed Distance Piece.


Figure 11-16. Reciprocating Compressor Showing Damage to the VVP Follow-ing Failure of the AdjustWer Screw.

Figure 11-17. Reciprocating Compressor Hydrodynamic Bearing Shoe Show-ing Normal Wear.


rainwater accumulating in the lube oil storage drums and than beingintroducedintothecompressorlubeoilreservoirduringperiodicmain-tenance.

A NOTE OF CAUTION: The photographs used are taken from actual projects. Therefore, the reader may recognize the manufacturer of the equipment shown in these photographs. However, the problems discussed are not uncommon and may involve any manufacturer’s equipment. The reader should not assume that these problems are attributable to a specific manufacturer.

Figure 11-18. Reciprocating Compressor Hydrodynamic Bearing Shoe Showing Distress Following Bearing Failure.

Chapter 12

Controlling Reciprocating andCentrifugal Compressors in

Identical Processes—by Tony Giampaolo

Thiscasestudyaddressesmeldingtogetherthecontrollersoftwoseparate compressor booster systems: the dynamic—centrifugal com-pressorandthepositivedisplacement—reciprocatingcompressor.Thisis accomplished through theuseof a supervisory control system.Thesupervisorycontrollerimplementstheoperators’selectionforPrimaryandSecondarycompressorsystemsunlessaproblemdevelops. If thathappens the supervisory controller can take alternate actionbasedonpre-programmed controller logic. An edited version of this study entitled “A UniqueApproachToControllingReciprocatingAndCentrifugalCompressors InA FuelGasBoosterSystem”waspublishedinWestern Energy Magazine in thesummerof 1994.

INTRODUCTION

A unique supervisory control system has enabled Harbor Co-generation Co., operator of an 80MW cogeneration plant in the LosAngelesarea, tosubstantially increasetheefficiencyof its turbinefuelgas booster system. Normally, plant fuel gas is available at or abovethe 300 psig required by the turbine. However, pressure frequentlydrops into the200psig range,withanoccasional excursionas lowas130psig.Duringthese timesa1,700horsepowerelectricmotor-drivencentrifugal compressor is employed to boost the pressure to at leastthe required 300psia. However,sincethecentrifugalcompressorisdesignedtooperateat a suctionpressure of 119psig, the lowestpressure expected, its ef-

131


ficiencydeterioratesastheinletpressureincreasesabove119psigandless compression is required. In otherwords the centrifugal compres-sor,whichisdesignedtooperateatacompressionratioof2.3,isactu-ally operating,much of the time, at a compression ratio of 1.1 to 1.3.Centrifugalcompressors,whencontrolledwithsuctionvalvethrottlingandgasrecycle,donotsignificantlyreducethehorsepowerconsumedas suctionpressure increases above its designpressure level. In theseinstances the compressor continues topump the sameamountofgas,mostlyintothegasrecycleline,resultinginlostenergy(approximately700BHP)andreducedefficiency.Tomaintainasatisfactorylevelofeffi-ciencyintheplantfuelgasboostersystemwithvaryinginletpressuresthe company has employed a custom designed supervisory controlsystem to integrate the existing motor-driven centrifugal compressorinparallelwith a reciprocating compressor. Thechoiceofareciprocatingcompressorwasmadeafterconsid-eringavariablefrequencydrivefortheexistingcentrifugalcompressor,or a second, stand-alone compressor.

• The variable frequency drive (VFD) would add another controlelement—speed control—and would reduce the need to recyclegasatallbutthehighsuctionpressureconditions(whererecyclewould still be required for surge protection).However, theVFDwould also add complexity as yet another system: a systemwithoutbackup,asystemsubjecttofailure.Inshort,theadditionof the VFDwould improve only efficiency.As a non-redundantsystem, the VFD would have an adverse affect on reliability oravailability. Finally,VFDs are expensive!

• Under varying load conditions, the reciprocating compressoroperates very efficiently at constant speed with suction valveunloading.Themethodprovidesgoodflexibility for theprice asautomaticsuctionvalveunloadersdonotaddsignificantlytothepackageprice. In this configuration, and for this application, thereciprocatingcompressorisslightlymoreexpensivethantheVFD.But as a complete and independent system, it is a 100%SPARE!Power savings of up to 690 BHPwere expected (and have beenachieved)withthereciprocatingcompressorcomparedtothecen-trifugalcompressor.However,addingareciprocatingcompressorin parallel with the existing centrifugal compressor did presentanunusual control interfaceproblem.

ControllingReciprocatingandCentrifugalCompressors 133

DISCUSSION

The reciprocating compressor chosen has automatic suction valveunloading and a small recycle line. The four-cylinder compressor pro-vided eight load steps (two per cylinder), each step representing 12.5percentofthetotalthroughput.Asmallrecyclelinemodulatesthrough-put within this 12.5 percent step to maintain smooth flow to the gasturbine.A simplified flow sketch of the centrifugal and reciprocatingfuelgasbooster compressor inparallel is shown inFigure12-1. The 1,500 horsepowermotor-driven reciprocating compressor op-erates at a constant 880 rpm.Throughput control is accomplishedwithfully automated suctionvalveunloaders and amodulated recycle line.Both the loaders/unloaders and recycle valve are pneumatically oper-atedvia solenoidvalves andan I/P converter, respectively.Themotor,compressor,frameandcylinderlubesystems,coolingsystemforgasandoil and suctionknockoutdrumareall locatedon the compressor skid.

THECONTROLOVERVIEW

In designing this new operating system, Harbor Cogenerationworked with Power & Compression Systems of El Toro, CA. The su-pervisory control system (Figure 12-4) functions as a central controlsystem. Itprovides theoperatorwith theability topre-selecteither the

Figure 12-1. Parallel operation of compressors.


reciprocating or the centrifugal compressor; selection is made at thefront of the supervisory control panel (Figure 12-5). Front panel lightsindicatewhichcompressor,ifany,isrunningandifeithercompressor’smainbreakerhasbeen lockedout.

Figure 12-2. Electric motor-driven reciprocating fuel gas booster compressor package

Figure 12-3. Reciprocat-ing compressor control panel


Undernormaloperatingconditionstheprimaryselectedcompres-sor is the running compressor. In the event the primary selected com-pressor is not available (aborted its start sequence, or experienced anunplanned shutdown), the secondary compressor is started. Selectionof either compressor as theprimaryoneautomaticallyplaces theothercompressor in the secondarymode.

METHODOFCONTROL

General Thestart sequenceforthecompressorsiscontrolledbythesupervi-sorycontrollocatedinthemaincontrolroom.ThiscontrolwasdesignedspecificallybyPower&CompressionSystemstofunctionasthecentralcontrolsystemtointegratethecontrolsofthecentrifugalandreciprocat-ing compressors. It provides the ability to pre-select either the reciprocating or thecentrifugalcompressor;selectionismadeatthefrontofthesupervisorycontrol panel (Figure 12-5). Front panel lights indicatewhich compres-sorhasbeenselectedasprimary.Thesepanel lightsalso indicatewhichcompressor, if any, is running and, if either compressor main breakerhasbeen lockedout. Under normal conditions the primary selected compressor is therunning compressor. In the event the primary selected compressor is

Figure 12-4. Control system interface


not available (abortedits start sequence,or ex-perienced an unplannedshutdown),thesecondarycompressor is automati-cally started. Selectionof either compressor asprimary automaticallyplaces the other com-pressor in the secondarymode.

RECIPROCATINGCOMPRESSORCONTROL

The reciprocatingcompressor is controlledby an Allen-BradleyPLC5-20 mounted in acontrol panel locatednext to the compressorskid (Figure 12-3). ThePLC controls START,LOAD,UNLOAD,STOPandCAPACITYbasedonvaryingsuctionpressureand constant dischargepressure. The controlpanel is located off skidto isolatethepanelfromthe vibration commonwith reciprocating com-pressors. Using analogand digital signals fromthecompressor,thiscon-trolperformsdataacquisition,monitoring,anddiagnosticsof the com-pressorand the control functions.

Figure 12-5. Supervisory control


Automatic Operation In the “compressor-auto-mode” position the controller initiatesthe start sequence as a function of a permissive from the supervisorycontroller,andgaspressure.Onceinitiated,thestartsequenceautomati-cally starts the cylinder lube pump and (after lube oil pressures havebeenestablished)themaindrivemotor.Thepre-lubepumpissettoruncontinuously tomaintainoil temperatureandpressure. Thecompressorstartswithallendsunloadedandtherecyclevalvein the full open (100%-full recirculation) position. After a warm-upperiod, the compressor loadsone, then twocylinder ends in 30-secondintervals until “setpoint” pressure is reached and the recycle valve ismodulating between 20% and 70%. If the compressor remains at thisconditionforanextendedperiodoftime(thatis,therecyclevalvestaysbetween20%and70%),thecontrollerwillloadothercylinderends(andunloadtheinitialends)toavoidheatbuild-upinthecylinders.Thisheat-preclude sequence is repeatedatevery loadstepwhen therecyclevalvemovement stabilizes between 20%and 70%.Each load step is initiatedwhen the recycle valve moves to <20% open position. Similarly, eachunload step is initiated when the recycle valve moves to >70% openposition.At each transition the recycle valve will modulate from theclosed to a partially open position tomaintain smooth flow to the gasturbine.Assuctionpressure increases, thecylinderendswillunload inthe reverseorder.

Manual Operation Inthe“compressor-manual-mode”position,thecompressorcanbestarted and loaded andunloadedmanually. This provision is availableformaintenancepurposesonly.

Remote Terminal Display Alsoprovidedisaremoteworkstationlocatedinthemaincontrolroom.ThisremoteworkstationinterfaceswiththefieldPLC,toprovidecontinuous operating and status information (operating conditions,parameters, and alarms). The remote workstation, a PC-based unit,continuously stores data intomemory andmakes these data availablewhen requested.

Centrifugal Compressor Control The centrifugal compressor is controlled by the plant distributed


controlsystem(DCS)andthesupervisorycontrol.Thiscontrolsystemisaugmentedbyanantisurgecontrollermountedonthecompressorskid.

Automatic Operation Inthe“compressor-automatic-mode”position,thecompressoriniti-atesastartasafunctionofgasturbinesuctionpressureandapermissivefromthesupervisorycontrol.Thecontrollerthenmodulatesthesuctionthrottlevalveandtherecyclevalve tovarygasflowto thegas turbine.The recycle valve also functions as a surge control valve.As suctionpressuredecreases, the recycle valvemodulates to the full closedposi-tion.Assuctionpressurecontinuestodecrease,thesuctionvalveopens.

Manual Operation In the “compressed-manual-mode” position, the recycle valvemust be placed in the full opened position and the suction throttlevalve in the near closed (it never goes full closed) position prior tostart. Once started, the recycle valve can be closed, and the suctionthrottle valve opened until the required throughput is reached. Caremust be taken that the valve is moved slowly to avoid driving thecompressor into surge.

Figure 12-6. Remote Workstation


POTENTIALPROBLEMSTHATWEREAVOIDED

Severalpotentialproblemswereanticipatedandstepswere takenin thedesign stage toaddress these.

1) As each cylinder end is loaded, that volume of gas is sent to thegasturbine.Thisincreaseinflowcouldresult inover-temperatureand/or instability. Either conditionwould be very detrimental tothegas turbine,andtoavoid themthe followingcontrolelementswereutilized:a) Time delays were incorporated into the cylinder loader/

unloaderlogictofacilitateadjustingeachsequenceduringfieldlooptuning.

b) Using open-loop response, the recycle valve was preposi-tionedprior to each sequence change.Also, timedelayswereincorporated into the recycle valve logic to facilitate fieldadjustments.

During commissioning, these timers were easily adjusted tominimizeimpactonthegasturbine.

2) When flow, suction and discharge pressure remain constant, thecylinderendsmustbe sequentially loadedandunloaded inordertopreventheatbuild-upintheunloadedcylinders.This“heatpre-cludesequence”isrepeatedoverthe8-cylinderends(andthrough-out the various load steps) so that no cylinder end is completelyunloaded long enough for temperature to rise to anunacceptablelevel. To complicate this task, the torsional study recommendedagainst loading certain end configurations (e.g. 5 ends loaded =withall4headendsloaded,loadcrankend#3or#4,never#1or#2).a) Eachheatprecludestepwasprogrammedintothelogicperthe

resultsofthecompressormanufacturer’storsionalstudy.Sincetheactualheatbuild-uptimewasuncertain(themanufacturerclaimedaminimumof15minutespercylinderend)timerswerebuiltintoeachlogicstepsothattheycouldbeeasilyadjustedinthefield.Thisprovedtobemorehelpthanoriginallyanticipated,sincetheheatbuild-upwasactuallycloserto10minutesatthe2and3endsloadedconfigurations.


3) The recyclevalvehad tomove fast enough to compensate for theactionof thesuctionvalve loader/unloaders.Theactuationspeedoftheloader/unloaderswasnotknown,eventothemanufacturer.However, therecyclevalvecouldbemadetomove, fullstroke, inless thanone second.a) Avolumeboosterwasdesignedintotheoperationoftherecycle

valvetoimproveitsresponsetime.Recyclevalve“fullstroke”speed of response was initially about 5 seconds. This wasconsidered too slow as the loader/unloader responses, whenmeasured,waslessthan1second.Tomatchtheresponsesoftheloader/unloader valves and the recycle valve itwas easier toimprovetheresponseoftherecyclevalve.

b) By adjusting the timers in the suction valve loader/unloaderlogic,valveoperationcouldbestaggeredslightly.

4) The supervisory control had to insure that one andonly one com-pressorwasallowedtostartinautomatic.Furthermore,thesecondarycompressorhadtostartintheeventthattheprimarycompressordidnot start as planned. Finally, regardless of the other compressor’sstatus (shutdown, running, etc.) each compressor had to have theabilitytobestartedmanuallyformaintenancepurposes.Figure12-7showssomeoftheladderlogicemployedinthesupervisorycontrol.

5) Control loop tuningwas accomplishedwith the aidof a 20-chan-nel high-speed recorder. The recorder indicated themovement ofeach unloader (16 channels), the recycle valve, and suction anddischarge pressure. Initially the control gain and reset rate wereset.Once thisprovideda stableoperationat each load level, thentheunloaderswerealsocontrolledbyutilizingopenloopcontrolofthe recycle steps. The load stepswere also controlled byutilizingopenloopcontrolof therecyclevalve.Allof thesevariations,andadjustmentstoindividualcontrols,weremonitoredwiththehigh-speed recorder. The recorder (Astro-MedMT95K2) provided easyprogramming,calibration,chartspeedsselection,andlaserprinterresolution. The chart speeds utilizedwere 1mm/sec, 5mm/sec,and25mm/sec.

Proportional gain and reset rate was calculated using theZiegler&Nicholstechniqueforclosedlooptuning.Initialandfinalrecyclevalve responseare shown inFigures12-8and12-9.

ControllingReciprocatingandC

entrifugalCom

pressors141

Figure 12-7. Supervisory Control Ladder Logic

142

Com

pressorHandbook:PrinciplesandPractice

Figure 12-8. Initial Recycle Valve Response

ControllingReciprocatingandC

entrifugalCom

pressors143

Figure 12-9. Final Recycle Valve Response


CONCLUSION

Performance tests conducted during commissioning confirm theBHPatthedesignconditionwasasexpected(410BHP).Thisisasignifi-cantimprovementcomparedtothe1100BHPrequiredbythecentrifugalcompressorat the samedesign conditions. Variations in pressure ratio and throughput are accomplished byautomatic unloaders installed on each suction valve. This reduces thetotalmotorloadrequiredcomparedtothecentrifugalcompressorathighsuctionpressures(at lowsuctionpressuresefficienciesarecomparable).Also each load step (8 steps total) represents 12.5% of the throughput.Therefore, thesizeof therecycle lineandvalvearereduced, improvingefficiencybyeliminatingexcessiverecycle,andthesmallerrecyclevalvecontributes to improvingvalve response time. While each case isdifferent and shouldbe consideredon its ownmerits,thecontroltechniquesemployedherecanbeutilizedoncompres-sors(singleormultipleconfigurations,operatinginseriesorparallel)inanyenvironment that requiresvariation inpressureand throughout.

145

Chapter 13

Optimization & Revitalization ofExisting ReciprocatingCompression Assets

—W. Norman Shade, PEACI Services Inc.

Compressors,withoutadoubt,havethe longest lifeexpectancyofanypieceofmechanicalequipment.Thisstudydiscusseshowcompres-sorscanberevitalizedandreturnedtoservice(seealsoFigures3-16and3-41—Chapter3—foranexampleof compressor reconditioning).————————————————————————————————

ManyreciprocatingcompressorsinNorthAmericaandthroughoutthe world have been in service for decades, some operating for morethan half a century. Most older models are no longer optimal, andmany are not dependable for current process requirements.Over time,technology has improved and energy costs have increased, making iteconomically attractive to optimize and revitalize existing compressionassetswith components that take advantageofmodern technology.Ef-ficiency,reliabilityandmaintainabilitycanbegreatlyimprovedthroughcustom engineered solutions. Changes may range from conversion tobettercompressorvalves,totheadditionofproperlydesignedautomaticunloadingdevicesforbettercontrol,totheinstallationofnewpurpose-built compressor cylinders and components, to the complete reapplica-tionof anentire compression system toanewservice. Even relatively new compressors can become inefficient andmis-matchedasproductionneeds change. Inother cases, from the time theunits were installed, actual operating conditions never quite matchedthe design conditions specified when the equipment was ordered. Instill others,manufacturersmay havemissed themarkwhendesigning


and applying their products, so that the equipment underperforms orisunreliable for its intended service. Whateverthereason,changingvalvesorcylinders,addingunload-ersandautomaticcontrol,orreapplying,existingreciprocatingcompres-sors is frequently necessary and justifiable. In the case of newer com-pressors, theOEMsmayhavestandardcylinders thatcanberetrofittedtotheir frames,howeverthatusuallyrequiresexpensivechangestotheprocesspiping,pulsationvessels,andcylindermounting.Inthecaseofmaturecompressormodels thatareno longer inproduction,newOEMcylinders tendtobeexpensiveandhave long leadtimes, ifavailableatall. In some applications, when new compressors are selected or usedcompressorsareredeployed,thestandardOEMcylinderline-upmaynotincludetheoptimalcylinderboresizeorworkingpressureratingforthedesiredoperatingconditions.Anyof thesesituationsarecandidates forcustom engineered cylinders and unloading devices that provide cost-effective solutions forpreserving the value and reducing the operatingcostof existing compressionassets.

NEWCUSTOMENGINEEREDREPLACEMENTCYLINDERS

With over 30 years of designing and manufacturing custom en-gineered compressor cylinders, internal components, valves, and un-loading systems, the author’s company has helped a large number ofcompressorusers improvetheircompetitiveedgebyprovidingcreativesolutions that improve their existing compression assets to improveperformance, efficiency, reliability, maintenance cost, and safety. Theseefforts have provided solutions to operating problems, have improvedthe performance and have increased the reliability of reciprocatingcompressors.Theextensiveexperience indesigningandmanufacturingcustom engineered compressor cylinders is represented in Figures 13-1and13-2.Customengineered cylindershavebeenappliedonallmajorbrands of compressors ranging from small 3.0 in. (76.2 mm) stroke,1800 rpmhigh-speedmodels to giant 20.0 inch (508.0mm) stroke, 330rpm frames,withbore sizes ranging from1.33 to36.0 in. (33.8 to914.4mm),andworkingpressuresfromaslowas50psig(3.5bar)toashighas 12,000 psig (827.4 bar) for upstream, gas transmission, gas storage,chemicalprocess, refinery,high-pressureair andother services. Newcompressorcylinderscanbecustomengineeredandpurpose

Optimization&RevitalizationofExistingAssets 147

built tomeetcustomerspecificationsortosolveproblemswithexistingOEMdesigns.Inmostcases,thecylinderscanbe“bolt-in”replacementsto match existing process piping connections andmounting locations,which avoids expensive system piping and foundation modifications.Cylinderscanbejacketed(watercooled)ornon-jacketed(aircooled)andmadefromappropriatematerialfortheapplication.Ductileironcastingsare most commonly used in recent years, although gray iron, steel orstainlesscastingsarealsoused,asaresteelandstainlesssteel forgings.Reducedhardnessmaterials areavailable for sourgasandotherappli-cationsrequiringNACEspecifications.Cylinderboresmaybesuppliedin the virgin material condition, ion nitride hardened for improvedwear resistance, or lined with a replaceable sleeve. Depending on therequirements, existing internal components may be reusable, howevernewvalves,pistonrodpackingandmaintenancefriendlypistonandrodassemblies are often included.Awide range ofmanual and automati-callycontrolledclearancevolumepocketsandenddeactivatorsarealsoavailable, customengineered for the specificapplication requirements.

Case History 1: Increased Pipeline Compressor Capacity In the first example, a mid-continent USA natural gas pipelineoperator had several existing Clark HBA and HRA integral enginecompressors with cylinders that no longer served the operating needs

Figure 13-1. Custom cylinder experience—bore dia. vs. stroke


of the compressor station. The customer specifications dictated an ap-proximate 32% increase in bore size in a new cylinder that providedmaximum operating efficiency. In addition, the cylinder was to haveexternal dimensions that permitted the use of the existing piping sys-tem, cylinder supports and foundation. It was readily apparent that aconventional cylinderdesignwouldnevermeetallof the requirementsof the specification. The OEM, who supplied the engines, agreed, butoffered a conventional design that did notmeet the specifications. TheFour-Poster™cylinderofferingwassufficientlyuniquethattheauthor’scompany receivedapatenton thenewdesign. Inadditiontomeetingallthespecificationrequirements,thecylin-derswereofferedata lowerpriceandshorterdelivery thanofferedbytheOEM. In thisdesign, all valveswere accessible from the topof thecylinderandthefrontandrearhead,pistonassembly,linerandpackingcase were accessible from the front of the cylinder.A total of 42 new12.50 inch (317.5mm)bore,1000psig (69.0bar) cylinderswere retrofit-ted ontomultiple 17.0 in. (431.8mm) stroke compressors. Closed looptestscertifiedtheefficiencyofthecylinders.Significantsavingsresultedfrom the re-use of the piping and cylinder supports and the absenceof foundation changes. Subsequent to the supplyof these 42 cylinders,otherorderswerereceivedforfourmoreidenticalcylindersandsixteencylinders with slight modifications for another OEM’s 14.0 in. (355.6

Figure 13-2. Custom cylinder experience—pressure rating vs. bore diameter


mm) strokeengines.All cylindersdelivered trouble free service.

Case History 2: Improved Service Life AUSAGulfCoastchemicalplanthadaWorthingtonBDCoffgascompressorwithcastironcylindersthatwereconstantlyunderchemicalattackfromextremelycorrosivegas.Cylinderlifewastypicallyonly3to5years, and theOEMhadnot resolved theproblem to the satisfactionofthecustomer,whorequiredafunctionallyequivalentcylinderdesignthat would directly replace the existing cylinder to the extent that allotherpartsoftheassemblycouldbere-usedwiththenewcylinderbody. The customerhadattempted toproduce replacement cylinders ofthe samegeometry in ani-resistmaterial, however, the cylindergeom-etry and jacketed design were not compatible with ni-resist foundrypracticesandrequirements.AsshowninFigure13-4,theauthor’scom-pany redesigned the exterior of the jacketed, 400 psig (27.6 bar), 12.0in. (304.8mm)bore,9.5 in. (241.3mm)strokecylinderbodytosimplifyand enhance foundry procedures. Proper core support and clean outand pattern orientation in the mold resulted in sound castings whenproduced inASTMA316 series stainless steel.

Figure 13-3. Retrofitting custom engineered cylinders increased bore size by 32% on this 17 in. (431.8 mm) stroke Clark integral compressor.


ThechangetoA316stainlesssteelresolvedthecorrosionproblemsand eliminated the periodic safety inspections that had been requiredwith the original cast iron cylinders. For more than 20 years, the re-placementcylindershavebeenmoreproductive,andmaintenancecostshavebeenreduced.Similarreplacementsofthistypeweresubsequentlyprovided for 19.5 and 7.75 in. bore (495.3 and 196.9mm) cylinders foradjacent stages.

Case History 3: Providing Safe and Reliable Production Operation A large Cooper Bessemer V250 integral engine compressor inanAustral-Asian fertilizer production plant had experienced repeatfailures of the 3rd stage syn gas cylinder. The 6.50 in. 5 (190.5 mm)lined bore diameter, 20 in. (508.0 mm), 6000 psig (413.7 bar) forgedsteelcylinderwasadesignthatemployedtransversetie-boltsforrein-forcement.Afterseveralyearsofservice,crackswouldoriginateinthedischargevalvepockets andmigrate to the tie-bolt clearanceholes asshown in Figure 13-5.A thorough engineering analysis revealed veryhighstressesandunacceptablefatiguesafetyfactorsinthevalveseatsas shown inFigure 13-7.

Figure 13-4. This custom engineered ASTM A316 cast stainless steel cylinder eliminated a severe corrosion problem with the cast iron cylinder that it replaced.


Since the high-pressure 3rd stagewas normally operating at 93%of rated rod load, and the 1st and 2nd stageswere lower at 77% and59%, respectively, a performance study was completed to determinean optimal bore size for the new 3rd stage cylinder. Interacting withthe plant engineering staff, it was agreed that the new cylinder borediametercouldbereducedto6.00in.(152.4mm),whichcutthenormaloperating rod load to 84% of rated, while still maintaining acceptablerod loads of 79% and 68%, respectively, on the 1st and 2nd syn gasstages.Itwasalsodeterminedthatthelowmolecularweightpermitteda reduction in valve diameter,which significantly reduced the preloadon the valve cap bolting,while also increasing the section thickness inthe critical cornerswhere thevalvepockets intersected themainbore. A new 6.00 in. (152.4mm)AISI 4140 forged steel bolt-in replace-mentcylinderwasdesignedandextensivelyanalyzedtoachieveaccept-able fatigue safety factors. Thenew jacketed cylinder, shown in Figure13-6,was fullymachined from a solid block forging thatwas 3 in. (76

Figure 13-5. 3rd stage syn gas cylinder showing location of leak detection.


mm) thicker than the original cylinder. The problematic tie-bolts wereeliminated, and the critical valve pocket fillet geometry was carefullycontrolled and then shot peened aftermachining to provide beneficialcompressive preload stresses.Water jacket side covers were fully ma-chined fromaluminumplatematerial to eliminate a corrosionproblemthatwasprevalentwith theoriginal cast iron side coverdesign.

Case History 4: Eliminating Extremely Serious Piston Failures Amajoroil companywas experiencing frequent and rapidpistonfailuresonthe2ndstageofareapplied4000hp(2983kW)SuperiorW76compressor used for sour gas lift service on an offshore platform.Notonlywasdowntimecausingsixfiguredailyproductionlosses,operatingpersonnelwerefearful that thepistonfailurescouldleadtoabreachofthelethalsourgas,whichwouldbeextremelydangerousonanoffshore

Figure 13-6. New AISI 4140 forged steel cylinder with water jacket, weighing more than 10,000 lb. (4335 kg).


platform.TheOEMandanothercompressorservicecompanyhadbothtried to solve the problemwith new pistons, but failures continued inaslittleas8daysofoperatingtime.Uponbeingconsultedforhelp,theauthor’scompanyconductedathoroughassessmentandfoundthatveryhigh2ndstageoperatingtemperature,coupledwithrodloadsthatwereatorabovethemanufacturer’srating,weretherootcausesoftherapid,catastrophicfailuresofthe18.0in.(457.2mm),7.0in.(177.8mm)strokemulti-piecealuminumandductileironpistons.AsshowninFigure13-9,the cylinderwas anunusual shape that resulted in avery shortpistonlengthrelativetothediameter.Afiniteelementstressanalysis,asshownin Figure 13-8, showed that the original piston design had a fatiguesafetyfactoroflessthan1.0attherequiredoperatingconditions,whichexplainedtherapidfailures.Unfortunately,thealternativesforresolvingtheproblemwerevery limited. The short cylinder andOEM limits on thepermissible reciprocat-ing weight precluded the use of a stronger, less temperature sensitiveferrousmaterial piston.Adding interstage cooling to reduce the 194°F(90°C) suction temperature was not practical on the offshore platformdue to space limitations. The OEM did not have an alternative cylin-der that would solve the problem, and lead time for a complete new

Figure 13-7. Finite element stress analysis of high-pressure cylinder body showing high stresses in valve seat fillets.


replacement compressor package was about 18 months at the time.Acustomengineeredreplacementcylinderwasdeterminedtobe thepre-ferred solution, but lead time was also prohibitively long. Concurrentdesignandanalysiseffortsestablishedthataone-piecesolidaluminumpiston,althoughnothavinginfinitelife,wouldoperateforalongerandreasonably predictable period of time. This was adopted as a tempo-rarymeasure until a new replacement cylinder could be designed and

Figure 13-9. Original 18.0 in. (457.2 mm) OEM 2nd stage cylinder [left] that ex-perienced frequent rapid, catastrophic piston failures due to a mis-application.

Figure 13-8. Multi-piece Piston finite element stress plot.


manufactured.Withtheone-piecepistonsprovidingadditionaltimefordevelopment of a permanent solution, further review of the operatingconditionsof the 1st and2nd stages showed that furtherbenefitswereachievableby changing the two1st stage cylinders alongwith the 2ndstage.Bychangingthetwo1ststagecylindersfrom20.0in.(508.0mm)to21.0in.(533.4mm)andthe2ndstagefrom18.0in.(457.2mm)to18.125in.(460.4mm),abetterbalanceofrodloadsandinterstagetemperatureswasachieved,whileprovidingthecompressorwithanincreasedcapac-ity rating thatwasanunexpectedbonus for theoperator. All three of the newASTMA395 ductile iron cylinderswere de-signed longer to accommodate longer ductile iron pistons that werecapable of the required loading. In order to do this while also main-tainingprocesspipingconnectionpoints, specialcastductile ironoffsetadaptersweredesignedandmanufacturedforthesuctionanddischargeconnections.Thepistonswerehigh-performance, two-pieceductile ironcastings with internal ribs. Both the cylinders and the pistons wereextensively analyzedwith finite element stressmodels. StandardArielvariable volume clearance pockets were adapted to the new 1st stagecylinders.Thenewcylindersreceivedhydrostatictestsat1.5timeswork-ingpressure,followedbyheliumsubmersiontestsatworkingpressure.Field-installedontheoriginalframeasshowninFigures13-10and13-11,thenewcylindershavecompletelyeliminatedthepistonfailures,whileproviding the production platformwith several percentmore capacitythanwasachievablewith theoriginalOEMcylinder configuration.

Case History 5: Increasing Compressor Efficiency and Throughput Asingle-stage,gasenginedriven,DresserRand6HOS-6compres-sor was applied in natural gas pipeline service. The operating condi-tionsof thepipelinewere such that theenginedriverwas fully loadedmaking it the limiting factor for throughput of the compressor station.Performance tests of the original OEM cylinders, which were misap-plied, showed thatvalve losseswereveryhigh for the lowratioappli-cationbecauseof insufficientvalve sizes andgaspassages in theOEMcylinders being used on the frame. New 10.5 in. (266.7 mm) bore, 6.0in. (152.4mm) stroke, 1200psig (82.7 bar) bolt-in replacement cylinderassembliesweredesignedwith largervalvesandgenerous internalgaspassages.As shown in Figure 13-12, the new cylinders were installedinplaceof theoriginal cylinders in less than twodays,maintainingallexistingprocesspipingconnectionsandmounting.Performancetestsof


LEFT: Figure 13-10. New 18.125 in. (460.4

mm) bolt-in 2nd stage replacement cylinder.

BELOW: Figure 13-11. New 21.0 inc. (533.4

mm) bolt-in 1st stage replacement cylinders

with manual vari-able volume clearance

pockets.


thenewcylindersshowedanefficiencygainofabout10%,betteringtheguarantee of 8% and leading to a corresponding 10% throughput gainfor the compressor station.

Case History 6: Increasing Cylinder Working Pressure AUSAnaturalgasstoragefacilityhadarequirementforincreasedoperatingpressure.ThecompressionwasprovidedbyfourCooperBes-semer GMVA-10 integral engine compressors each having four olderstyle vertically-split valve-in-head cylinders. The OEM, as well as theauthor’s company,wasasked to re-rate theoriginal cylinders;howeverwith largegasketed jointsasshown inFigure13-13, thecylinderswereonlymarginallyadequatefortheiroriginalpressureratingandthereforeincapable of further uprating. The OEM offered new higher-pressurecylinders, however, they would require major modifications to theASME coded pulsation bottles and process piping, greatly increasing

Figure 13-12. The first of six new high-efficiency bolt-in replacement cylin-ders [left] is installed on a six-throw, single-stage D-R 6HOS-6 frame. Two of the original OEM cylinders are shown on adjacent throws [right] prior to replacement.


thecostandthe installationtimefortheproject.Theauthor’scompanymeasured critical interface dimensions on site without removing theold cylinders and thendesignednew12.0 in. (304.8mm)bore, 14.0 in.(355.6 mm) stroke, 1200 psig (82.7 bar)ASTMA395 cast ductile ironbolt-inreplacementcylindersasshowninFigure13-14.Thedesignwasable to accommodate theuse of the originaldoubledeckvalves, pack-ing cases and automatic fixed volume clearance pockets.As the newcylinderswerecompleted,theoriginalvalves,packing,andpocketswerereconditioned, and the valve pocket volume bottles were re-rated forthe higherworking pressure. The field conversionwent smoothly andthecustomer’s facilitywasavailable,onschedule, tohandle thehigheroperatingpressure. AnotherUSAnaturalgasstoragefacilityrequiredincreasedoperat-ingpressure.Thefacilityhadfive1940svintageCooperBessemerGMVcompressorswithvalve-in-headcylinders.AsshowninFigure13-15,theoriginalcylindershadtwoseparatesuctionandtwoseparatedischargeconnections,andthecylinderswereverticallysplitsothattheoutboardsuctionanddischargeflangeconnectionshadtobedisturbedwheneveritwas necessary to access the piston, rod or packing formaintenance.Although theOEM indicated that amodestpressure re-rate of the cyl-inders might be possible, the station operators were concerned aboutthe safety since the existing cylinders sometimes had gasket leakageproblemsat theoriginalpressurerating.Theywantedtoavoidanygaspiping or pulsation bottle changes and they wanted higher pressurecylinders that eliminated theproblematic gaskets and the cumbersomeandcostlymaintenanceprocedures that requireddisturbinggaspipingconnections toaccess thecylinderbore. Inorder tomeet thischallenge,a special valve-in-barrel cylinder design was developed, and 20 new7.0in.(177.8mm)bore,14.0in.(355.6mm)stroke,2000psig(137.9bar)ASTMA395castductileiron,jacketed,cylindersweremanufacturedandinstalledon theexisting compressorsas shown inFigure13-16.Valves,packing andunloaderswere reconditioned and reused in the new cyl-inders.

Case History 8: Emergency Replacement of Obsolete Cylinders A chemical plant had an emergencywhen a critical PennsylvaniaATP off gas compressor unexpectedly developed internal process gasleaksintothecoolingwaterjacket.TheOEMnolongerhadtheoriginalcylinderbodypattern,sotheirleadtimeforanewcylinderapproached


Figure 13-13. Original OEM 1100 psig (75.8 bar) valve-in head type cylinders.

Figure 13-14. New custom engineered 1200 psig (82.7 bar) bolt-in replacement valve-in-barrel type cylinders.


Figure 13-16. New bolt-in replacement valve-in-barrel cylinders reused origi-nal valves, packing and cylinder unloaders. Piping connections are no longer disturbed for piston, rod, and packing maintenance.

Figure 13-15. Original problematic valve-in-head cylinders required disturbing the outboard piping connections to access the piston, piston rod and packing.


oneyear.Sincelossofthiscompressorrequiredshuttingdowntheentireplant,emergencytemporaryrepairsweremadetothecylinder,andtheauthor’s company was contacted to provide a new cylinder. Criticalinterfacemeasurementsweremade on site, and a new design 20.5 in.(520.7mm),11.0 in. (279.4mm)stroke,150psig (10.3bar)cylinderwasexpedited fromASTMA395 cast ductile iron to strengthen the weak,corrosion sensitive sections. The new cylinder was delivered in only20 weeks, including the time for design, pattern construction, casting,machining, hydrostatic testing and installation of a new ni-resist liner(Figures13-17&13-18).

Case History 9: Extending OEM Standard Cylinder Offerings AnupstreamproductioncompanyhadarelativelynewArielJGJ/6compressorused for acidgas injection to stimulateoilproduction.The600 hp (447 kW), 1500 rpm compressorwas originally configured as a5 stageunit.Once thefieldwas operating, however, it became evidentthatproduction couldbe enhancedby increasing the injectionpressureto a level that exceeded the capability of the compressor as originallyconfigured.Itwasdeterminedthata6thcompressorstagewasrequired,but the OEM did not have a suitable high-pressure cylinder for that

Figure 13-17. Liner installation. Figure 13-18. Finished cylinder.


particular frame. Inaddition,theaggressivenatureoftheacidgasrequiredtheuseofmaterialproducedtoNACEspecifications;andthelimitedphysicalsizeconstraints required a highermaterial strength than could be achievedwiththeuseofreducedhardnessalloysteels.17-4PHstainlesssteelwasthereforethematerialofchoiceforthenewcylinder.Withtheapprovalof theOEM, the author’s companydesigned andmanufactured a spe-cial 2.75 in. (69.9 mm) bore, 3.5 in. (88.9 mm) stroke, 3000 psig (206.8bar) cylinder.The tail roddesign,which is shown inFigure13-19,wascompletely machined from a solid 17-4PH double H1150 heat treatedstainlesssteelforging.Thiscylinderwasaddedtotheoriginalcompres-sorasa6th stage to increasefinal injectionpressure.

Case History 10: Filling gaps in OEM Standard Cylinder Offerings A distributor packager of new Gardner Denver and remanufac-tured Joy 7.0 in. (177.8mm) strokehigh-pressure air compressorsusedin portable diesel engine driven compressor packages for the rotarydrillingindustrywasencounteringapplicationconditionsforwhichtheOEM cylinders were not optimally suited. Given the maturity of thecompressorproductline,theOEMwasnotveryinterestedinmodifyingtheexisting cylinderdesignsor creatingnewcylinderdesigns. Theauthor’scompanyreviewedtherangeofrequiredapplicationconditions, measured the mounting interface dimensions, and subse-quentlydevelopedafamilyofnew7.0in.(177.8mm)cylindersthatfitthecompactvee-typeJoyandGardnerDenvercompressorframesandthatwereoptimallysizedfortheairdrillingapplicationrequirements.

Figure 13-19. Special 3000 psig (206.8 bar) 17-4PH stainless steel tail rod cyl-inder designed for acid gas injection service.


Theboresizesof theASTMA395castductile iron, jacketedcylinders,ranged from3.5 to4.5 in. (88.9 to114.3mm)withaworkingpressureof2500psig(172.4bar).Cylindersarenowsuppliedonaregularbasistomeetthisdistributorpackager’sneeds,andadditionalboresizesarebeing considered to further expand the range of pressures and flowrates offered to meet the needs of the drilling market as shown inFigure 13-20. Manymore examples canbeprovided, however, the forgoing listshows the various kinds of applications for which new custom engi-neered,purposebuilt reciprocating compressor cylinders canbe attrac-tive alternatives for optimizing and revitalizing existing compressionassets.

RE-APPLIEDCOMPRESSORCYLINDERS

Although new, custom engineered and purpose-built cylinderscan provide optimal performance and reliability, there are frequent

Figure 13-20. 2500 psig (172.4 bar) cylinder family developed to fill a gap in the OEM’s offerings for high-pressure air drilling service.


examples where surplus or previously decommissioned cylinders canbe reconditioned and reapplied onto existing compressor frames.NewOEMcylinderscanalsobeinstalledondifferentOEMframesthathavesimilar strokes and roddiameters.Of course, diligent and experiencedengineeringdesignisrequiredtomatchthedifferentmounting,bolting,pistonrodand,ifthestrokeisdifferent,thepiston.Considerationmustalso be given to matching reciprocating weights with the compressorframe’s capability.Maximum pressure ratings, Internal gas and inertiarod load ratings, and pin reversal requirementsmust be considered inthe reapplicationaswell.Doneproperly, reapplicationof cylinders is areliablealternativethatcanoftenreducethecapitalcostandwillalmostalwaysreducethecylinderleadtime.Severalexamplesdemonstratethepotentialof this engineeredapproach.

Case History 11: Adapting RemanufacturedCylinders to An Existing Compressor Anaturalgaspipeline transmissionstationhadtwo1950svintageI-RKVG10/3 integralenginecompressors in service.Althoughhavinga provision for three compressor throws, only two of the throwswerepreviouslyequippedwithcylinders, thecenter throwbeingunused formore than 50 years. Conditions had changed over that period of timesothatthe600hp(447kW),330rpmcompressorswerenolongerfullyloaded. Thepipeline operator haddetermined that, by adding anotheridenticalcylindertotheinactivethrowinparallelwiththeexistingcyl-inders in a single-stage configuration, the engine couldbe fully loadedto increase the station’s capacity. Inquiriesweremade to theOEMandseveral other companies. The OEM still had the original pattern andwas able to quote new identical cylinders, however lead time for thatoptionwasabout40weeks.Theauthor’scompanyalsoofferedtodesignandmanufacture new bolt-in replacement cylinders, however, the bestpossible leadtimewas26weeks.Neitherof thesealternativesweread-equatefortheoperator’srequirementtohavetheconversioncompletedso that both compressorswould be in servicewithin nomore than 20weeks after the release of the order.An extensive search for identicalused or surplus cylinders produced no results either, although the au-thor’s companywas able to find two similar 15.0 in. (381mm) strokeI-R process cylinders, Figure 13-21, that had the sameflanges anddis-tancebetweenflanges.Itwasdeterminedthatwithmodificationstothecombination crank end head and distance piece, the process cylinders


could be adapted to fit the existing KVG compressors.A performancemodel,Figure13-22,showedthatthecylinderswouldmeettherequiredoperating conditions if all cylindersweremodifiedby lining to a 10.16in. (258.1mm)borediameter. Having no time to lose, two used cylinders were purchased andtaken to the factory for cleaning, inspection and hydrostatic testing toconfirm that the cylinder bodieswere sound. The cylinderswere thenmodifiedtofullyrestoreallfitdimensions,sealingsurfacesandthreadstoensuretheirintegrityforthereapplication.Allnewbolting,sealsandgasketswereinstalled.Aftermachining,asecondhydrostatictestoftheremanufacturedcylinderswasconductedat1.5timesmaximumworkingpressure, as shown inFigure 13-23.Concurrentwith the completionofthe “new” remanufactured cylinders, new liners, non-lube piston androd assemblies, double deck poppet valves, valve cages, valve caps,outer end automatic fixed volume clearance pocket heads, and highlyefficient radial poppet suction valve deactivators were designed andmanufacturedforthefourexistingcylindersonthetwocompressorsaswellasforthetworemanufacturedcylinders.AcompleteAPI618designapproach 2 pulsation study was also conducted on the entire system

Figure 13-21. One of two used I-R cylinders prior to the remanufacturing process.


withthenewconfiguration.Thisstudyrevealedhighpressuredropsintheexistingplantsuctionpulsationcontrolarrangementwiththehigherflow, three cylinder configuration, andmitigating actionswere quicklyidentifiedandadded to the conversionplan. Whenallcomponentsandcylinderassemblieswerecompleted,thestationwastakenoff linefortheconversion.ThesuctionanddischargebottleswereremovedandtakentoanASMEcodeweldingshopforad-ditionof the incremental connections, followedby recertification.Somesimple piping modifications were also made to mitigate the pressuredrop problem that had been revealed by the pulsation study. Usingspecialfieldmachiningequipment,theexistinglinersweremachinedoutof the existing cylinderswithout removing them from the compressor.New liners and the other new components were then installed alongwith the incremental remanufactured cylinders. The revised bottleswere reinstalled and the systemwas completed and commissioned onschedule.Oneof the completedunits is shown inFigure13-24.At thistime the compressorshavebeen in service forover3years.

Figure 13-22. Performance model for 3-cylinder, single-stage I-R compressor showing all load steps required for system flexibility without overloading at the maximum specified discharge pressure.


Figure 13-23. Remanufactured cylinders with new fixed volume clearance pocket heads and valve caps undergoing an 8-hr factory hydrostatic test.

Figure 13-24. One of two reconfigured KVG units after adding a 10.16 in. (258.1 mm) diameter remanufactured cylinder [center] and adding new clearance pockets, radial poppet suction valve deactivators, and other new components to all the cylinders.


Case History 12: Mixing Used OEM Cylinders andFrames to Optimize Performance Anaturalgasprocessingplantwantedtoreapplyanexisting2000hp(1491kW),900rpmgasengineandcompressorpackageforapropanerefrigeration process. Process conditions were fixed, so that cylindershadtobecarefullyselectedtomatchtheoperatingrequirements.Usingthepreferred6.0in.(152mm)strokeSuperiorWH64compressorframe,initial performance sizing resulted in a selection of 25.5 in. (648 mm)diameter Superior cylinders for both the 1st and 2nd stages, howeverthe 2nd stage cylinderswould have had to operate very close to theirworking pressure limit. The same concern existed for the 20.0 in. (508mm)3rdstagecylinder.TheOEMwasnot interestedinproducingspe-cialcylindersforasingleunit,andnousedcylinderswithanadequateworkingpressure couldbe found. As a result, a new 6.5 in. (165mm) strokeAriel JGC/4 compres-sorwas considered,but this resulted in concernabout the reliabilityofthevalveswiththeheavypropanegas.Switchingtoa5.5 in. (140mm)strokeAriel JGD/4compressormadethevalvereliabilitymorepromis-ing,but itcouldnotproduce therequiredflowwith theavailablestan-dardArielcylinders.ItwasfinallydeterminedthatthesolutioncouldbeoptimizedbyadaptingArielJGDcylinderstofitontheSuperiorWH64compressorframe.Ariel22.0in.(559mm)diameterJGDcylinderswereselectedfor the1ststageandthe two2ndstagecylinders,andanAriel19.625in.(499mm)diameterJGDcylinderwasselectedforthe3rdstage.UsedArielcylinderswerefoundthatcouldbereconditionedforallbutone of the four that were required, however since the JGD cylindersare designed for a 5.5 in. (140 mm) stroke, their use on the Superiorframe required redesigning the piston and piston rod to accommodatethelongerstroke.Inaddition,specialmountingadapterswererequiredtomatch the different bolt patterns on the cylinders and the frame, asshowninFigure13-25.Theauthor’scompanyprovidedthespecialengi-neeredcomponentstoaccomplishthisadaptation.Atthistime,theunit,shown inFigure13-26,hasbeen in service forapproximately3years.

Case History 13: Extending Obsolete Compressor Life withNew OEM Cylinders A natural gas production company had an obsolete 6.0 in. (152.4mm)strokeKnightKOCA-2compressorframewithfabricated(welded)steelcylinders,asshowninFigure13-27.Thecylindersweresusceptible


Figure 13-25. Adaptation of 5.5 in. (140 mm) stroke Ariel cylinders to a Supe-rior 6.0 in. (152 m) stroke Superior compressor frame.

Figure 13-26. 2000 hp (1491 kW), 900 rpm, gas engine driven 3-stage propane refrigeration compressor.


to cracking problems in service and were no longer safely repairable.The original compressorOEMhad been out of business formore thantwodecades,soneithernewnorrepairableusedcylinderswererealisticalternatives. Lead time for new compressor packages was close to 52weeks and production losses would be substantial if the compressorcould not be returned to service in amuch shorted time. The author’scompany was asked to evaluate alternatives. Although new customengineered cylinders were an option, lead time was still somewhatlonger than desirable, considering the cost of lost production.Amuchshorter lead timewaspossiblebyadaptingnewAriel JGKcylinders tothe KOCA frame.As long as changeswere going to have to bemadeto the cylinders, bottles andpiping anyway, theproducerdecided thatit would be desirable to optimize the compression system to accom-modate an anticipated suction pressure decline over time, while stillutilizingasmuchof thegasenginedriverpoweraspractical todeliverflow.Thisultimatelyrequiredconvertingfromasingle-stagetoa2-stageconfiguration with 8.375 in. (212.7 mm) and 6.25 in. (158.7 mm) borecylinders,whichprovidedoptimalcoverageoftheoperatingconditionsandmaximumdriverpowerutilizationasshowninFigure13-28.Evenmoreimportantlyinthiscase,thesolutioncouldbeimplementedinlessthan20weeks time. New 5.5 in. (139.7 mm) strokeAriel cylinders were converted to6.0 in. (152.4mm) strokewith new custom-engineered piston and rodassemblies, rodpacking andwiper assemblies, andmounting adaptersto match the KOCA frame. NewASME coded pulsation bottles werealsodesignedandmanufactured,asshowninFigure13-29,andpipingmodifications were made to adapt to the original skid and utilize theprocessgascooler.AnAPI618DesignApproach2pulsationstudywasconducted,andthesystempipingdesignwastunedtoprovideeffectivepulsationcontrolover thebroadoperatingrange.Theunitwasbeen inservice for more than two years, and additional units are now beingconsidered for similar conversions.

Case History 14: Completely Reconfiguring an Existing Compressor Anexisting600hp(447kw),588rpm,4-throw,3-stage,7 in. (177.8mm) stroke JoyWB14 compressorwas originally configured as showninFigure13-30 tocompress1303cfm (42.5m3/min)ofbonedrynitro-gen.Requirements changed such thatonly314 cfm (10.3m3/min)wasrequired.Thecostof recycling76%of theflowwasnotacceptable,but


Figure 13-27. Obsolete KOCA-2 compressor prior to conversion.

Figure 13-28. Loading map for new 2-stage configuration.


thecostofacompletenew,butsmaller,machinewasequallydifficulttojustify.Theowner/operatorwantedtoreconfiguretheexistingcompres-soratminimalcosttooperatemoreefficientlyatthelowerflowrate.Inorder to evaluate possible alternatives, a complete performancemodelwasdeveloped for theexistingconfigurationhaving two16.0 in. (406.4mm),one10.5 in. (266.7mm)andone7.0 in. (177.8mm)cylinders.Themodelwasthenrevisedinaniterativemannertoevolveasolutionthatletreplacingthetwo1ststagecylinderswithreconditionedused8.0 in.(203.2mm) cylinders,moving the original 7.0 in. (177.8mm) 3rd stagecylinder to the 2nd stage, and installing a reconditioned used 4.0 in.(101.6mm) cylinder on the 3rd stage. The bottles weremodifiedwithnewflanges toadapt to thesmallercylinders.Mostof theexistingpip-ingwas reusedbyaddingadapter sections. Byreconfiguringthecompressorwithreconditionedusedcylindersandreapplyingoneof theexistingcylinders,as shown inFigure13-31,this solution provided a cost effective solution that saved significantoperatingcostandpreservedtheoriginal investmentof thecompressorframe,motorand thebalanceof the system.

Figure 13-29: Obsolete KOCA/2 frame with new Ariel cylinders applied.


Figure 13-30: Prior to reconfiguring, this oversized 3-stage Joy WB14 nitrogen compressor required 76% of the flow to be recycled.

Figure 13-31: The reconfigured unit was sized to meet the process requirements without significant recycled flow.


COMPRESSOROPTIMIZATIONANDAUTOMATIONWITHCUSTOMIZEDUNLOADINGDEVICES

Inadditiontohundredsofsuccessfulexamplesofcustomdesignednew cylinders and countless innovative adaptations and reapplicationsofnewandusedOEMcylinders,thereareevenmoreexamplesofcom-pressor optimization with automatically actuated devices, often calledunloaders, that control the loadandflow.

Case History 15: Automation of Propane Refrigeration Compressor Anaturalgasprocessingplantuseda400hp (298kW)Ajaxcom-pressorwith no automatic control. The 2-stage, 2-throwunit served asthe low-pressure stageofamulti-unitpropane refrigerationsystem.Asambientconditionschanged,thepropaneboil-offratevaried,necessitat-ingmanualadjustmentsofthecompressor’sclearancepockets.Sincetheunitwasnotattended24/7,theoperatorskepttheunitunder-unloadedsothatitcouldhandleupsetswithoutshuttingdown.Unexpectedshut-downsstoppedtheliquidstrippingprocessandledtoseriousproblemsdownstream from thegasplant. Thesolution involvedthecombinationofadiscretestepunloaderand an infinitely variable unloader providing, in essence, infinite vari-ability over a broad range of operating conditions, as shown in Figure13-32.Thisgavetheplantmaximumcompressorloading(andmaximumplant refrigeration and liquid stripping capacity) without the risk ofunexpected shutdowns. The compressor conversion involved installing a hydraulicallyactuated head end variable volume clearance pocket on the 1st stagecylinder along with a pneumatically actuated fixed volume clearancepocket on the head end of the 2nd stage cylinder as shown in Figure13-33. The control pressure for the hydraulically actuated variable vol-ume unloader was provided by the use of discharge pressure actingon an external accumulator containing hydraulic fluid separated by abladder. This approach eliminated the need for separate pumps andpower requirements. The control pressure for the fixed volume pocketwasprovidebyplantairpressure. Controlalgorithms, thatwerederived froma robust reciprocatingcompressorperformancemodelingprogramdevelopedby the author’scompany,wereprogrammedintoaPLC,whichthenautomaticallycon-trolled theunitwith the twounloaders.Theconversionrequiredabout


Figure 13-32. Fine unload-ing steps for a wide range of operating conditions are provided by a combination of fixed and variable un-loaders.

Figure 13-33. New automatic clearance pockets added to propane refrigeration compressor.


twodaysof installation time,and ithasbeenoperatingcontinually formore than3years.

Case History 16: Pipeline Compressor Automation Apipelineoperatorrequiredthereapplicationoffive largeClarkTCV-12 and TLA-6 19.0 in. (483mm) stroke integral engine compres-sors. An increased range of pressure ratios required a reduction ofexisting fixed clearance. The budget was tight and required reusingasmuch of the existing hardware as possible. The author’s companymodeledthecompressorperformanceanddevelopedanoptimumun-loading configuration. The front and rear heads and liner portswerereconfigured to reduce the cylinder fixed clearance. The large 14.0in. (355.6mm) diameter cylinderswere lined down to 12.5 in. (317.5mm).Thecompressorvalveswereredesignedtomaximizetheliftareaand minimize fixed clearance. A total of 116 fixed volume clearancepocket unloaders having volumes of 150 in3, 300 in3, 600 in3, or1500in3(2.5,4.9,9.8or24.6L)providedtheflexibilitynecessarytomeetthewiderangeofcompressionratiosassociatedwiththenewapplication.Novel piston designs permitted the use of existing piston rodswhilepro-viding piston to rod attachment features that resulted in easierand safer assembly techniques. The revampedunits, shown in Figure13-34,providedtherequiredperformance,whichwasmoreflexibleandreliable than theOEM’soffering.

Figure 13-33. One of five large 19.0 in. (483 mm) stroke integral engine com-pressors that were completely configured for automatic operation with a broad range of pressure ratios.


Case History 17: Replacement ofVariable Speed Engine with Synchronous Electric Drive A natural gas producer had a 2-throw, 2-stage Superior W-72reciprocating compressor driven by a 1600 hp (1193 kW), 900 rpm gasengine thathadbecomeobsolete,was inefficient andwas expensive tomaintain. In addition, due to local air quality restrictions, and electricmotor drivewas considered as a preferred replacement for the enginedrive. Theprimarymodeofflow controlwas via speed control on theengine,which provided a high level of flexibility.However the opera-tordidnotwant thecomplexityorcostofavariable frequencyelectricmotor drive. Use of a constant speed, synchronousmotor drive, then,appeared tobe impractical for the application thathad tohave awiderangeofflowcontrolcapability.Theauthor’scompanywasrequestedtoproposeunloadingequipment thatwouldprovideoptimalflexibilityatreasonable cost.Afterweighing various options, the preferred solutioninvolvedsimply replacing theouterheadsofeachcompressorcylinderand adding plug-type suction valve deactivators to the head ends ofboth cylinders.As shown in Figure 13-35, the new 1st stage headwasconfiguredwithacustomengineereddesignhaving3differentclearancepocketsof 73, 146and292 in3 (1.2, 2.4 and4.8L) respectively.

Figure 13-35. Two-throw compressor speed with multiple clearance pockets on each compressor cylin-der head end.


Figure 13-36. The multiple pocket unloaders, together with valve deactivators on the two head ends, provided more flexibility with a synchronous motor drive than the same compressor had with a 33% speed turn down.

The 2nd stage head was configured with a custom engineereddesign having 2 identical clearance pockets of 170 in3 (2.8 L). EachpocketwascontrolledwithitsownintegralpneumaticactuatorthatwascontrolledbyaPLCcontrolpanelmountedonthecompressorpackage.Togetherwith the twosetsofheadendsuctionvalvedeactivators, thisarrangement resulted in 31 discrete load steps that provided the veryfinecontrolshowninFigure13-36.Asshown,thearrangementactuallyprovidedabout36%moreturndowncapabilitythantheoriginalvariablespeeddrive.

Case History 18: Automation of Field Gas Gathering Compressors A U.S.A.Appalachian natural gas producer was operating three2-stageAjaxDPC-600compressors.Suctionpressureswings lead to theunits shutting down to prevent overloading, thus reducing each unit’srun time and requiring personnel to travel out to the remote units torestartthem.Theoperatorthereforehadbeenrunningtheunitsconser-


vativelyinordertoaccommodatethepressureswingswithoutshuttingdown,however that resulted in lessoverallproduction. An engineering study was commissioned to evaluate the com-pressor configuration to identify the best means for operating eachunitviaautomaticcontrolandunloadingtocovertheentireoperatingrange. Goals were to keep the engines running safely as field condi-tion changed, and also to increase capacity during normal operationtominimize thenumberofunits thathad tobeoperatedatany time.Various alternatives were evaluated including a single 685 in3 (11.2L) fixed volumepocket; parallel 192 and 383 in3 (3.1 and 6.3 L) fixedvolumepockets;parallel114,228,and342in3(1.9,3.7and5.6L)fixedvolumepockets,andanautomaticvariablevolumepocket.Inordertoevaluatethealternativesolutions,performancewascomparedatmorethan1000operatingpointswithintheoperatingrangenotedinFigure13-38.Thehighestflowpossibleateachpointwasplottedforeachcaseateachconditionandthecurveswereoverlayedforcomparison.Over

Figure 13-38. Two-throw compressor speed with multiple clearance pockets on each compressor cylinder head end.


the rangeof 160 to240psig (11.0 to15.5bar) suctionand900 to1000psig (62.1 to 69.0 bar) discharge, the highest average flow was pos-siblewith the automatic variable volume pocket, however, the 3-vol-ume pocket (the 2nd from the top surface in Figure 13-38) providedaverageflow thatwaswithin 1%of the variablepocket (top surface).The3-pocketheads,Figure13-39,werealsoslightly lower incostandsignificantly easier to control. This solutionwas also amorepracticalalternativeforaremotefieldapplicationwithlimitedtechnicalsupportandmaintenance available.

SUMMARY

Thelistofexamplescangoonandon,buttheforegoingcasehis-toriesprovidearangeofgoodexamples.Optimizationandrevitalizationhas been successful on compressors used in air drilling, nitrogen, airprocession, air separation, chemicalprocessing,fieldgasgathering,gasboosting,gaslift(EOR),gastransmission,gasinjectionandwithdrawal,refinery, refrigerationand specialty applications.Virtually any size andmodel can be considered for improvements, whether or not the com-

Figure 13-39. 3-volume clearance pocket unloader.


pressors are currently supported byOEMs or are obsolete. ExperienceincludesAjax,Alley,Ariel,ChicagoPneumatic,Clark,Cooper-Bessemer,Dresser-Rand, Delaval, Enterprise, Gardner Denver, GE Gemini, Hur-ricane, Joy, Knight, Neuman Esser, Norwalk, Penn Process, Superior,Thomassen, andWorthington.

Acknowledgement Theauthoracknowledgesthelegacyprovidedbymanydecadesoftechnicalleadership,innovationanddedicationofrespectedindustryex-pertsWilliamHartwick,RichardDeminski,EdwardMiller,andRichardDoup, as well as the ongoing contributions of Chad Brahler, DwayneHickman,CharlesWiseman,LarryBurnett and their capable teams.

References1. Brahler,C.D.;Hickman,D.A.&Shade,W.N.,PerformanceControlofReciprocating

Compressors -Devices forManagingLoadandFlow,GasMachineryConference,Albuquerque,NM,Oct. 7, 2008.

2. Brahler,C.D.&Schadler,H.C.,OneCompressorOEMMayNotHave theOpti-mumEquipmentSelection,GasMachineryConference,Dallas,TX,Oct. 1, 2007.

3. Harbert,P.;Hickman,D.;Mathai,G.,&Miller,R.,AnAutomationMethod forOp-timizingandControllingAReciprocatingCompressorUsingLoadStep,SpeedandSuctionPressureControl,GasMachineryConference,Dallas,TX,Oct. 2, 2007.

4. Shade,W.N.,CustomEngineeredHigh-PerformanceCylindersRevitalizeExistingCompressorAssets,GMC Today,Oct. 4, 2004.

5. Shade,W.N.,DCPMidstreamUpgradesLaGloriaGasPlant,COMPRESSORTech-Two,Oct. 2007.

6. Shade,W.N.,ModernReplacements forOldValve-In-HeadCompressorCylinders,GMC Today,Oct. 2, 2007.

7. Shade,W.N.,Re-EngineeringExtends theUtilizationofPipelineCompressorAs-sets,GMC Today,Oct. 3, 2006.

183

Chapter 14

Piston Rod Run-out is a KeyCriterion for Recip Compressors

—Gordon Ruoff

INTRODUCTION

Proper assembly, installation and maintenance are the keys tocompressor efficiency and life.And the major key is piston rod run-out.Eventhoughthecompressormayhavebeentestrunatthefactory,pistonrodrun-outmustberecheckedonsiteaspartof thecompressorinstallation.Thereafter, any time thepistonor rod is serviced, rod run-outmustbechecked.Thispaperestablishedthecriterionforrodrun-outandwas included inAPI“Standard618ReciprocatingCompressors forPetroleum,Chemical, andGas IndustryServices”,AppendixC. Thisarticle,publishedonOctober12,1981wasprovidedcourtesyofOil & Gas Journal.ThearticlewaswrittenbyGordonRuoff.————————————————————————————————

Piston rod run-out is a criterionusedby engineers andoperatingpersonneltodeterminepiston-rodalignmentinrelationtocylinderandcrossheadalignmentonhorizontal reciprocatingcompressors.Stringentrod run-out requirements are frequentlypart of thepurchase specifica-tions for reciprocating compressors. Factorywitnessofarodrun-outcheckisoftenarequirementoftheinspectionofanewcompressor.Rodrun-outisusuallyalwayscheckedasapartofnormalrecipcompressormaintenance,especiallyafterover-hauland reassemblyof thegas ends.

HORIZONALRUN-OUT

Horizontalrodrun-outreadingscanbeusedasadirectindicationofhorizontalalignment fromthecrossheadthroughthedistancepieces


to the cylinder. Horizontal rod run-out is measured by placing a dialindicatorpickupon the sideof the rod. Forperfectalignment, the indicatorshouldreadzeroastherodismoved through the length of the stroke. Variations of ±0.00015 in./in.of strokeareusually consideredacceptable. Horizontal rod run-out should be the same whether the unit iscoldorhot.Excessivehorizontalrodrun-outiscorrectedbyrealignmentof components involved. This may include cylinders, heads, distancepieces,crossheads,crossheadguides,rod,andpiston.Squarenessofrodthreadsrelativetopiston-rodnutthreadsandfaceandcrossheadthreadsand face is a critical factor.

VERTICALRUN-OUT

Normal expected cold vertical rod run-out is not an indication ofmisalignment. For perfect alignment, it is the result of the differencebetweenthenormalcoldrunningclearanceofthepistontoboreandthecrosshead-to-crossheadguide,whichinlargecylinderscausesthepiston

Figure 14-1.

PistonRodRun-out is aKeyCriterion forRecipCompressors 185

centerline to laybelow thecrossheadcenterline.Thisbasicgeometry isillustrated inFigure14-2. Notethatthepistonandcrossheadcenterlineslaybelowtheperfectalignmentcenterlinebyonehalfof therunningclearances. Incylinderswhere the running clearance is greater (or less) than the crossheadrunning clearance, the piston will lay below (or above) the crossheadcenterlinebyonehalf thedifference in the cold runningclearances. Theresultisnormalcriticalrodrun-out.Forpurposeofcalculatingnormal vertical rod run-out, this one-half clearancedifference is desig-natedas thedifferentialdrop (∆drop). FromFigure14-3, it canbe seen that thevertical rod run-outhadadirect relationshipbetween thesenormal running clearances, ∆drop,rod length, and stroke.This creates anearlyproportional right-trianglesituation such that the normal expected cold vertical run-out can becalculatedwith sufficient accuracywhen thesevaluesareknown. Vertical rod run-out is measured by placing a dial indicator onthe top or bottom of the rod.As the rod ismoved through the entirestroke length, itwill “run-out” by an amount equal to the ratio of thestroke-to-rod length times the ∆ drop, i.e., half the difference betweenthe runningclearances (Figure14-3). Itisgenerallydesirablethatthemeasuredverticalcoldrodrun-out

Figure 14-2. Basic geometry of cold vertical rod run-out(Asmeasuredbyadial indicatorduringamanualbar-over)


bewithinthecalculatedvalue(forthe involvedcylinderandcrossheadclearances, rod length, and stroke) by ±0.00015 in./in. of stroke.Anyamount of rod run-out outside of these limitsmay indicate somemis-alignment. While it may be thought by some that “zero” vertical cold rodrun-out is highly desirable and is a sign of perfection, itmay actuallybean indicationof somemisalignment,particularlyat the timeofnewunit assembly where large nonlube cylinders with aluminum pistonsand Teflon rider rings are involved. It is necessary, in fact, to considerthe run-outatbothambientandnormaloperating temperatures.

NORMALRUN-OUT

For largecylinderswithaluminumpistonsandTeflonrider rings,therecanbeasignificantdifferencebetweenthenormalcoldrodrun-outand the run-out at normal operating temperatures. This is due to thehighexpansionrateofthealuminumpistonandTeflonriderringswhichcanmake a significant difference in the differential clearance betweenthepistonand the crosshead. Ontheotherhand,theremaybeoperatingconditionswithlowsuc-tion temperatures such that normal operating temperaturesmay be no

Figure 14-3. Vertical rod run-out relationships.


greater than theambient temperatureuponwhich thenormalexpectedcoldverticalrodrun-outisbased.Thesesituationsmustbegivenspecialconsiderationwith the cold rod run-outadjustedaccordingly.

ACCEPTABLEVERTICALRUN-OUT

Whatisanacceptableverticalrodrun-outfigure?API-6182ndEdi-tionrequires0.00015in./in.ofstrokeatnormaloperatingtemperatures.Although this figure is deemed by most manufacturers as being toostringent,especiallywherelargecylindersareinvolved,itcanusuallybeaccomplishedby shimadjustmentof the crosshead shoes. Shimadjust-mentmustbemadeon thedeterminationofboth thenormal expectedcold rod run-out and the expected run-out at the normal operatingtemperatures. Athoroughstudyof thefiguresandreadings isrequiredtodeter-minewhatadjustments,ifany,areneededtoobtainthedesiredrun-outatoperatingtemperatures.Itisfeltbysomethat0.001to0.002in./in.ofstokemaybeperfectlyacceptablefornormaloperatingverticalrun-out,especially where large cylinders are involved. This amount of run-outcreates just enoughmovement in thepacking rings to keep them fromsticking in cups, especially indirtygasapplications. Where there ismuch concern about rod run-out, each applicationneeds to be studied carefully to make the right decision and properadjustment. There are many compressors operating successfully withthehigher rod run-outfigures.

MEASURINGRODRUN-OUT

With regard to vertical rod run-out, some clarification of positiveandnegativemagnitudeisneeded.Rodrun-outshouldalwaysbemea-sured startingwith the rodat theextremeendof the stroke. If the rod is all theway “back,” i.e.,with the piston at the crankend of the cylinder, and the dial indicator placed on top of the rod atthe 12 o’clockposition and set at zero as in Figure 14-3,when the rodis stroked “forward,” i.e., out toward the head end, the indicator willread positive, indicating that the rod is sloping downward from thecrosshead to the piston as shown in the illustration. (If the indicator


is placed on the bottom of the rod at the 6 o’clock position, it wouldreadnegative.)Thisisthenormalsituationencounteredwherecylinderclearanceexceeds crossheadclearance. Iftherodisallthewayout(pistontoheadend)andtheindicatoris set at zero, then it will read negative as the rod is stroked “back.”This negative reading in this direction still indicates the rod is sloping“downward,” as does the positive reading obtained when the rod isstrokedoutwardduring the reading. To avoid confusion, it is suggested that all rod run-out readingsbe taken on a forward strokewith the dial indicator set at zerowhenthe rod is all theway back. Onmultithrow units, this is important sothatrelativereadingsamongthethrowscanbeproperlycomparedandevaluated.

CORRECTINGVERTICALRUN-OUT

Verticalrodrun-outshouldneverbeadjustedorcorrectedbyforc-ing the cylinder and/or distance piece up or down by use of supportadjustingscrews.Thiscanputexcessiveforcesandstressesonthecom-ponents involved. Cylinder alignment and cylinder level at both coldandnormaloperatingtemperatureconditionsshouldbedeterminedandadjusted so that componentswill be free of harmful stresses at normaloperating temperatures. Then rod run-out should be checked and cor-rected ifnecessary. Ifalignmentisdeterminedtobebasicallycorrect,anditisdecidedthat some adjustment to the normal vertical rod run-out is absolutelynecessary, this can be done by means of shims under the crossheadshoes,e.g., takingshimsfromthebottomshoeandplacingthemunderthe top shoe drops the centerline further below the perfect alignmentcenterlinewhilemaintainingthesamerunningclearance.Thisdecreasesthe∆dropand thusdecreases thepositive rod run-out. Once crosshead shims have been shuffled to adjust rod run-out,it is important to always install the crosshead “top” side up followingremoval formaintenance. SometimesmucheffortandcostisexpendeduselesslytoobtaintherequiredAPI-618rodrun-outlimitof0.00015in./in.ofstroke,especiallywhenadjustmentsaremadetoreducethenormalexpectedrodrun-out. Takethecaseofa24-in.nonlubecylinderona9-in.strokecompres-


sor where the normal vertical cold rod run-out would be about 0.006in.with a 60-in. long rod and about 0.0045 in.with an 80-in. long rod(cylinderrunningclearanceis0.090in.andcrossheadrunningclearanceis 0.012 in.,making∆dropequal to (0.090-0.012)/2or0.039 in. If rod run-out is adjusted to zero when cold, it will be about anegative 0.003 in. when hot, i.e., it will be laying upward, instead ofdownwardas shown inFigure14-3. However, a 24-in. nonlube piston has a rider ring design withabout 0.060 in. of allowablewear built into it. Over time, as the riderringwears, the pistonwill drop below the centerline of the crossheadby the amount of the allowablewear. Thus,when it is time to replacethe rider rings, rod run-outwill be about 0.009 in.with the 60-in.-longrod. If periodic crosshead shimadjustments have beenmade in an at-tempt tokeeprodrun-outatornearzero, thenall theshims thatwerechangedmust beput back in their originalpositions.Todo this is un-necessaryandcostly inrelationtoanybenefitsgained.Further,verticalrodrun-outthatissetpreciselyatthefactorywithmuchtimeandeffortusuallyhas tobe readjusted in thefieldafter shipmentandhandling.

EVALUATINGVERTICALRUN-OUT

Inevaluatingverticalrodrun-out,normalrodsagmayoccasionallyhavetobetakenintoaccount.Considera9-in.strokecylinderendwitha1-3/4 in.diameterrodwitha lengthof70 in.betweencrossheadandpiston(cylinderendusesadoubledistancepiecearrangementbetweencrossheadguideandcylinder, resulting ina long rod). Ifpistonclearanceis0.040in.andcrossheadclearanceis0.020in.,the normal expected vertical rod run-out (from the formula of Figure14-3) is (9/70)(1/2) (0.040-0.020)or about0.001 in.±0.0014 in. Considering normal sag, the maximum deflection is calculatedfrom the formula (5/384)(Wl3El) for a beam supported at both ends.Assuming the rod to be steel with aweight of 8.17 lb/ft, E (modulusof elasticity)or 30x106, and1 (momentof inertia)of 0.40604; then:

normal sag= (5/384)x [(47.7x703)/(30x106x0.4606)]=0.0115 in.

However,theprojectedstraightlinedeflectionisabout1.5timesthe


calculated sag, or about 0.023 in. This is the figure thatmust be usedto calculate the expected rod run-out relative to sag conditions (Figure14-4).Theeffective rod length isnowonly1/2 the full lengthor35 in.Since short rods result in greater run-out, the effective run-out due tosag is “amplified.” Withthissagcondition,andwiththepistonandcrossheadinper-fect alignment, i.e., with no ∆ drop, when an indicator is placed nextto the crosshead, or next to the packing area (closest to the piston aspossible)expectedrodrun-outduetosagwillbe9/35x0.023,orabout0.006 in.This,ofcourse, isaconditionthatcannotbereadilycorrected. Todetermineifrodsagiscontributingtowhatappearstobeexces-sive rod run-out, an indicator placed on the rod next to the crossheadwill result in a positive reading as the rod is stroked outward towardtheheadend,while another indicatornext to thepressurepackingwillresult inanegative readingduring the samestroking. Anindicatorplacedontherodataboutitsmidpointwillresult inlittleorno run-out reading.

Figure 14-4. Rod run-out relative to sag.


Inreality,whenthecrossheadandpistonaresubjecttorodsagtheytendtolayinthecrossheadguideandcylinderasshownexaggeratedinFigure5.Duringtheoutwardstrokewhentherodisincompression,thecolumneffectcreatesevenfurther“sag.”Thenwhenthestrokereversesandtherodisintensionittendstostraightenout.Thissometimescausestherodtoappeartojumpupanddownwhenviewedinoperationwiththenakedeye. Instruments used to detect roddynamics during actual operationsometimes record the reversalat theendofeachstrokeasa significant“jump.” Ifadial indicator isplacedat the lowestpointof rodsagdur-ingabar-overtest, thedifferenceindeflectionfromrodcompressiontorod tension can often be seen, resulting from the frictional drag of thepiston in thebore, especially if it is a largenonlube cylinder. Insummary,horizontalrodrun-outshouldideallybezero,withavariationofnomorethan±0.00015in./in.ofstroke.Excessivehorizontalrod run-out canbean indicationofmisalignment. Normalvertical rod run-out isnot an indicationofmisalignment.Ratheritistheresultofthedifferencebetweenthepistonandcrossheadrunningclearances,withitsmagnitudefurtheraffectedbythelengthofthepistonrod,andbythestroke.Itneedstobecalculatedforeachcyl-inderend.Normalvariationfromthecalculatedvalueshouldbewithin±0.00015 in./in.of stroke. Someaxioms relative tovertical rod run-outareas follows:

• The larger the cylinder and the more the running clearance, themore the run-out.

Figure 14-5. Crosshead and piston subjected to rod sag.


• The longer the stroke, themore the run-out.

• The longer the rod, the less the run-out.

• Short rod, long stroke=greater run-out.

• Long rod, short stroke= lesser run-out.

• The larger the cylinder, the longer the stroke, the shorter the rod=greater run-out.

• Thesmallerthecylinder, theshorterthestroke, thelongertherod= lesser runout.

There are many compressors successfully running with rod run-outs thatexceedanyof the limitsandparametersestablishedabove. Insomerarecasesrodsagmaycontributetoamisleadingexcessiverun-outreadingwhich cannot be improved by further alignment or crossheadadjustment Rod run-out that is set precisely to stringent requirements at thefactorywith theexpenditureofmuch timeandeffortusuallyhas tobereadjusted in the field after shipment, handling, and installation if thesameprecise run-out is required.

193

Chapter 15

Effect of Pulsation Bottle DesignOn the Performance of aModern Low-speed Gas

Transmission Compressor Piston—Christine M. GehriRalph E. Harris, Ph.D.

Southwest Research Institute

INTRODUCTION

Pulsationbottledesignis typicallythoughtofasameanstomini-mize harmonic responses in the suction and discharge piping.As thischaptershows,thereisalotmoretopulsationbottledesign.Withproperdesignnotonlywillharmonicsbe controlledbut compressorefficiencywillbe improved. This studywas presented at the 2003GasMachinery ConferencebyChristineM.Gehri andRalphE.Harris,Ph.D.————————————————————————————————

Dynamic pressure drop refers to the calculation of instantaneouspressuredropateachpointthroughthecrankrotation.Thehorsepowercost iscomputedby integratingthe instantaneouspressuredropoveracylinder cycle. If the instantaneousvelocityacrossa square lawrestric-tion is much higher than the mean velocity, then the cycle integratedhorsepowercostwillbemuchhigherthanthehorsepowercostestimatedby mean flow estimates. Field analyses have documented instanceswheredynamicpressuredropeffectshave incurredhundredsofexcesshorsepowerand, insomecases,weresohighas to inhibit full loadop-


erations. In one such case, strong cylinder interactions combinedwitha significant pressure drop restriction downstream of the compressorcylindervalves resulted inexcessivehorsepower losses. This paper presents a case study of a low-speed reciprocatingcompressor unit installed in a natural gas transmission service. On-site testing following start-up indicated total losses of approximately55percentoftheindicatedhorsepowerwhileoperatingonthebottomloadstepatratedspeed.Interimmodificationstotheexistingprimarysuctionanddischargepulsationbottlesweremade,whichallowedtheunitstooperatemuchmoreefficientlythroughthewinter.Redesignoftheprimarypulsationbottleswasnecessarytofurtherminimizecylin-derinteractionandtheassociatedhorsepowercost.Fulldocumentationofthedesignprocessispresentedinthispaper,alongwithsupportingfielddata.

INTRODUCTION

Bottle design is a key component in a successful compressor in-stallation. Traditionally, bottle design focused on controlling pulsationlevels,andultimatelyvibration, throughresonanceavoidance.Thema-jorityofthelowspeedintegralunitsinstalledintheUSgastransmissionindustryweredesignedusing analog technology. The analog approachfocusedonlocatingacousticresonancesoutsideofthenormaloperatingspeed range and limiting bottle unbalanced shaking forces to a targetvalue. Empirical loading values, and not stress calculations, were keydesign parameters. The success of this approach is evident in the con-tinuedoperationand long lifeof thiscompressorfleet.With traditionaldesign tools, theeffectof thebottledesignoncompressorperformanceis difficult to establish. Because of the linear nature of the solution,the role of valve performance and pulsation levels on the predictedpressure-volume cards was not accurately represented. Moreover, theeffectsofpressuredrop throughthebottlesystemwerenotreflected inthe predicted horsepower, or any equivalent accounting in the design.Following installation, abottledesignwas typicallyviewedas success-ful if it achieved acceptable vibration and compressor performance, asmeasuredby cylinderperformance testing. It is very difficult to measure the impact of the pulsation bottledesign on cylinder performance in the field.Although pulsation levels

EffectofPulsationBottleDesign 195

are easilymeasured, differential pressuremeasurements between vari-ous locations in the bottle design or between the cylinder nozzle andheaderarenotonlyunreliable, thevaluesaredifficult to interpretwithregard to horsepower costs. Pressure drop through the cylinder valvesandorificepressurelossmayshowupasdifferentialindicatedpressureonthepressure-volumecard,butthroughthebottlesystem,theselossesaretypicallyseenasaspreadinthecylindertoepressures.Asurveyofcompressorefficiencies1 in the installedbaseof integralunits in theUSgastransmissionindustryrevealedasignificantspread,asshowninFig-ure1.This spread inefficiencyrepresentsa lossofpipelinecapacityaswellasexcessfuelconsumption.Givenourimprovingabilitytopredictand understand the role of bottle design on compressor performance,a significant component to the observed spreadmust be attributed topulsationbottledesigns. The increasinguse of high-speed compression in various applica-tions has prompted the development of design tools that better meetthe demanding requirements of today’s high-speed compressor pack-ages.Pulsationbottledesign,inparticular,hastakenonnewchallenges.Resonanceavoidanceonvariablehigh-speedunitsismuchmoredifficultto achieve, and resonance management is required. High-speed unitshaveamoredifficulttimeachievingtraditionalcompressorperformanceefficiencies,andpulsationcontrol throughpressuredropcandriveunitefficienciesdown further. Dynamicpressuredropreferstoanincreaseinstaticpressuredropthroughacomponentoranentiresystemofpulsationcontrolelements,resulting frompulsatingflow.Becausepressuredrop isproportional tothe squareof thevelocity throughan element,dynamicflow results inmore pressure drop on the swings above themean flow than is savedon the portion of flow below the mean flow value. When acousticresonances are involved, the dynamic flow can be significant and pro-duce surprisingly large horsepower losses. Since the compressor is theprimemover of the gas, thesedynamic pressure losses are reflected inthe horsepower consumed by the compressor cylinder and seen in thepressure-volume cards.With the ongoing transition to digital acousticanalysis, calculation of dynamic pressure drop can be automated intothedesignprocessof futuresystems.However, the largebodyofexist-ingcompressorsmaybenefit fromreevaluationof thepulsationcontrolsystem,particularlywhenelectricdrivesareinvolvedorcapacitylimita-tionshavebeenmet.

196

Com


Figure15-1.EfficiencySurvey in the InstalledBaseof IntegralUnits in theUSGasTransmission Industry.


CASESTUDY1

Figure 2 is a photograph of a Cooper GMW330 installed in 1996attheEnterpriseCompressorStationoperatedbySouthernNaturalGasCompany.Thisisoneofthelastlow-speedintegralsinstalledintheUSgastransmissionindustry.Theunitisratedat3,920HPandhastwo21-inchborecylinders.Thecylindersaredownconnectedwiththeprimarysuction and discharge bottles located below the compressor deck. Theunitoperatesoveraspeedrangeof255to330rpm,withfixedclearancepocketsandatotalof19loadsteps.Theoriginalprimarybottledesign,shown in Figure 3, used a baffled, center fed bottle with a perforatedchoketoreducedsecondorderpulsations.This isoneofthreeidenticalunits installedaspartof anearlyhorsepower replacementproject.Fol-lowing installation, the unit could not run off of the bottom load stepwithoutexceedingtheratedenginetorque.Figure4presentsthecylinderpressure data from Cylinder 2 at 330 rpm.As shown, compressor ef-ficiency(basedonidealcompressorhorsepowerbetweentoepressures)

Figure15-2.CooperGMW330with21-inchBoreCylinders.


Figure15-3.CooperGMW330OriginalBottleDesign–1996.

Figure15-4.CylinderPerformancefor theOriginalBottleDesignat330 rpm.


wasontheorderof46percent.Notethatthedifferentialindicatedpower(DIP)aboveandbelowtoepressuresisconsistentwithverylargevalvelosses, or perhaps nozzle orifice losses. Simultaneousmeasurements ofpressurebehindthesuctionanddischargevalves,overlaidwithcylinderpressure data clearly indicated that high valve loss was not the rootcauseof theexcessive losses. Using the measured dynamic pressures, combined with theacousticmodeldeveloped for the system, an estimate of the static anddynamic losses through the choke tubewasmade. Figure 5 illustratesthenatureof thesecalculations.Theflowthroughtheperforatedchoketube is first established. From this flow estimate, the pressure drop isestimatedusingthestaticlosscoefficient.Usingthemeanflowandstaticloss coefficient, the estimated pressure drop is 2.8 psi. Using the sameloss coefficient and the dynamic flow, the average estimated pressuredropis17psi.Usingmeanflowandmeanpressuredropestimates,thecostoftheperforatedchokeisestimatedtobe30HP.Usingthedynamicflow and dynamic pressure drop estimates, the cost of the perforatedchokeisestimatedat600HP.Fieldmodificationsofboththesuctionanddischargebottlesconsistedofshorteningthechoketubesandremovingthe perforated section, as shown in Figure 6. Shortening the choke re-sultedinsignificantlyraisingthechoke’sresonantfrequency,anacousticmode involving strong cylinder-to-cylinder interactions. Figure 7 presents the cylinder pressure data followingmodifica-tions to the suction and discharge primary bottles. Based on the im-provedcylinderefficiency,440oftheestimated600HPwererecovered.Thisparticular station typicallyoperatesata lowratio,whichmakes itdifficult to achieve low pulsation levels and consistently high cylinderefficiencies.Basedondataacquiredatvariousspeedsandoperatingcon-ditions, a simple relationship between pulsation driven cylinder lossesandcombinedsuctionanddischargenozzlepulsationswasdeveloped.AsshowninFigure8,combinedsuctionanddischargenozzlepulsationsof less than 40 psi peak-to-peak would be required to drive losses toless than50HPper cylinder end. The units operatedwith this bottle configuration until 2002,wheninterest was expressed to recovermore of the lost horsepower throughthe pulsation control system.A bottle systemwas designed that wouldcompletely isolate cylinder-to-cylinder interactions.A goal of the designwas touse theexisting secondarybottle locatedoutsideof the compres-sorbuilding.Figure9presents a schematicof thenewdesignapproach.

200

Com

pressorHandbook:PrinciplesandPracticeFigure15-5.SourceofExcessiveHorsepowerLoss for theOriginalBottle.


Figure15-6.CooperGMW330ModifiedBottleDesign–1997.

Figure15-7.CylinderPerformance for theModifiedBottleat 330 rpm.


Bothcylindersareacousticallyisolated from each other byusingalongchoketoconnectthe primary bottle chamberforeachcylindertotheexter-nal bottle. Figures 10 and 11present pictures of the dualchoke configuration leavingthe primary bottle and thesmall bottle used to recom-binetheflowpriortoenteringthe exterior secondary bottle.Figure 12 presents cylinderpressure traces followingmodifications to the bottles.Table 1 summarizes the cur-rent performance of the unit.Compressor performance isnowupto92percentatcom-parableoperatingconditions.

Figure 15-9. Current Bottle Layout for theCooperGMW330Unit.

Figure 15-8. Effect of Residual Pulsations on Cylinder Performance for theCooperGMW330Unit.


Figure15-10.DualChokeConfiguration.

Figure15-11.Spe-cialFitting forDualChokes.

A final concern expressed by the unit operatorswas an apparentvariationindischargepressurebetweenthethreeunits.Thecenterunit(Unit12)displayedhigherdischargepressurethanUnits11and13.Thepressuresensorsforallthreeunitswerelocatedinthecenterofthechoketubesbetweentheprimaryandsecondarybottles.Again,dynamicflowlosseswere thought to be driving the variation in discharge pressures.Dynamic pressure datawere acquired in the center of the suction anddischarge choke tube as a function of unit speed. Figure 13 presentsthe results and clearly shows the choke tube resonance excited at 300rpm. Figure 14 presents the static discharge pressure recorded at thesame location.Clearly, the dynamic response is driving the increase inapparent staticpressure.


Figure15-12.CylinderPerformancefor theCurrentBottleLayoutat330 rpm.

Table15-1.PerformanceSummary forCurrentGMW330BottleLayout.


Figure15-13.Unit 11Choke4thOrderDynamicPressureVersusRPM.

Figure15-14.Unit 11DischargePressureVersusRPM.


CASESTUDY2

Figure 15 is a photograph of a Dresser-Rand HHE reciprocatingcompressor unit operating in a storage application. There are threeidentical units at this station, installed in the late 1970’s. The units areseparable electric driven compressors operating at 327 rpm and ratedat 6,000HP. The units can be operated in either a single or two-stageinjection/withdrawalservicewithdischargepressuresuptoabout3,000psig or a single stage withdrawal service with discharge pressures ofabout900psig.Theoperatingcompanyplanstoinstallafourthidenticalunitandrequestedareevaluationof theexistingbottledesigns.SwRI’sdigital Interactive Pulsation-Performance Simulation2,3

(IPPS) codewas

usedtoevaluatethecurrentbottledesign.TheIPPScodeautomatesthecalculation of dynamic flow losses at pressure drop elements and theanalysis indicated excessive losses on both the first and second stages(Figure16).Theprimarydischargebottlesarethree-chamberwithchokesin series, as shown in Figure 17. Based on the model predictions, theexcess losses appear to be associated with dynamic flow through theexit chokebetweenCylinders2and4.Thischoke tubecarries theflowof twocylinders, and is sized such that thevelocity isgreater than the

Figure15-15.Dresser-RandHHELow-SpeedSeparableUnit.


Figure15-16.IPPSPressureDropPredictionsforSecond-stageDischargeChoke.

flowthroughthechokeconnectingtheCylinder4andCylinder6bottlechambers. Aspartof anongoingGMRCresearcheffort to evaluateanden-hance the SwRI/GMRC design tools, a field test effortwas completedtodocumentsystemperformance.Cylinderperformancedatawereusedtoquantifyunit losses.Figure18presents theheadendandcrankendtraces forCylinder 2.Table 2 summarizes theunit’s cylinder efficiency.Notethat,aspredicted,thereisexcesslossofhorsepowerforCylinders2and4.Predictedlossesareapproximately160HPonthehigh-pressuresideand60HPonthelow-pressureside.Combined,thetotallossesareapproximately4percentofratedhorsepowerattheconditionsanalyzed.

Figure15-17.ExistingBottleConfigurations forHHEUnits.


We anticipate that the losses will be greater under different operatingconditions. Modifications were developed that should minimize thehorsepowercost.Atthistime,wearerecommendingthattheexitchoketubesbeshortened toshift theacoustic response frequency.Forelectricdriveunits,thepotentialsavingsinhorsepowerismagnifiedbydemandcharge costs,which can significantly alter the economicsofunit opera-tion.

SUMMARY

There is awide range of compressor efficiencies in the installedbase of horsepower in the US gas transmission industry. Traditionalbottledesignmethods typicallydidnot quantify the impact ofpulsa-tioncontroloncompressorperformance.Thischapterhasshownhowdynamic flow losses can significantly reduce compressor efficiencies.Basedonourcurrentunderstandingoftherelationshipbetweenpulsa-

Figure15-18.Cylinder2Performance forExistingBottleDesign.


Table15-2.CylinderEfficiency forExistingHHEBottleDesign.

tion control and compressor performance and our continually devel-opingpredictive capabilities,we suspect that there aremany existingcompressor installations thatwouldbenefit fromareevaluationof thebottle design. In some cases, even marginal increases in compressorperformance can translate directly to significant savings in operatingcosts due to decreases in fuel consumption and increases in systemcapacity.

References1. Berry, Renee, “Gas Transmission Compressor Efficiency Survey.” GMRC Research

ProjectConductedbySouthwestResearch Institute, 1995.2. Gehri,ChristineandHarris,Ralph,Ph.D.,“HighSpeedReciprocatingCompressors–

TheImportanceofInteractiveModeling.”PresentedattheGasMachineryConference,Houston,TX,October1999.

3, Gehri,ChristineandHarris,Ralph,Ph.D.,“TechnologyfortheDesignandEvaluationof High-Speed Reciprocating Compressor Installations.” Presented at the EuropeanForum forReciprocatingCompressors,Dresden,Germany,November1999.

Chapter 16

Resolution of a CompressorValve Failure: A Case Study

—Steve ChaykoskyDresser-Rand Company

Painted Post, NY

Thisstudyaddressesthevariouscomponentsofreciprocatingcom-pressorvalves (i.e.,valveplate type,valveplate lift,valveplate impactvelocities)theeffectsofparticulates inthegasandhoweachinfluencescompressorperformanceandoveralloperation. This study has been provided courtesy ofDresser Rand, Inc., theowner of this material. Furthermore, Dresser Rand, Inc., has grantedcopyrightpermission for theuseof this study in thisbook.————————————————————————————————

ABSTRACT

The commissioning date of three critical path ethylene compres-sors at a Belgian chemical plant was in jeopardy because the originalcompressor valves failedwithin hours of start-up on the final stage ofcompression.Thevalvesupplierattemptedtoresolvethefailurestwice,without success.The failed concentric ringvalvedesignwas evaluatedby the compressor OEM and determined to have marginal reliabilitycharacteristics.A different type of compressor valve, the ported platevalve,wasproposedasapotential solution.Althoughaslightdecreasein compressor efficiencywas expectedwith the alternative valve tech-nology, the client deemed this an acceptable compromise to meet hisproduction schedule. The new valveswere installed inApril 1999 andcontinue to runwithoutproblems.

211


INTRODUCTION

Valveproblemsarereportedastheprimaryreasonforreciprocatingcompressorshutdowns.Theanecdotalevidenceisconfirmedbyarecentsurveyofhydrogencompressors,whichfinds that36%ofunscheduleddowntime isdue tovalve failures. Compressorvalvesfailforavarietyofreasons,whichcangenerallybe divided into design factors and operational factors. Design factorsincludeapplicationand selectionerrors.Operational factors are relatedto the process gas conditions and to how the compressor is run andmaintained. Discovering root causes of failure depends on a full exchange ofinformation between the valvemanufacturer and his client. The clientmust be willing to provide complete operational data and the valvemanufacturermust bewilling to review andmodify his valve design.Sinceeachreciprocatingcompressor isunique in theway it isoperatedand serviced, the circumstances surrounding each case of valve failureand its resolutionarealsounique. This case study illustrates how a valve failure can be attributedboth to valve design considerations and to operational problems.Asthe case is discussed, emphasiswill beplacedon the analysis of valvedynamics, the relationshipofvalve lift tovalve lifeandcompressoref-ficiency, andelementsofvalve failureanalysis. The client reported that failures of the 3rd stagedischarge valvesoccurredwithin hours of achieving full pressure load on 2 of 3 newlycommissionedethylenecompressors.Thefailedvalvecomponentsweremultipleconcentric ringsmadeofPEEK,ahigh-strength thermoplastic.

Figure 16-1. Concentric Ring Valve Figure 16-2. Ported Plate Valve

Resolutionof aCompressorValveFailure 213

Figure 16-1 shows an example of a concentric ring valve. The originalvalve supplier reported that excessive stresses were imparted in theringsdue tounexpected radial,flutterymovement.Their responsewasto design andmanufacture a different type of concentric ring elementtobetter resist the lateralmotion. As the compressorOEMand a valvemanufacturer,Dresser-Randwas solicited for comments but did not have enough information toperformaformalfailureanalysis.However,plantstartupwasimminent,sotheclientcontractedustoprovidevalvehardwareincasetheoriginalsupplier’s modified valves were unsuccessful.We had the freedom toprovide whatever valve style we believed would work best. By day’send,aportedplatevalvewasselected,adynamicvalveanalysis(DVA)wasperformedbasedonthereportedfieldconditions,newvalvedraw-ingsandprograms for theN/C (numerically controlled)machine toolswerecreated,andthemanufacturingfloorwasalertedtoexpectacrisisorder.Figure16-2 showsaportedplatevalve. Twodays later, the client contactedus in themorning to say thatthe original 3rd stage suction ring valves had also failed and wantedtoknowifportedplatesuctionvalvescouldbedeliveredwiththenewdischarge valves. Suction and discharge valves for all 3 compressorswere manufactured and shipped that afternoon.A set of ported platevalveswas installed inone compressor immediatelyupon receipt. Tendaysaftertheproblemarose,theoriginalvalvesuppliertrans-mitted simulatedP-V (pressure-volume) indicator cards andvalvemo-tiondiagramsofthefailedvalvesforourevaluation.Thedatasuggestedthattheoriginalvalveswereclosingontime.However,theoriginalvalvesupplier’smodifiedvalves continued to fail. By the secondweek, a second set of portedplate valveswerefit-ted to another compressor, and a site visit by Dresser-Rand personnelyielded thefirst reports of oil sticktionproblemsandofparticulates inthegas stream. Since the3rd stagedischargevalves failedfirst, the client initiallybelieved debris and/or liquids could be eliminated as the problemsourcebecause the3rdstagesuctionvalves,whichareupstreamof thedischargevalves,remainedintact.However,the3rdstagesuctionvalvesfailedthedayafterthedischargevalvesfailed,andtheclienteventuallyacknowledged that therewasdebris in thegas stream. Theoriginalvalvesupplierexplainedthattheradialmovementoftheconcentricringswascausedbyvalveflutter,whichoccurredbecause


the compressor did not achieve full pressure fast enough. Flutter is aphenomenon inwhich themoving elements of a valve open and closeseveraltimesthroughoutonecompressioncycle.Tominimizeflutter,anacceleratedloadprofilewasattempted,whichbroughtthecompressortofullpressure faster.However, the3rdstagedischargevalverings failedagainafter just a fewhoursat fullpressure. When asked to comment on the cause of failure, we could onlyconjecture because failed hardware was not available to evaluate, adynamicsstudyof theoriginalvalvedesignwasnot initiallyprovided,andnotall of theoperational factswereknown. ProprietaryDVAsoftwareisthetoolthatwasusedtogeneratesim-ulatedliftdiagramsoftheconcentricringandportedplatevalves.Sincethefailedconcentricringvalvedatawasnot immediatelyavailable,weestimated the configuration of rings and springs in order to perform aDVA study. Figure 16-3 shows the estimatedhead enddischarge valvemotion of the concentric rings. Note that all of the rings close beforereaching the topdeadcenter (TDC) crankangleof 360°.This is impor-tantbecauseitmeansthattheoriginalringsshouldhaveclosedontime.The original valve supplier’s simulated lift diagramswere transmittedtouslater,andtheirvalvemotionstudyalsoindicatedon-timeclosing.SincewedevelopedtheportedplatevalvedatafortheDVA,therewasahighdegreeofconfidence in thevalvemotionsimulation,andFigure16-4 shows that theportedplatesdo indeedclosewellbeforeTDC. Liftdiagramsillustratetheimportanceofpropervalveclosingbutdo not indicate whether the impact velocities of the moving elements(plates, rings) are acceptable. DVA software calculates impact velocity,but thisparameter isusefulonly if the application limitsof eachvalvetypearewellknown.Differentplateandringgeometrieshavedifferent

Figure 16-3. Lift Diagram Concentric Ring Valve Head End Discharge Valve


allowable impactvelocities,whichare experientiallyderived. Ported plate valves were selected for this high-pressure ethyleneapplication with the assumption that oil sticktion and excessive plateimpactsfactoredintothefailures.Confirmedreportsofoilsticktionanddebris in the gas streamwere received almost 2weeks after the initialfailuresoccurred. Theemergentnatureofthisfieldproblemforcedarapidevaluationof potential valve designs. The final selection was certainly based onpartsavailabilityandshopcapacity,but theforemostcustomerrequire-mentwasalsoachieved: reliability. The ported plate valve has less flow area than the original failedring valve because the lift is lower.Although the client preferred nottohave compressor efficiencydegradedby avalve change,hedeemedit an acceptable compromise.After all, a compressor that shuts downon valve failures has zero efficiency. The effective flow area (EFA) ofthe ported plate valve is 41% less than the reported EFA of the failedconcentric ring valve.A valve’s EFA is determined by laboratory flowtestingand is typically40%to70%of thevalve liftarea.Valve liftareais a calculated quantity basedmerely on valve geometry and is opentointerpretationbydifferentvalvemanufacturers.Therefore,EFAistherelevantflowarea for thevalvedesigner,not lift area. The original valve supplier’s 2nd modification of the concentricring valves involved lowering the lifts. The ported plate inlet valveshave17%lessflowareathantheringvalveswithreducedlift;theportedplatedischargevalves29% less.Although theEFAsof theportedplatevalvesarelessthanthemodifiedringvalves’,anOctober1999monitor-ing report basedonRecip-Trapmeasurements states that the 3rd stagevalve losses were “very low,” as were the plate impacts. Simulations

Figure 16-4. Lift Diagram Ported Plate Valve Head End Discharge Valve


predicteda4.5% increase in3rdstagepowerconsumptionwithportedplatevalvesinstalled,butthemeasuredlosseswereactuallymuchless.Thispointsoutacommonfallacyincalculatingacylinder’spowerloss-es.Algorithms that only consider the valve losses and ignore pressurelosses over the cylinder passages, cylinder ports, and valve cages arenotcomprehensive.All these lossesareadditive.Therefore,areductionjustinvalveflowareadoesnotnecessarilytranslateintoaproportionalpower increaseandconsequent efficiencydecrease. Avalvewithverylowpressurelossappearsveryefficientonpaper.Evenwhenproperspringsarechosentoclose thevalve, theremaynotbeenoughpressuredroptoholdthemovingelementscompletelyopen,whichwillcauseflutter.Flutterdecreasestheeffectiveliftandthereforeincreases the predicted power losses. Flutter can be eliminated withlightersprings,but then there isa riskof latevalveclosing,whichwillreducecylindercapacityandcausethemovingelementstoslamshut.Ifthegas forces slamavalve closed, themovingelements could fracturefrom theassociatedhigh closing impactvelocities.

FAILUREANALYSIS

Design factors working against reliability of the original valvedesign:

1. High valve lift. Lift is simply the distance traveled by the valve’smoving elements from the fully closed position to the fully openposition.Thehigher the lift, thehigher theflowareaof thevalve.The higher the valve’s flow area, the lower the pressure dropacross the valve. The lower the pressure drop across the valve,the lower the horsepower consumed by the compressor’s driver(motor,engine,orturbine.)Thelowerthepowerconsumption,themoreefficient thecompressor,and thehappier theclient.Bysuchtransitivelogic,theclientshouldbehappierwithhighervalvelift,but not in this case. The lift of the original failed concentric ringvalvewas.079inches,whereastheliftoftheportedplatevalvethathasbeen runningproblem-free forover ayear is .040 inches.Thelift chosen by the valve designer depends on the valve type andthe compressor operating conditions. Different valve styles havedifferent flow characteristics. Ported plate valves have excellent


flowareasatrelativelylowlifts,buttherearequicklydiminishingreturnsas lift is incrementally increased,and there isacertain liftbeyondwhichnomoreflowareacanbeobtained.Concentricringvalves are often used at higher lifts than ported plate valves.Attheupperendoftheirusefulliftrange,concentricringvalveshaveequivalentorslightlyhigherflowareasthanportedplates,buttheimpactvelocitiesoftheringsatthosehigherliftswillexceedthoseof theportedplates. Poppet valves have yetmoreflowarea thanconcentric ring valves when used at their highest practical lifts,but the poppet’s impact velocities may be markedly higher thanthe rings’.Therefore, choosing the appropriate lift for agivenap-plicationdepends on knowing the impact velocities that differentvalvetypescanwithstand.Inthiscase,a.040inchliftportedplatevalve works well at the 3rd stage conditions, yet a .079 inch liftconcentric ringvalveprovedunreliableandnotmeasurablymoreefficient.

2. Excessive impact velocities. The laws of physics rarely allow some-thing for nothing, and the price to pay for higher lift is higherimpact velocities of the valve’smoving elements. Since increaseddurability of themoving elements depends on relatively low im-pact velocities and therefore low lifts, the relationship betweencompressor efficiency and valve reliability is inverse. Reliabilityand efficiency are competing properties of compressor valves,and an acceptable compromise often means sacrificing efficiencyfor reliability. In this case, the original valve supplier did not di-vulge the predicted impact velocities of the concentric rings thatfailed.However,ourestimationof thevalvemotion indicates thatthe opening impacts of the original .079 inch lift concentric ringdischarge valves may have been 40% higher than the openingimpacts of the .040 inch lift ported plate valves, and the closingimpactsmayhavebeen38%higher.Lift isnottheonlyfactorthatdetermines impact velocity. Compressor speed, cylinder pressure,valve springing, and sticktionare also contributors. Impactveloc-ityincreaseswithhigherspeeds,highercylinderpressures,heavierspring rates, and increased sticktion. In this case, the compressorspeedof 328 rpm is considered low,but the cylinderpressuresof1000 psi inlet and 3200 psi discharge are relatively high, and theincreased3rdstageoilviscosityandlubricationratecontributedto


a relativelyhigh sticktion level.

3. Lateral ring motion at off-design condition (reduced pressure break-in phase.) Although a compressor valve is often able to workover a range of operating conditions, it is optimized for a singledesign case, i.e. a single combination of pressures, temperatures,speed, cylinder clearance, gas molecular weight, lubrication rate,and pulsation level.When one of these parameters changes, thevalvenecessarily runsat anoff-designcondition, andproper tim-ing of the opening and closing events is disrupted.Conditions ofservice thatareslightlyoff-designareusually tolerated,butmajorchanges to any of the parameters can pose a severe detriment tothe valvemotion and sometimes result in failures of springs andmoving elements. In this case, the original valve supplier blamedthe fluttery, non-synchronous movement of the rings on a slowrate of pressurization of the 3rd stage. The supplier stated thatthe ringswerevery sensitive to sticktion effects and special start-up conditions and subsequently redesigned the valve a 3rd timewith radiused rings of a different thermoplastic material to fightthe sticktion problem and the high impact loading.However, theredesigned concentric ring valves were not installed, so it is notknownwhether thiswouldhave solved the failures.

Asthemovingelementsofavalveopenandclose,theynevermoveperfectlyperpendiculartobothstoppingsurfaces.Theplatesandringsalwayswobblebeforetheymomentarilyrestagainsttheseatsandstopplates.Consequently,theinitialopeningandclosingimpactsalwaysoccuronanouteredgeoftheplateorring,whichis usuallywhere plate or ring fractures initiate. The ported platethatsolvedthisfieldproblememploysscallopededgesontheouterperiphery.Sincetheplatetipsasitmovesofftheseatandstopplate,the initial impacts will be forced to one of the scalloped edges.Figure16-2 illustrates thescallopedplate,which isapatentedde-sign.Each scallopprovidesa regionofhigher cross-sectional areato better absorb the opening and closing impact velocities. Sincethis increasedarea canbetter resist impact stresses, the likelihoodof plate failures is reduced. The ported plate is also precision-guided with a center ring and with rollpins to minimize lateralmovement.Inthiscase,theplatevalvedesign,whichconsistsofa


singleportedPEEKplate,resistsnon-uniform,lateralmotionbetterthan the concentric rings.

Client requested the original valve type that failed. Manyclientsareaccustomed to using a particular style of valve in their compres-sors, and they often develop a comfort level with a given valvetype. Preferences are usually experience-based, but price has alsobeenknowntoenterintothespecificationprocess.Whilecustomerinputiswelcomedandencouraged,thevalvesuppliermustknowhis product’s application limits, which are primarily based onpressure, temperature, and compressor speed. Valve designs arecustomized to satisfy specific operating conditions, and not allvalvetypeswillworkforallconditions,soitisthevalvedesigner’sresponsibility to select and advocate the appropriate hardware.While preferences should be honored whenever possible, clientsareadvisednot todictatefinalvalvedesign.

Operational factorsworkingagainstvalve reliability:

1. Particulate ingestion. The suction strainers on all 3 stages wereregularly plugged with small metallic particles. The 2nd & 3rdstage accumulations were associated with cylinder lube oil hold-ing the particles together. The source of the particulates, whichwere reported to be a reality of the plant’s gas process for years,wasnotdivulged.Incompressiblesubstancesinthegasstreamcanbecome lodgedbetweenamovingelement (plateor ring)and thestationary parts (seat and stopplate) of a compressor valve. Theplatemaybecomeoverstressedatthepointwhereithingesontheforeign particles and fail. Plastic plates, such as PEEK or Nylon,willoftenembedmetallicdebrisandcontinuetooperateproperly,buttheembeddedmetalfragmentscanprematurelyweartheseal-ing surfaceof thevalve seat.

2. Oil sticktion.The3rdstagecylinderlubricationsystemisindependentof thefirst two stages,which is atypical for reciprocating compres-sors.The3rdstageusesaslightlydifferentblendofoilandfeedsitatahigher rate than the lowerstages.The3rdstagepressuresandtemperatures are 1000 to 3200 psi and 95° to 225° F, respectively.Ethylene evidently takesoil into solutionat thispressure and tem-perature combination, so an adhesive agentwas added to the 3rd


stage lubricating oil to ensure that it would stick to the cylinderbore,pistonrings,andpistonrodpacking.Theadditivealsocausedtheoiltosticktothemovingpartsofthevalve.Sticktionisdefinedas the viscous adhesion of oil between themoving and stationarypartsofavalve,andisdetrimentaltovalvereliabilityintwoways.First,sticktioninhibitsplateopening,causingtheplatetoslamopenwhenthepressureforceexceedsthesticktionforce,whichcancausedamagenotonlytotheplatebuttotheunderlyingsprings.Second,sticktiondelaysplateclosing,whichelevates theplate’sclosing im-pactvelocity.Iftheplateclosessolatethatthereisbackflowthroughthevalve,thenthereisnotonlyagreaterpotentialforplatedamagebutthereisalsoanactuallossofcylindercapacity.Inthiscasestudy,the high oil viscosity and increased lubrication rate contributed tothesticktioneffects.

CONCLUSIONS

1. Thereasonsforvalvefailurescanbedividedintodesignproblemsandoperationalproblems.

2. Valvelift,plateimpactvelocity,operatingatoff-designconditions,valveflutter,oilsticktion,andliquidsanddebrisinthegasstreamare factors to considerwhenevaluatingvalve failures.

3. There are several different types of compressor valves, each ofwhich has application limits, so a good understanding of how toreliablyapplyeach type isparamount.

4. Successful resolution of valve failures oftenmeans having to em-ploy new or redesigned valves that are less efficient. Sacrificingsome compressor efficiency for increasedvalvedurability is oftenanacceptable compromise.

5. Timelyresolutionofvalvefailuresdependsonaccuratelyanalyzingcompleteoperationaldataandanyavailable failedhardware.

Acknowledgements TheauthorwishestothankDr.DerekWoollattforhisthoughtfulcommentsandsuggestionsonthetechnicalcontent.ThanksarealsoduetoAnneLaceyforherinsightfulreview.Finally,thankstoDresser-RandCompany for theopportunity topresent this case study.

221

Appendix A1

Compressor Manufacturers

The following tables list compressor equipment manufacturersand compressors by model. The list is divided into two categories:dynamic compressors and positive displacement compressors. Theinformation provided is instrumental in identifying units with com-parable capability. Note thatmost positive displacement compressors are packagedby independent companies. These companies package a compressorwith a driver per the purchaser specifications (included in the pack-ager’s scope of supply are suction & discharge vessels, valves andcontrols). This information, provided courtesy of COMPRESSORTechTwomagazine “Compressor Technology Sourcing Supplement” publishedby Diesel & Gas Turbine Publications. Current compressor informa-tion, aswell asNaturalGasEngines,MechanicalDriveGasTurbines,Electric Motors, Variable Speed Drives, andMechanical Drive SteamTurbinescanbefoundatwww.compressortech2orwww.CTSSNet.net.—––————————Compressor specifications in Appendix XX From the 2009 Compressor Technology Sourcing Supplement, courtesy COMPRESSORTechTwo magazine, published by Diesel & Gas Turbine Publications. Current compressor information can be found at www.compressortech2 or www.CTSSNet.net.

222

Com


Dynam

ic Com

pressor Specifications

DataC

ourtesyCOMPRESSO

RTechTwom

agazine.Additionalcurrentdataavailableatw

ww.com

pressortech2.comor

www.CTSSN

et.net

Appendices

223

Dynam

ic Com


DataC

ourtesyCOMPRESSO

RTechTwom


ww.com

pressortech2.comor

www.CTSSN

et.net

224

Com


Dynam

ic Com


DataC

ourtesyCOMPRESSO

RTechTwom


ww.com

pres-sortech2.com

orwww.CTSSN

et.net

Appendices

225

Positive Displacem

ent Com


DataC

ourtesyCOMPRESSO

RTechTwom


ww.com

pressortech2.com

orwww.CTSSN

et.net

226

Com


Positive Displacem

ent Com


DataC

ourtesyCOMPRESSO

RTechTwom


ww.com

pressortech2.com

orwww.CTSSN

et.net

Appendices

227

Positive Displacem

ent Com


DataC

ourtesyCOMPRESSO

RTechTwom


ww.com

pressortech2.com

orwww.CTSSN

et.net

228

Com


Positive Displacem

ent Com


DataC

ourtesyCOMPRESSO

RTechTwom


ww.com

pressortech2.com

orwww.CTSSN

et.net

Appendices

229

Positive Displacem

ent Com


DataC

ourtesyCOMPRESSO

RTechTwom


ww.com

pressortech2.com

orwww.CTSSN

et.net

230

Com


Positive Displacem

ent Com


DataC

ourtesyCOMPRESSO

RTechTwom


ww.com

pressortech2.com

orwww.CTSSN

et.net

Appendices

231

Positive Displacem

ent Com


DataC

ourtesyCOMPRESSO

RTechTwom


ww.com

pressortech2.com

orwww.CTSSN

et.net

232

Com


Positive Displacem

ent Com

pressor SpecificationsData Courtesy COMPRESSORTechTwomagazine.Additional current data available at www.compressortech2.com orwww.CTSSNet.net

Appendices 233

Appendix A2

COMPARISONTABLEOFTHETHREETYPESOFCOMPRESSORS*

This chart provides a brief comparison of three compressortypes: Reciprocating, Screw and Centrifugal relative to pressure andflow rate. For specific details permanufacturer andmodels the readerisdirected toAppendixA.

*ProcessGasApplicationsWhereAPI619ScrewCompressorsReplacedReciprocatingAndCentrifugalCompressors,TakaoOhama,YoshinoriKurioka,HironaoTanaka,TakaoKoga;Proceedingsof theThirty-FifthTurbomachinerySymposium—2006


Appendix B1

List of Symbols

A Area,Ft2

Btu British thermalunit C Velocityoftheairenteringthecompressororairandcombus-

tionproducts leaving the turbine, ft per sec CDP Compressordischargepressure,psia CDT Compressordischarge temperature, °R cp Specificheat at constantpressure,Btu/lb°F cp Specificheat at constantpressure,Btu/lb°F cv Specificheat at constantvolume,Btu/lb°F °E Engler,degrees gc Accelerationdue togravity, ft/sec2

h1 Enthalpyof thefluidentering,Btu/lbm h2 Enthalpyof thefluid leaving,Btu/lbm hi Enthalpyof thefluidentering,Btu/lb ho Enthalpyof thefluid leaving,Btu/lbm HP Horsepower Hr Hours J Ratioofworkunit toheatunit, 778.2 ft lbf /Btu k Ratioof specificheats, cp/cv KE1 kinetic energyof thefluidentering,Btu/lbm KE2 Kinetic energyof thefluid leaving,Btu/lbm KJ Kilojoules KW Kilowatts Mo Massof agivenvolumeofoil Mw Massof agivenvolumeofwater MW Poweroutput (electricgenerator)

Appendices 235

MWMolecularweight N1 Singlespoolor lowpressurecompressor-turbinerotorspeed,

rpm N2 Dualspoolhighpressurecompressor-turbinerotorspeed,rpm N3 Power turbine rotor speed, rpm NOx Oxidesofnitrogen P2 Compressor inletpressure,psia P3 Singlespoolcompressordischargepressure,singlespool low

pressure compressordischargepressure,psia P4 Dual spool lowpressure compressordischargepressure,psia P5 Turbine inletpressure,psia P3/P2 Single spool compressorpressure ratio,dual spool lowpres-

sure compressorpressure ratio P4/P2 Compressorpressure ratio P4/P3 Dual spoolhighpressure compressorpressure ratio PAMB Atmospheric totalpressure,psia PE1 Potential energyof thefluidentering,Btu/lbm PE2 Potential energyof thefluid leaving,Btu/lbm Pi Inletpressure,psia Po Dischargepressure,psia cP AbsoluteorDynamicViscosity,Centipoises PPT Power input to the loaddevice,Btu/sec Ps3 Lowpressure compressoroutpressure (static) Ps4 Highpressure compressoroutpressure (static) Pt7 Exhaust totalpressure, psia Q Cubic feetper second (CFS) 1Q2 Heat transferred toor from the system,Btu/lbm Qm Mechanicalbearing lossesof thedrivenequipment,Btu/sec Qm Mechanicalbearing lossesof thepower turbine,Btu/sec Qr Radiationandconvectionheat loss ,Btu/sec Rc Compressorpressure ratiosecsR.I. RedwoodNo1 (UK), seconds


cST Kinematicviscosity=Absoluteviscosity/density. centistokes SUS SayboltUniversal (USA), seconds T1DB Drybulb temperatureupstreamof the cooler, °R

T2DB Drybulb temperaturedownstreamof the cooler, °R T2WB Wetbulb temperaturedownstreamof the cooler, °R T2 Compressor inlet temperature, °R T3 Single spool compressor discharge temperature, dual spool

lowpressure compressordischarge temperature, °R T4 Dual spool low pressure compressor discharge temperature,

°R Tamb Ambientair temperature, °R Ti Inlet temperature, °R To Discharge temperature, °R U Velocity, feet per sec D Diameter, in2, ft2

V Vibration W Massflowrate, lbm/sec W2 Relativevelocity Wa Airflowentering the compressor, lb/sec Wf Fuelflow, lbs/sec Wf/cdP Gas turbineburner ratiounits, lbs/hr-psia Wg Turbine inletgasflow, lbs/sec 1W2 Workperunitmassonorby the system, ft-lbf /lbm 1W2/J Work,Btu/lbm Zave Average compressibility factorof air, β2 Relativedirectionof air leaving the statorvanes ηc Compressor efficiency ηt Gasgenerator efficiency ηpt Powerextraction turbineefficiency θ1 Directionof air leaving the statorvanes

Appendices 237

Appendix B2

Glossary of Terms

Absolute pressure—The totalpressuremeasured fromabsolutezero inunits such as pounds per square in absolute (psia), Inches ofMercuryAbsolute (in.ofHGA).

Absolute temperature—The temperature of a body at absolute zero(Fahrenheit scale is minus 459.67°F/Celsius scale is minus273.15°C).

Absolute viscosity (Dynamic)—Is the tangential force per unit areanecessary tomove one layerwith respect to the other at unitvelocity whenmaintained a unit distance apart by the fluid.Units are pound force sec per foot squared (lbf sec/ft2) or ki-logram/meter sec (kg/msec)

Absolute zero—The lower limitof temperature -459.69°For -273.15°C

Acidity—Thequality, stateordegreeofbeingacid.Asubstance is con-sidered acidic if its pH is less than 7. In oils, acidity denotesthe presence of acid-type constituentswhose concentration isusuallydefined in termsof aneutralizationnumber.

ACFM—Actual cubic feet per minute. The quantity of gas actuallycompressed and delivered at the temperature & pressure atthat location.

Activated alumina—Anadsorption typedesiccant.

Adiabatic compression—Compression where no heat is transferred toor from thegasduring the compressionprocess.

Adiabatic efficiency—The ratio of the actualwork output to theworkoutputthatwouldbeachievediftheprocessbetweentheinletstateand theexit statewas isentropic.

Adsorptive filtration—Theattractiontoafiltermediumbyelectrostaticforces, or by molecular attraction between the particles andthemedium.


Aeration—Aprocessbywhichair is circulated throughordissolved ina liquidor substance.

Aerosol—Asuspensionoffine solidor liquidparticles inagas.

Aftercooler—Heat exchangers for cooling air or gas discharged, typi-cally, fromcompressors.

Air actuator—A device to convert pneumatic energy to mechanicalenergy.

Air bearing—Afluiddynamic(hydrostatic)bearingusingairtosupportthe rotating shaft loadswithin thebearing.

Air borne—Supportedor transportedbyair.

Air cooled compressor—A compressor cooled by atmospheric air cir-culated around the cylinders or casing usually incorporatingcoolingfins.

Air flow—Themovementof air fromonepoint toanother.

Air lock—An interlocking device that permits passage of air (or gas)between regions of different pressureswholemaintaining thepressure in each region.

Air nozzle—Afitting incorporating an orificewhereby a liquid or gaspassing through the orifice undergoes a change in pressureandvelocity.

Air receiver—Areceptaclethatstorescompressedairforheavydemandsinexcessof compressor capacity.

ALT—Altitude

Altitude—Theelevationabove sea level.

Alternating current (AC)—An electrical current that periodically re-verses its direction. Standard in US and Canada is 60 cyclespersecond.Europeandothercountriesis50cyclespersecond.

Amagat’s law—States that the volume of a mixture of gases is equalto thesumof thepartialvolumeswhich theconstituentgaseswouldoccupyifeachexistedaloneatthetotalpressureofthemixture.

Ambient—Thetemperature,pressureandhumidityatspecificsiteloca-tions.

Appendices 239

Ammonia—A colorless gas with a strong pungent odor made up ofnitrogenandhydrogen (NH3).

Amonton’s law—States that the pressure of a gas, at constant volume,variesdirectlywith theabsolute temperature.

Ampere (AMP)—Aunitofelectricalcurrentorrateofflowofelectronsthrough a conductor. One volt across one ohm of resistancecausesa currentflowofoneampere.

Ancillary equipment—Components in addition to the primary equip-ment (suchas the compressor,generatorordriver).

Anhydrous—Asubstance containingnowater.

Anion—Anionwithmoreelectrons thanprotons,giving itanetnega-tive charge..

A.P.I.—AmericanPetroleum Institute

Approach temperature—The difference in temperature between thatof the heatedmedium exiting from a heat exchanger and theinlet temperature of the cooling medium entering the heatexchanger.

ASHRAE—TheAmerican Society of Heating, Refrigerating andAir-ConditioningEngineers, Inc. .

A.S.M.E.—AmericanSocietyofMechanicalEngineers

A.S.T.M.—AmericanSocietyForTestingandMaterials.

Atmosphere (ATM)—The layer of gas surrounding the earth that isretainedbygravity.

Atmospheres absolute (ATA)—It is the weight of the column of airexisting above the earths surface at sea level. It is equivalentto14.696psiaor1.0333kg/sq cm.

Atmospheric dew point—Thetemperatureatwhichwatervaporbeginsto condenseat atmosphericpressure.

Attenuation—Thegradual loss in intensityof soundwavesdue todis-tanceoranblockingmedia (suchasawall).

Avogadro’s Law—States thatequalvolumesof idealgases,at thesametemperature and pressure, contain the same number of mol-ecules.


Axial compressor—A compressor belonging to the group of dynamiccompressors. Characterized by having its flow in the axialdirection.

Boyle’s Law—States that for a fixed amount of an ideal gas kept at afixed temperature,pressure andvolumeare inverselypropor-tional.

Bag house—Adust-collectionchambercontainingnumerouspermeablefabricfilters throughwhich thegasespass.

Bar—Aunitofpressure equal to0.986925atmospheresor14.5039psi.

Barometric pressure—Is the absolute atmospheric pressure existing atany given point in the atmosphere. It is theweight of a unitcolumnofgasdirectlyabovethepointofmeasurement.Itvar-ieswithaltitude,moistureandweather conditions.

Base plate—Ametallic structure onwhich a compressor,motor, driverorothermachineunits aremounted.

Bernoulli’s principle—States that forafluid,with littleornoviscosity,increases in the speedof thefluidflowoccurs simultaneouslywithadecreaseinpressureoradecreaseinthefluidspotentialenergy

BHP—Breakhorsepower

Blower—Acompressordesigned tooperateat lowerpressure ratios.

Blow off control—When themaximumpressure is reached, the deliv-eredair isblownoff to theatmosphere.

Body—Thestationaryvalve seating surface

Bonnet—Theportionof avalve that surrounds the spring.

Booster compressor—Compressorsusedtoincreasetheinitialgaspres-sure, usually low or near atmospheric pressure, to a higherpressure.

Boyle’s law—States that the volume of a gas, at constant temperature,varies inverselywith thepressure.

Breather—Afilteringunitforventedenclosuresinstalledtopreventdirtand foreignmatter fromentering theenclosure.

British thermal unit—Btu.Theamountofheatrequiredtoraisethetem-

Appendices 241

perature of one pound ofwater one degree Fahrenheit underset conditionsof temperatureandpressure.

Buna N—Asyntheticrubberfrequentlyusedforvesselandliquidfilterelementgasket.

Burst pressure—Maximum pressure a vessel will withstand withoutbursting.

By pass—Conditionthatexistswhentheair,gas,orfluidnormallypass-ing throughanelement isbeing shuntedaround theelement.

Capacity—Capacityofacompressor is the full ratedvolumeofflowofgas compressedanddeliveredat certain set conditions.

Capacity gauge—A gauge that measures air flow as a percentage ofcapacity, used in rotary screw compressors as an estimatorduringmodulation controls.

Carbon dioxide—Aheavycolorlessgas thatdoesnot support combus-tion but is formed by the combustion and decomposition oforganic substances.

Carbon monoxide—Acolorlessodorlessverypoisonousgasformedbythe incompleteburningof carbon.

Casing—The pressure containing stationary element that encloses therotorandassociated internal componentsof the compressor.

Cation—An ionwithmoreprotons thanelectrons.

Celsius—The international temperature scalewherewater freezes at 0(degrees) andboils at 100 (degrees).Alsoknownas the centi-grade scale.

Centrifugal compressor—A dynamic compressor where air or gas iscompressedbythemechanicalactionofrotatingimpellersandstationarydiaphragms imparting velocity andpressure to theairorgas.

CFM—Cubic feetperminute.Anairflowmeasurementofvolume.

Charles Law—States that the volume of a gas is directly proportionalto its temperature.

Chatter—Abnormal, rapid reciprocating movement of the disc on theseatof apressure reliefvalve.


Check valve—Avalve thatpermitsflow inonedirectiononly.

Choke—This term is used for turbo compressors and represents themaximum flow condition. It is sometimes also referred to asstonewalling.

Clean pressure drop—Thepressure lossacross thefilterelementdeter-mined under steady state flow conditions using a clean testfluidacrossa cleanfilter element.

Clearance—Themaximum cylinder volume on a working side of thepiston, minus the piston displacement volume per stroke. Itisusually expressedasapercentageof thedisplacedvolume.

Clearance pocket—Anadjustablevolumethatmaybeopenedorclosedto the clearance space for increasing or decreasing the clear-ance,usuallytemporarilyorseasonally,toreducethevolumet-ric efficiencyandflowof the compressor.

Collapse pressure—Theminimumdifferentialpressurethatanelementisdesigned towithstandwithoutpermanentdeformation.

CNG—Compressednaturalgas.

Coalescing filter—Afilterunitthatcombinesthreeprinciplestofilteroutoil aerosols: 1)Direct interception—A sieving action, 2) Inertial impaction—Collisionwith filtermedia fibers, 3)Diffusion -Par-ticles travel in a spiralmotion, presenting an effective frontalarea thus capturingparticleswithin thefiltermedium.

Cogeneration—The use of exhaust heat flow as the source of heat togenerate steam, hot water, hot oil, etc. to be used in anotherprocess.

Cold start—Startingacompressor fromastateof total shutdown.Usu-allydonewith“local” control at the compressor.

Compressed air—Air under pressure greater than that of the atmo-sphere.

Compressibility—Ameasureof the relativevolumechangeofagasasa response toapressure change.

Compressibility factor Z—Istheratiooftheactualvolumeofthegastothevolumedeterminedaccording to the idealgas law.

Compression efficiency—Istheratioofthetheoreticalworkrequirement

Appendices 243

to the actual work required to be performed on the gas forcompressionanddelivery.

Compression ratio—Theratioof theabsolutedischargepressure to theabsolute inletpressure.

Condensate—The liquid phase that separates from a vapor duringcondensation.

Condenser—Aheat exchangerused to changeavapor intoa liquid.

Conduction—Thetransferofthermalenergybetweenneighboringmol-ecules ina substancedue toa temperaturegradient

Constant speed control—A controlmethodwhereby load is varied ei-therbyuseof recyclevalves, suctionvalveunloadersorvari-ablevolumepocketsandenergyinputatconstantdriverspeed.

Contaminant—Foreignmatter carried in the air, gas or fluid to be fil-tered out. Includes air borne dirt,metallic particles producedbywearofmovingpartsoftheaircompressor,rustfrommetalpipelines.

Contaminant capacity—An indicatorof relative service lifeof afilter.

Control valve—Anyvalve that controlsflow.

Convection—Ameans of transferring heat through mass flow.Alsothe transfer of heatwithin a fluid bymovements of particleswithin thefluid.

Coolant—Fluid coolingagent.

Cooling tower—Acoolingwatersupplysystem.Therearetwodifferenttypes—Openandclosed loop systems.

Coriolis force—A force acting in adirectionperpendicular to the rota-tionaxisand to thevelocityof thebody in the rotating frameand isproportional to theobjects speed in the rotating frame.

CPM—Cycles per minute—a unit of measure of the frequency of anyvibration.

Cracking—To subject petroleum oil to heat in order to break it downinto lighterproducts.

Critical pressure—The pressure required at the critical temperature tocause thegas to change state. It is thehighestvaporpressure


that the liquid canexert.

Critical speed—Rotative speeds at which rotating machinery passesthrough its thenatural frequency.

Critical temperature—That temperature above which a gas will notliquefy regardlessof any increase inpressure.

Crosshead assembly—Theassemblyconnectingthecrankcaseandcon-nectingrodtothecylinderheadandpistonrodfortranslatingcircular to linearmotion.

Crosshead loading

Thetensileorcompressiveloadingonthecrossheadassemblywithcom-pressivepistonrod loadingon theoutwardstrokeand tensilepiston rod loadingon the inward stroke.

CSA—CanadianStandardsAssociation

Cubic feet per minute (CFM)—Anairflowmeasurementofvolume.

Cycle time—Amountof time fora compressor to completeone cycle.

Cylinder—Thepiston chamber ina compressororactuator.

Cyclone—Atypeofseparatorforremovaloflargerparticlesfromagasstream. Gas ladenwith particulates enters the cyclone and isdirected to flow in a spiral causing the entrained particulatestofalloutandcollectatthebottom.Thegasexitsnearthetopof the cyclone.

Cyclone separator—Ameansofpurifyingangasstreambyusingbothgravitational andcentrifugal forces.

Dalton’s law—Statesthatthetotalpressureexertedbyagaseousmixtureisequal to thesumof thepartialpressuresofeach individualcomponent in thegasmixture.

Degrees Celsius (°C)—Anabsolutetemperaturescale.((°F–32)x5/9).

Degrees Fahrenheit (°F)—An absolute temperature scale. ((°C x 9/5)+32).

Degrees Kelvin (°K)—An absolute temperature scale. The kelvin unitof thermodynamic temperature, is the fraction1/273,16of thethermodynamic temperature of the triple point of water. Thetriple point of water is the equilibrium temperature (0.01 °C

Appendices 245

or273,16K)betweenpureice,airfreewaterandwatervapor.

Degree Rankine (°R)—Anabsolute temperature scale. (°F+459,67).

Degree of saturation—Theratioofhumidityratioofmoistairtohumid-ity ratioof saturatedmoist air.

Deliquescent—Meltingordissolvingandbecomingaliquidbyabsorb-ingmoisture.

Delta P—Describes thepressuredrop through a component and is thedifference inpressurebetween twopoints.

Delta T—Describes the temperaturedifferencebetween twopoints.

Demand—Flow of a gas under specific conditions required at a par-ticularpoint.

Demulsibility—The resistanceof ahydraulicfluid to emulsification.

Density—Theweightofagivenvolumeofgas,usuallyexpressedinlb/cu ft at standardpressureand temperature conditions.

Desiccant—An adsorption type material used in compressed air dry-ers. Industry standards are activated alumina, silica gel andmolecular sieves.

Design pressure—Themaximumcontinuousoperatingpressure asde-signedby themanufacturer.

Desorption—Oppositeof absorptionoradsorption.

Dew point—Ofagas is the temperatureatwhich thevapor ina space(at a given pressure)will start to condense (form dew). Dewpointofagasmixtureisthetemperatureatwhichthehighestboilingpoint constituentwill start to condense.

Diaphragm—A stationary element between rotating stages of amulti-stage centrifugal compressor.

Diaphragm compressor—Isapositivedisplacement reciprocating com-pressor using a flexible membrane or diaphragm in place ofapiston.

Diaphragm cooling—Amethod of removing heat from the flowingmedium by circulation of a coolant in passages built into thediaphragm.Alsoknownasdiaphragmrouting.

Differential pressure—The difference in pressure between any two


pointsof a systemor component.

Differential pressure indicator—An indicator, or differential pressuregage,whichsignalsthedifferenceinpressurebetweenanytwopointsof a systemora component.

Diffuser—Astationarypassagesurroundinganimpeller,inwhichveloc-ity pressure imparted to the flowmedium by the impeller isconverted into staticpressure.

Digital controller—Control device whose operation may be reducedto binary operation such as ON = 0, OFF = 1 or OPEN andCLOSED.AlsoknownasLogic controls

Diluent—An inert substance added so as to reduce the activity of asubstance.

Direct current—DC.Acontinuous,onedirectionalflowof electricity.

Directional control valve—A valve to control the flow of a gas in acertaindirection.

Dirt holding capacity—ThequantityofcontaminantafilterelementcantrapandholdbeforethemaximumallowablebackpressureordeltaP level is reached.

Disc—Themovable seating surface inavalve.

Discharge piping—The piping between the compressor and the after-cooler, theaftercooler separatorand thedownstreamelement.

Discharge pressure—The totalgaspressure (staticplusvelocity) at thedischarge port of the compressor. Velocity pressure is consid-eredonlywithdynamic compressors.

Discharge temperature—Thetemperatureexistingatthedischargeportof the compressor.

Displacement compressor—Amachine where a static pressure rise isobtainedbyallowingsuccessivevolumesofgastobeaspiratedintoandexhaustedoutofaclosedspacebymeansof thedis-placementof amovingmember.

Displacement of a compressor—Thevolumedisplacedbythecompress-ingelementof thefirst stageperunitof time.

Disposable filter—A filter element intended to be discarded and re-placedafterone service cycle.

Appendices 247

DOE—TheU.S.DepartmentofEnergy.

Double acting compressor—Apositivedisplacementcompressorwherethepistonscompressgasonboththehead-endandcrank-endstrokes.

Downstream—Theportionoftheflowstreamwhichhasalreadypassedthrough the systemor the portion of the system located afterafilteror separator/filter.

Drag—Occurs when a valve does not close completely after poppingandremainspartlyopenuntil thepressure is furtherreduced.

Drain valve—Avalvedesignedtoremovesurplusliquidfromthecom-pressedgas system (scrubbers,pulsationbottles, etc.).

Dripleg—Apipeextendingdownwardfromthebottomofthevesseltocollect any condensationflow.

Dry bulb temperature—Theambientgastemperatureasindicatedbyastandard thermometer.

Dry gas—Anygasorgasmixture thatcontainsnowatervaporand/orinwhichalloftheconstituentsaresubstantiallyabovetheirre-spectivesaturatedvaporpressuresattheexistingtemperature.

Dry unit (oil free)—Isone inwhich there isno liquid injectionand/orliquid circulation for evaporative coolingor sealing.

Dynamic losses—Frictionagainstductwalls, internal friction in thegamassanddirectionvariationswillcauseaspeedreductionandare therefore calleddynamic losses.

Dynamic type compressors—Machines in which air or gas is com-pressedbythemechanicalactionofrotatingvanesorimpellersimpartingvelocityandpressure to theflowingmedium.

Dynamic viscosity(Dynamic)—Thetangentialforceperunitareaneces-sarytomoveonelayerwithrespecttotheotheratunitvelocitywhenmaintainedaunitdistanceapartby thefluid.Units arepound force sec per foot squared (lbf sec/ft2) or kilogram/meter sec (kg/msec)

Durometer—This termrefers to thehardnessor softnessofgaskets.

Dust cake—Alayerofdustbuiltuponanairfilter.


Dust holding capacity—Theamountofatmosphericdustwhichafilterwill capture.

Effective area—The area (in sq inches) of the filter element that is ex-posed to theflowofairorfluid for effectivefiltering.

Efficiency—filter—Ability of a filter to removeparticlematter from anair stream. Measured by comparing concentrate of materialupstreamanddownstreamofthefilter.Typicalparticulatesizesrange from .3micron to50micron.

Efficiency- compression—Is the ratio of the theoretical work require-ment to the actualwork required to beperformedon the gasfor compressionanddelivery.

Efficiency- isothermal—The ratio of the theoreticalwork calculatedonan isothermal basis to the actual work transferred to the gasduring compression.

Efficiency mechanical—The ratio of the thermodynamicwork require-ment in the cylinder toactualbrakehorsepower requirement.

Efficiency—polytropic—Theratioofthepolytropiccompressionenergytransferred to the gas to the actual energy transferred to thegas.

Efficiency—volumetric—Theratioofactualcapacitytopistondisplace-ment, statedasapercentage.

Element—Themediumormaterialthatdoestheactualfilteringorsepa-rating.May be paper, wire mesh, special cellulose, inorganicplastic, or a combination.

Emulsibility—Theabilityofanon-water-solublefluidtoformanemul-sionwithwater.

Emulsifier—Additive that promotes the formation of a stablemixture,oremulsion,ofoilandwater.Commonemulsifiersare:metal-lic soaps, certainanimalandvegetableoils,andvariouspolarcompounds.

Emulsion—Intimate mixture of oil and water, generally of a milky orcloudy appearance. Emulsions may be of two types: oil-inwater (wherewater is the continuous phase) andwater-in-oil(wherewater is thediscontinuousphase).

Appendices 249

Energy audit—Asurveythatshowshowmuchenergyyouuseinyourcompressedairgeneration.

Energy conservation—Practices andmeasures that increase energy ef-ficiency.

Energy kinetic—Theenergya substancepossessesbyvirtueof itsmo-tionorvelocity.

Energy storage—Theabilitytoconvertenergyintootherforms,suchasheat or chemical reaction, so that it can be retrieved for lateruse.Alsothedevelopment,design,constructionandoperationofdevicesforstoringenergyuntilneeded.Technologyincludesdevices suchas compressedgas.

Enthalpy—Thesumof the internal andexternal energies.

Entropy—Ameasureof theunavailabilityof energy ina substance.

Environmental contaminant—All material and energy present in andaroundanoperatingsystem,suchasdust,airmoisture,chemi-cals, and thermal energy.

Evaporation—The escape of water molecules from a liquid to the gasphaseat the surfaceof abodyofwater.

Exothermic—Atermusedtodescribeachemicalprocess inwhichheatis released.

Expanders—Turbinesorenginesinwhichgasexpands,doeswork,andundergoesadrop in temperature.

Ferrography—An analytical method of assessing machine health byquantifying and examining ferrous wear particles suspendedin the lubricantorhydraulicfluid.

Filter—Adevicethatremovessolidcontaminants,suchasdirtormetalparticles,fromaliquidorgas,orthatseparatesoneliquidfromanother,or a liquid fromagas.

Filter breather—Afiltering unit for vented enclosures installed to pre-ventdirtandforeignmatter fromenteringtheenclosure.Alsopreventsoil lossby retainingoildropletsanddraining theoilback to the sump.

Filter efficiency—Ability of a filter to remove particle matter from anair stream. Measured by comparing concentrate of material


upstreamanddownstreamofthefilter.Typicalparticulatesizesrange from .3micron to50micron.

Filter element—The porous device which perform the actual processoffiltration.

Filter housing—Something that coversorprotects thefilter assembly.

Filter separator—Filteringunitthatseparatessolidsandliquiddropletsfromgas (air).Widelyused in removingoil fromagasorair.

Filtration—The physical ormechanical process of separating insolubleparticulatematter fromafluid, suchasairor liquid,bypass-ing the fluid through a filtermedium thatwill not allow theparticulate topass through it.

First law of thermodynamics—The amount of work done on or by asystemisequaltotheamountofenergytransferredtoorfromthe system.

Flange—Abolted rimused forattachment toanotherobject.

Flash point—Thelowesttemperaturetowhichoilmustbeheatedunderstandardizedtestconditionstodriveoffsufficientinflammablevapor to flashwhen brought into contactwith a flame. Flashpoints of petroleum based lubricants increasewith increasingpressure.

Flexible mounting—Vibration isolation mount. Provides reductions invibration transmission.

Flow—The volume of a substance passing a point per unit time (e.g.,metersper second,gallonsperhour, etc.).

Flow control valve—A valve that controls the flow of air that passesthrough thevalve.

Flow diagram—A schematic flow sheet showing all controls involvedwith the system.

Flow meter—Aninstrumentformeasuringtheamountofgasflowofacompressor.Measured inCFM.

Flow rate—The rate (in liters or gallons per minute, cubic meters orcubic feetper second,orotherquantityper timeunit)

Flushing—Acirculationprocessdesigned to removecontamination.

Appendices 251

Forced draft fan—A fan that generates (by pushing) a flow of ambi-ent air over the exterior of the finned pipes to dissipate thesensibleheat.

Friction—Surfaceresistancetorelativemotion,whichslowsdownmove-mentandcausesheat.

Full load—Achievedwhen the air compressor is running at full RPMwith a fully opened inlet and discharge, delivering themaxi-mumvolumeat the ratedpressure.

Gay-Lussac’s Law—States that if the mass and pressure of a gas areheld constant than gas volume is directly proportional to thegas’s temperature.

Gag—A device attached to a safety relief valve that prevents it fromopeningat the setpressure.

Galling—A form of wear in which seizing or tearing of intermittentcontacting surfacesoccurs.

Gas—Oneof threebasicphasesofmatter (solid, liquid,gas).

Gas bypass—All or a portion of gas that is redirected upstream ordownstream of a component to compensate for process flowdemands.

Gas drying—The drying of compressed gases other than air. Equip-mentsize, choiceofmaterialsandotherspecificationsmaybedecidedlydifferent fordryinggases other than air because ofthe specificpropertiesof thegases.Properties include specificgravity,specificheat,viscosity, thermalconductivity,explosivecharacteristics, toxicity, corrosionandothers.

Gas laws—Thebehaviorofperfectgases,ormixturesthereof,followsasetof laws.Boyle’ law,Charle’s law,Amonton’s law,Dalton’slaw,Amagat’s law,Avogadro’s law,Poisson’s law.

Gate valve—A type of valve in which the closing element (the gate)is a disc thatmoves across the stream in a groove or slot forsupportagainstpressure.Agatevalvehasrelatively largefullports anda straight lineflowpattern.

Gage—An instrument formeasuring, testing,or registering.


Gage pressure—Is pressure as determined by most instruments andgages.

Glycol dehydration—Amethodofdryingnaturalgas.

Governor—A engine or motor controller that regulates the fuel flowor amperage to the engine ormotor to control the speed andpower.

GPM—Gallonsperminute

Guide—Theportionof avalveused toguide thedisc.

Guide vane—A stationary element that may be adjustable and whichdirects theflowmedium to the inletof an impeller.

Head adiabatic—Theforce(orpressure)thataadiabatic(Noheattrans-ferredtoorfromthefluid)compressionprocessraisesthegasfromsuctionleveltodischargelevelmeasuredininchesoffeetofwater column.

Header—Adistributionpipe.

Head polytropic—Theforce(orpressure)thatapolytropic(heattransferto/fromthefluidcanexistsandthefluidmaynotbehaveasanidealgas)compressionprocessraisesthegasfromsuctionleveltodischargelevelmeasuredininchesoffeetofwatercolumn.

Head pressure—The force (or pressure) that the compression processraises the gas from suction level to discharge levelmeasuredin inches of feet of water column. This term is also referredtoasHead.

Heat capacity—TheheatenergytransferredfromsystemAtosystemB.

Heat exchanger—Adeviceinwhichtwofluids,separatedbyaheatcon-ductingbarrier,transferheatenergyfromonefluidtotheother.

Heat recovery—Recoveringandutilizing theheat contentof afluid.

Horsepower (HP)—A unit of work equal to 33,000 foot pounds perminute, 550 footpoundsper second,or746Watts.

Horsepower brake (BHP)—The horsepower output from the driverand input to the compressor shaft.Also referred to as ShaftHorsepower.

Horsepower gas—Theactualwork required to compress anddeliver a

Appendices 253

givengasquantity, includingall thermodynamic, leakageandfluid friction losses.

Horsepower ideal—Thehorsepowerrequired to isothermallycompressgas.

Horsepower indicated—Thehorsepowercalculatedinpositivedisplace-ment compressorsusing compressor-indicatordiagrams.

Horsepower peak—The maximum power available from a driver orrequiredbya compressor.

Horsepower theoretical—The horsepower required to adiabaticallycompressgas.

Hot gas bypass valve—A valve which connects the compressor dis-charge to the compressor suction andadjusted so as tomain-tainaspecificpressureonthesuctionsidebycontrolledbypassor to regulate compressor throughput.

Humidity—Themoisture contentof air.

Humidity specific—Theweightofwatervaporintheairvapormixtureperpoundofdryair.

Humidity relative—The relative humidity of air vapor mixture is theratioofthepartialpressureofthevaportothevaporsaturationpressureat thedrybulb temperatureof themixture.

Hydrocarbons—Chemicals containing carbonandhydrogen.

ICFM—Gasflow incubic feetperminuteat the compressor inletpres-sure& temperature conditions

I.D.—Pipeorvessel insidediameter.

Ideal gas—Agas that follows theperfectgas lawswithoutdeviation.

IGV—Inletguide-vanevalve.Valve(s)installedatthecompressorsinlet.

Immiscible—Fluids that are incapable of being mixed or blended to-gether.

Impeller—Therotatingelement(s)ofadynamiccompressorthatimpartsenergy to theflowingmediumbymeansof centrifugal force.

Inches of water—Apressuremeasurement – for example 407.258 in ofwater (at 60 0F)=1Atmosphere=14.696psia.


Indicated power—Power calculated from a compressor-indicator dia-gram.

Indicator card—Apressure-volumediagramforacompressororenginecylinderproduceddirectly frommeasurementsmadeusingofadevice called theengine/compressor indicator.

Inducer—Acurved inlet sectiononan impeller.

Inerting—Process of purging a compressor or a compartment with anon-reactive such Inert gas—Agas that has little or no heat-ingvalue.

Inertia forces—Forcesgeneratedby themovingparts (such aspistons,rods,crossheads,connectingrods)withinthecompressorandacton the compressor caseand the support system.

Influent—Thefluidenteringa component.

Inlet pressure—The total pressure measured at the inlet flange of acompressor.

Inlet temperature—Thetemperaturemeasuredattheinletflangeofthecompressor.

Inlet throttle—A compressor control device (valve) that controls theoperationof the compressor.

Instrument air—Compressed air (usually dry and free from contami-nants) used specificallywithpneumatic instruments and con-trols.

Intercooler—Heat exchangers for removing the heat of compressionbetweencompressor stages.

Internal energy—Energy which a substance possesses because of themotion and configuration of its atoms, molecules, and sub-atomicparticles.

International Organization for Standardization—ISO.

Irreversible process—Aprocess inwhich thesystemand/orsurround-ings cannotbe returned to theiroriginal state.

Isentropic compression—Compressionat constant entropy

Isentropic efficiency—The ratioof isentropicwork toactualwork

ISO—InternationalOrganization forStandardization.

Appendices 255

Isobar—Constantpressure.

Isochor—Constantvolumn.

Isotherm—Constant temperature.

Isothermal compression—Compressionwherethetemperatureofagasremains constant.

Isothermal efficiency—The ratioof the isothermalpower consumptionto shaftpower input.

Joule—Theamountofworkdonebya forceofonenewtonmovinganobject through a distance of onemeter. One joule is equal tooneWATT—secondor0.737 footpounds.

Joule effect—The conversion of mechanical, electrical or magnetic en-ergy intoheat.

Joule Thompson effect—The temperature change that occurs when anon-idealgassuddenlyexpandsfromahighpressuretoalowpressure.TheratioofthechangeintemperaturedividedbythechangeinpressureisknownastheJoule-Thomsoncoefficient.

Journal—That part of a shaft or axle that rotates in a stationary fixedor tilt-padbearing.

Journal bearing—Afixedtwo-pieceortilt-padcylindersurroundingtheshaft(journal)andsuppliedwithafluidlubricant.Thefluidisthemediumthatsupports theshaftpreventingmetal-to-metalcontact.

Kelvin (K)—Pertains to the absolute scaleof temperature inwhich thedegree intervals are equal to thoseof theCelsius scale and inwhich the triplepointofwaterhas thevalue273.16Kelvin.

Kilowatt (kW)—Aunitofpowerequal to1,000watts.

Kilowatt hour (kWh)—Aunit ofworkdone inonehour at the rate of1,000watts.

Kinematic viscosity—Is the ratio of the absolute or dynamic viscositytodensity.

Kinetic energy—The energyof a bodyor a systemwith respect to themotionorvelocityof thebodyorsystemorof theparticles inthe system.


Knock out—Atermusedtodescribethecondensateflowrate.Itcanbeexpressed as gallons per minute (GPM) or pounds mass perminute (lbm/min).

kPa—Kilopascal; ametricmeasure of pressure based on force per unitarea. (1kPa=4.01 inchesofwater).

kW—Kilowatt—Aunitofpowerequal to1,000watts.

kWh—Kilowatthour—Aunitofworkbeingdone inonehour.

Lift—The distance between valve plate and the seating surfaceswhenthevalvegoes fullopen.

Liquid ring rotary compressor—Aliquidpistoncompressor isa rotarycompressor in which a vaned rotor revolves in an ellipticalcasing,withtherotorspacessealedbyaringofliquidrotatingwith it inside the casing.

LNG—Abbreviation forLiquefiedNaturalGas.

Load Electrical—Electricpower consumed.

Load Mechanical—Mechanicalor shaftpower consumed.

Load factor—Ratioofthedesigncompressorloadtothemaximumratedloadof the compressor.

Lobe—Arotatingelementof ablower.

Lubrication—Amaterial (lubricant)usedbetweenmovingpartsofma-chinery to reduce frictionandcarryawayheat.

Mach number—The ratio of the actual velocity or speed to the speedof sound in the surroundingmedium.

Manifold—Acommonchamberwithmultiple inlets and

Man way—An inspection coverorport inavesselor tank.

MAWP—Maximumallowableworkingpressure.

Maximum operating pressure—Thehighestoperatingpressurethesys-temor component isdesigned towithstand.

Mechanical efficiency—The ratio of work or power input to work orpoweroutput1.

Mesh size—Meshis thenumberofopenings inasquare inchofscreenor sieve.

Appendices 257

Micron—Micrometer or onemillionth of ameter;micron is sometimesrepresented in filtration by theGreek letter µ (mu).Amicronis 0.000039”. Contaminant particles are measured by micronsizeandcount.

MMCFD—Millionsof cubic feetper24hour.

Modulating control—Controlling a compressor at various speeds,throughput,pressureor temperature.

Moisture separator—Aunit designed to separate condensate from thecompressedair stream.

Moisture trap—Adevicedesigned toenableaccumulated liquids tobeheld fordraining fromavessel.

Mole—Thatmassof a substancewhich isnumerically equal to itsmo-lecularweight

MSDS—MaterialsSafetyDataSheets

Multistage axial compressor—Amachinehavingtwoormoreimpellersoperating in seriesona single shaft and ina single casing.

Multistage centrifugal compressor—Amachine having two or moreimpellers operating in series on a single shaft and in a singlecasing.

Natural frequency—Thenaturalfrequency,orresonantfrequency,isthefrequencyatwhichanelementwillvibratewhenexcitedbyaone time force.

N.B.—NationalBoardofBoiler andPressureVessel Inspectors.

NEC—NationalElectricalCode

Negative pressure—Apressure below that of the existing atmosphericpressure takenasazero reference.

NEMA—NationalElectricalManufacturersAssociation.

NFPA—NationalFluidPowerAssociation.

No load—The air compressor continues to run, usually at full RPM,butnoair isdeliveredbecause the inlet is either closedofformodified,not allowing inlet air tobe trapped.

Nominal efficiency—Anarbitraryfilter efficiency rating.

Noncondensables—Arethoseconstituentsinthesuctiongasthatcannot


becondensed toa liquid ,, andremoved fromthegasstream,with the coolingmediumavailable.

Noncooled compressor cylinders—Reciprocating type compressor cyl-inders, with low compression ratios, used mainly in oil andgasfieldapplications.

Noncorrosive gas—Is one that does not attack normal materials ofconstruction.

Nonlubricated compressor—Acompressordesignedtocompressairorgaswithout contaminating theflowwith lubricatingoil.

Nozzle—Amechanicaldevicedesigned to control the characteristicsofafluidas it exits (or enters) anenclosedchamberorpipe.

NPT—National Pipe Thread standard.A description of a specific pipethread

NPTT—National Pipe Thread tapered.A description of a specific pipethread

NTP—NormalTemperatureandPressure.Thisisbaseduponatempera-tureof 32degF (0degC)andapressureof 1bara.

O.D.—Ameasurement.Outsidediameter.

OEM—Original equipmentmanufacturer.

Oil aerosol—Asuspensionoffine solidor liquidparticles inagas.

Oil free—The termgenerally applies to the condition of the air or gaseitherwhen it leaves the compressor,or afterfiltration.

Oil system—Aarrangementofcomponentssuchaspump,sump,sepa-rator, cooler andfilter.

Operating pressure—The pressure at which a combination of compo-nents ismaintained innormaloperation.

Orifice—Arestrictionplaced in apipe line toprovide ameans of con-trollingormeasuringflow.

OSHA—OccupationalSafetyandHealthAdministration.

Overhung type centrifugal compressor—Aaxial inlet compressorwiththe impelleror impellersmountedonanextended shaft.

Particulate type filter—Adevicedesignedtoremovesolids,suchasdirt,scale, rust andother contaminants from thegas.

Appendices 259

Pedestal type centrifugal compressor—Asingleordual inlet compres-sorwiththeimpellerorimpellersmountedonashaftbetweentwobearings.

Performance curve—Aplot of expected operating characteristics (e.g..,dischargepressureversusinletcapacity,shafthorsepowerver-sus inlet capacity, efficiency).

pH—Measure of alkalinity or acidity in water and water containingfluids.

Pinion—Thesmallerof twomatingormeshinggears;canbeeitherthedrivingor thedrivengear.

Piston displacement—Netvolumeactuallydisplacedbythecompressorpistonwhen traveling fromcrankend toheadend.

Pitot tube—Asensingdeviceusedtomeasuretotalpressures inafluidstream.

Pleated filter—A filter element whose medium consists of a series ofuniform folds.

Pneumatic power—Thepowerof compressedair.

Pneumatic tools—Tools thatoperatebyairpressure.

Positive displacement compressors—Compressors inwhich successivevolumesof airorgasare confinedwithina closed space, andcompressed.

Potential energy—Theenergyasubstancepossessesbecauseofitsrela-tiveelevation.

Pounds per square inch—PSI—Poundsper square inch.

Pour point—The temperature at which oil begins to flow under pre-scribedconditions.

PPB—Ameasurement.PartsPerBillion

PPBV—Ameasurement.PartsPerBillionbyVolume.

PPM—Ameasurement.Partspermillion

Precooler—Aheatexchangerlocatedimmediatelyupstreamofthecom-pressor inlet.

Pressure—Forceperunit area,usually expressed inpoundsper squareinch (PSI)orBAR.


Pressure absolute—Thetotalpressuremeasuredfromabsolutezero(i.e.,fromanabsolutevacuum).

Pressure cracking—The pressure at which a pressure operated valvebegins toopen.

Pressure critical—Thesaturationpressureat the critical temperature.

Pressure dew point—The temperature at which moisture begins tocondense ina compressedair system.

Pressure discharge—The total gas pressure (static plus velocity) at thedischargeportof the compressor.

Pressure drop—Resistancetoflow.Definedasthedifferenceinpressureupstreamanddownstream.

Pressure rated—Thequalifiedoperatingpressurewhich isrecommend-ed fora componentora systemby themanufacturer.

Pressure regulating valve—A valve which enables pressure to be re-duced,orkept constantat adesired level.

Pressure relief device—A device actuated by inlet static pressure anddesigned to open during an abnormal condition to prevent ariseof internalpressure in excessof a specifiedvalue.

Pressure total—The pressure that would be produced by stopping amoving streamof liquidorgas.

Preventive maintenance—Also known as PM,maintenance performedaccordingtoafixedschedule involvingtheroutinerepairandreplacementofmachineparts andcomponents.

Pseudo critical pressure—Thesaturationpressureatthecriticaltemper-ature.Itisthehighestvaporpressurethattheliquidcanexert.

PSI—Poundsper square inch.

PSIA—Poundsper square inch, absolute.

PSID—Poundsper square inch,differential.

PSIG—Poundspersquareinch,gauge.Pressureindicatedbyapressuregauge.

Psychrometry—Therelationshipofthepropertiesofairandwatervapormixtures in theatmosphere.

Appendices 261

Pulsation bottle or damper—Areceiverfittedontheinletordischargeofa reciprocating compressor. Thedevice isdesigned to removeresonancesfromthecompressorpipingtherebyreducingnoiseand stresses.

Purging—Processofexpellinganunwantedgasorliquidfromasystemthrough the introduction of a different gas until the last rem-nantsofunwantedgashavebeen removed.

Receivers—Tanksused for thestorageofairdischarged fromcompres-sors.

Reciprocating compressors—Machines in which the compressing ele-ment is apistonhavinga reciprocatingmotion ina cylinder.

Reduced pressure—The ratio of the absolute pressure of a gas to itscriticalpressure.

Reduced temperature—The ratio of the absolute temperature of a gasto its critical temperature.

Regulator—An automatic or manual device designed to control pres-sure,flowor temperature.

Relative clearance volume—The ratio of clearance volume to the vol-umesweptby the compressingelement.

Relative humidity—Theratiooftheactualwater-vaporpartialpressuretoitssaturationpressureatthesametemperature.(consideredonlywithatmospheric air).

Relative vapor pressure—The ratio of the vapor pressure to the satu-ratedvaporpressureat the temperature considered.

Relief valve—Aspringloadedpressurereliefvalveactuatedbythestaticpressureupstreamof thevalve.

Reversible process—An ideal process,which once having takenplace,can be reversed and in so doing leave no change in eithersystemor surroundings

Reynold’s number—The numerical value of a dimensionless combina-tion of four variables (pipediameter, density andviscosity oftheflowingfluidand thevelocityofflow)maybeconsideredtobetheratioofthedynamicforcesofmassflowtotheshearstressdue toviscosity.


Rings—Circularmetallicelementsthatrideinthegroovesofacompres-sorpistonandprovide sealing.

Rolling element—A type of anti-friction bearing in lightweight com-pressors.

Rotary blowers—Acompressorbelonging to thegroupofpositivedis-placement rotary compressors.

Rotor—The rotating element of a compressor composed of theimpeller(s), shaft, seals andmay include shaft sleeves and athrustbalancepiston.

SAE—SocietyofAutomotiveEngineers.

Safety valve—Adevice that limitsfluid (liquidandgaseous)pressuresbeyondapreset level.

Safety relief valve—Anautomaticpressurerelievingvalveactuatedbythestaticpressureofthesystem,whichopensinproportiontothe increase inpressureover theopeningpressure.

Saturated vapor pressure—Thepressureexistingat thesaturation tem-perature ina closedvessel containinga liquidand thevapor.

Saturation—Occurs when the vapor is at the dew point or saturationtemperature corresponding to itspartialpressure.

SCFM—Standardcubic feetperminute.

Screw compressor—Apositivedisplacement rotary compressor.

Sea level—This is thedesignation for theaverage ISOpressure=14.69psia.

Seals—Devicesusedbetweenrotatingandstationaryparts toeliminateorminimized leakagebetweenareasofunequalpressures.

Second law of thermodynamics—Aprocess canonlyproceed ina cer-tain direction but not in the opposite direction (heat cannotpass froma colderbody toahotterbody).

Set pressure—The gauge pressure atwhich a safety valve visibly andaudiblyopensora settingatwhicha reliefvalveopens.

Shaft—Thepartof therotatingelementonwhichtherotatingpartsaremounted and bymeans of which energy is transmitted fromtheprimemover.

Appendices 263

Shaft input—Thepower requiredat the compressordrive shaft.

SI—Systeme International. The international system of unit measure-ment.

Silencer—A device fitted at a valves outlet to reduce or dampen thenoiseproducedwhen thevalveventsoropens.

Single acting—Whenapiston compressesgasonly inonedirection.

Single stage compressors—Compressors in which gas is compressedineachcylinder from initial intakepressure tofinaldischargepressure.

Single stage centrifugal compressors—Compressors having only oneimpeller.

Sleeve—Atypeofhydrodynamic journalbearing.

Slip—The internal leakagewithinarotarycompressorrepresenting thepartially compressedgas that isnotdelivered.

Sole plate—Ametallicpad,usuallyembeddedinconcrete,onwhichthecompressor feet aremounted.

Sonic flow—The point (speed of sound) atwhich air flow through anorifice cannot increase regardlessofpressuredrop.

SPC—SpecificPowerConsumption.

Specific energy requirement—The shaft input per unit of compressorcapacity.

Specific gravity—Theratioof thedensityof thematerial to thedensityofwaterat a specified temperature.

Specific heat—Thequantityofheatrequiredtoraisethetemperatureofaunitweightof a substancebyonedegree.

Specific humidity—Theweightofwatervapor inanair-vapormixtureperpoundofdryair.

Specific power—Ameasureofcompressorefficiency,usuallyintheformofbhp/100acfmoracfm/bhp

Specific volume—Is thevolumeof agivengasdividedby itsweight.

Specific weight—Theweightof agasperunitvolume

Speed—Thenumberofrevolutionsperminuteofthecompressorshaft.


Speed of sound—The rateatwhich soundwaves travel.

Stages—Stepsinthecompressionofagas,Inreciprocatingcompressors,each stage usually requires a separate cylinder, in dynamiccompressors, each requiresa separate rotordisc.

Stem—The rodconnecting thedisc to the leveronavalve.

Suction pressure—This is the pressure found on the suction side of acompressor.

Surge—The reversal of flow within a dynamic compressor resultingfrom insufficientheadandflow

Surge limit—Inadynamiccompressor,surgelimitisthecapacitybelowwhich the compressoroperationbecomesunstable.

Synthetic lubricant—A lubricating oilmade from synthetic basedma-terials.

Temperature—Thephysicalpropertyofabodythatliesbehindthecom-monnotionofhot andcold.

Temperature absolute—The temperature of a body relative to theabsolute zero temperature, at which point the volume of anidealgastheoreticallybecomeszero.(Fahrenheitscaleisminus459.67°F/Celsius scale isminus273.15°C).

Temperature discharge—The temperature of the medium at the dis-chargeflangeof the compressor.

Temperature inlet—The temperatureof themediumat the inletflangeof the compressor.

Temperature static—Thetemperatureatastagnation(speedofthefluidis zero) point in thefluidflow.Also known as the stagnationtemperature.

Thermodynamics first law of—The amount of work done on or by asystemisequaltotheamountofenergytransferredtoorfromthe system.

Thermodynamics second law of—Heat cannot, of itself, pass from acolder toahotterbody.

Thrust balancing piston—Thepartofarotatingelementthatcounteractsthe thrustdevelopedby the impellers.

Appendices 265

Tilting pad—Atypeofhydrodynamic journalbearing inacompressor.

Total package input power—Thetotalelectricalpowerinputtoaelectricmotor driven compressor, including drivemotor, cooling fan,auxiliarymotors, controls, etc.

Torque—Torsional moment or couple. It usually refers to the drivingcoupleof amachineormotor.

Two stage compressor—Compressors inwhichgas iscompressedfromaninitialpressuretoanintermediatepressureinonestageandthancompressedfromtheintermediatestagetothefinalstage.

Two step control—Load/unloadcontrolsystemthat tries tomaximizescompressor efficiency bymatching delivery to demand. Basi-cally the compressor isoperatedat full loador idle.

UL—UnderwritersLab.

Ultrasonic leak detector—An instrument designed to detect the ultra-sonic emissionsandconvert them toaudible signal.

UNC—Thread—Unifiednational coarse.

UNF—Thread—Unifiednationalfine.

Unload—A condition where the compressor runs at full RPM but nogas isdeliveredbecause the inlet is closedoff.

Unloaded horsepower—The power that is consumed to overcome thefrictional losseswhenoperating inanunloadedcondition.

Vacuum pumps—Compressors that operate with an intake pressurebelowatmosphericanddischargepressureusuallyatmosphericor slightlyhigher.

Valves—Deviceswithpassagesfordirectingflowintoalternatepathsorrestrictingflowalongagivenpath.

Vane compressor—Asingleshaft,positivedisplacementrotarycompres-sor.

Vapor—Fine separated particles floating in the air and clouding it.Asubstance in thegaseous state.

Vaporization—Theprocessofchangingfromaliquidtoagaseousstate

Vapor pressure—The pressure exerted by a vapor confined within agiven space.


V belt drive—Adrivearrangementforpowertransmissiontocompres-sors.

Venturi—A reduction in fluid pressure that resultswhen a fluid flowsthrougha constricted sectionofpipe.

Viscosity—Ameasure of the resistance of a fluid which is being de-formedbyeither shear stressor external stress.

Viscosity index (VI)—Ameasureoftherateofchangeofviscositywithtemperature.

Volumetric efficiency—The ratio and percent of the actual deliveredcapacity (measured at inlet temperature, pressure and gascomposition) to thepistondisplacement.

Volute—A stationary, spirally shaped passage that converts velocityhead topressure.

Water cooled compressors—Compressors cooled by water/glycol cir-culated through jackets surrounding the cylindersor casings.

Water solubility—The ability or tendency of a substance to blenduni-formlywithwater.

Wet bulb temperature—The temperature recorded by a thermometerwhosebulbhasbeencoveredwithawettedwickandwhirledon a sling psychrometer. Takenwith the dry bulb, it permitsdeterminationof relativehumidityof theatmosphere.

Wet gas—Agasorgasmixtureinwhichoneormoreoftheconstituentsisatitssaturatedvaporpressure.Theconstituentatsaturationpressuremayormaynotbewatervapor.

Wicking—Theverticalabsorptionofa liquid intoaporousmaterialbycapillary forces.

WMW—WeldedMinimumWall tubing

WOG—WOG—A pressure rating in psi for valves. W.Water, O. Oil,G.Gas

Work—Energy in transition and isdefined inunits of Force timesDis-tance.Work cannotbedoneunless there ismovement.

Yoke—The portion of a safety/relief valve that surrounds the spring.The springhousing.

Appendices 267

Appendix B3

Conversion Factors

TOOBTAIN MULTIPLY BY————————————————————————————————bars atmospheres 1.0133cmHg@32°F (0°C) atmospheres 76.00ftofwater@40°F (4°C) atmospheres 33.90in.Hg@32°F (0°C) atmospheres 29.92in.Hg@60°F (15°C) atmospheres 30.00lb/sq ft. atmospheres 2,116lb/sq. in. atmospheres 14.696cmHg@32°F (0°C) bars 75.01in.Hg@32°F (0°C) bars 29.53lb/sq ft. bars 2,088lb/sq in. bars 14.50ft-lb Btu 778.26Horsepower-hr Btu 3.930x10-4joules Btu 1055kilogramcalories Btu 2.520x10-1kilowatt-hr Btu 2.930x10-4grams carats 2.0x10-1ounces carats 7.055x10-3ft. centimeters 3.281x10-2in. centimeters 3.937x10-1kilogram/hrm centipoise 3.60lb/sec ft centipoise 6.72x10-4Poise centipoise 1x10-2ftofwater@40°F (4°C) cmHg 4.460x10-1in.H2O@40°F (4°C) cmHg 5.354kilogram/sqm. cmHg 135.95lb/sq. in. cmHg 1.934x10-1lb/sq./ft. cmHg 27.85ft/sec cm/sec 3.281x10-2mph cm/sec 2.237x10-2


cu in. cu cm (dry) 6.102x10-2galU.S. cu cm (liquid) 2.642x10-4liters cu cm (liquid) 1x10-3cu cm cu ft (dry) 2.832x104cu in. cu ft (dry) 1,728liters cu ft (liquid) 28.32lb cu ftH2O 62.428galU.S. cu ft liquid) 7.481cum/min cu ft/min 2.832x10-2gallons/hr cu ft/min 4.488x102liters/sec cu ft/min 4.719x10-1cu cm cu in. 16.39galU.S. cu in. 4.329x10-3liters cu in. 1.639x10-2quarts cu in. 1.732x10-2cu ft. cumeters 35.31cu in. cumeters 61,023gasU.S. cumeters 264.17liters cumeters 1.0x103meters ft 3.048x10-1cm/sec ft/min 5.080x10-1km/hr ft/min 1.829x10-2mph ft/min 1.136x10-2cm/sec ft/sec 30.48gravity ft/sec 32.1816km/hr ft/sec 1.097knots ft/sec 5.925x10-1mph ft/sec 6.818x10-1Btu gram-calories 3.969x10-2joules gram-calories 4.184kg/m gram/cm 1x10-1lb/ft gram/cm 6.721x10-2lb/in. gram/cm 5.601x10-3kg/cum grams/cucm 1000lb/cu ft grams/cucm 62.43slugs grams/cucm 1.940Btu/hr horsepower 2,546Btu/sec horsepower 7.073x10-1ft-lb/min horsepower 33,000

Appendices 269

ft-lb/sec horsepower 550kilowatts horsepower 7.457x10-1m-kg/sec horsepower 76.04metrichp horsepower 1.014Btu-sec horsepower,metric 6.975x10-1horsepower horsepower,metric 9.863x10-1kilowatts horsepower,metric 7.355x10-1m-kg/sec horsepower,metric 75.0Btu horsepower-hours 2.545x103ft-lb horsepower-hours 1.98x106kilowatt-hours horsepower-hours 7.457x10-1m-kg horsepower-hours 2.737x105cmHg@32°F (0°C) inH2O@40°F (4°C) 1.868x10-1inHg@32°F (0°C) inH2O@40°F (4°C) 7.355x10-2kg/sqm inH2O@40°F (4°C) 25.40lb/sq ft. inH2O@40°F (4°C) 5.202lb/sq in. inH2O@40°F (4°C) 3.613x10-2atmospheres inH2O@60°F (15°C) 2.455x10-3inHg@32°F (0°C) inH2O@60°F (15°C) 7.349x10-2atmospheres inHg@32°F (0°C) 3.342x10-2ftH2O40°F (4°C) inHg@32°F (0°C) 1.133in.H2O@40°F (4°C) inHg@32°F (0°C) 13.60kg/sqm inHg@32°F (0°C) 3.453x102lb/sq ft. inHg@32°F (0°C) 70.73lb/sq in. inHg@32°F (0°C) 4.912x10-1cm in. 2.54Btu Joules 9.483x10-4ft-lb Joules 7.376x10-1hp-hr Joules 3.725x10-7kg calories Joules 2.389x10-4kgm Joules 1.020 x 10-1watt-hr Joules 2.778x10-4grams/cucm kg/cum 1x10-3lb/cu ft kg/cum 62.43x10-3inH2O@40°F (4°C) kg/sq cm 3.28x10-7inHg@32°F (0°C) kg/sq cm 28.96lb/sq ft. kg/sq cm 2.048x103lb/sq in. kg/sq cm 14.22Btu kilogram-calories 3.9680


ft-lb kilogram-calories 3087m-kg kilogram-calories 4.269x102grams kilograms 1x103lb kilograms 2.205oz kilograms 35.27cm kilometers 1x105ft kilometers 3.281x103miles kilometers 6.214x10-1nauticalmiles kilometers 5.400x10-1Btu/sec kilowatts 9.485x10-1ft-lb/sec kilowatts 7.376x102horsepower kilowatts 1.341kgcalories/sec kilowatts 2.389x10-1ft/sec, fps km/hr 9.113x10-1knots km/hr 5.396x10-1m/sec km/hr 2.778x10-1mph km/hr 6.214x10-1ft/sec, fps knots 1.688km/hr knots 1.853m/sec knots 5.148x10-1mph knots 1.151grains lb, avdp 7000grams lb, avdp 453.6ounces lb, avdp 16.0poundals lb, avdp 32.174slugs lb, avdp 3.108x10-2kg/cum lb/cu ft 16.02grams/cucm lb/cu in. 27.68lb/cu ft lb/cu in. 1728atmospheres lb/sq in. 6.805x10-2ftH2O@4°C lb/sq in. 2.307inHg@0°C lb/sq in. 2.036kg/sqm lb/sq in. 7.031x102cu cm liters 1x103cu ft liters 3.532x10-2cu in liters 61.03gal Imperial liters 2.200x10-1galU.S. liters 2.642x10-1quarts liters 1.057

Appendices 271

ft-lb meter-kilogram 7.233joules meter-kilogram 9.807ft/sec, fps meter/sec 3.281km/hr meter/sec 3.600miles/hr meter/sec 2.237ft meters 3.281in. meters 39.37miles meters 6.214x10-4in. microns 3.937x10-5ft miles 5280kilometers miles 1.609nauticalmile miles 8.690x10-1inHg@0°C milibars 2.953x10-2ft/sec, fps mph 1.467km/hr mph 1.609knots mph 8.690x10-1m/sec mph 4.470x10-1ft nauticalmile 6076.1m nauticalmile 1852miles nauticalmile 1.151grains ounces, avdp 4.375x102grams ounces, avdp 28.35lb, avdp ounces, avdp 6.250x10-2cu cm ounces.fluid 29.57cu in. ounces.fluid 1.805degree (arc) radians 57.30degrees/sec radians/sec 57.30rev/min, rpm radians/sec 9.549rev/sec radians/sec 15.92x10-2radians/sec rev/min. rpm 1.047x10-1radians revolutions 6.283lb slug 32.174sq ft sq cm 1.076x10-3sq in. sq cm 1.550x10-1sq cm sq ft 929.0sq in. sq ft 144.0sq cm sq in. 6.452Btu/sec watts 9.485x10-4


Appendix C

Gas Properties and Behavior

This information is reprinted from the Gas Processors SuppliersAssociation Engineering Data Book Courtesy of The Gas ProcessorsSuppliersAssociation. The Physical Constants of Hydrocarbons, Compressibility Chartsand Pressure-Enthalpy Diagrams have been reprinted in both EnglishandSIUnits.

PhysicalConstantsofHydrocarbons—EnglishunitsCompressibilityFactorsofGases—EnglishUnits

PhysicalConstantsofHydrocarbons—SIunitsCompressibilityFactorsofGases—SIUnits

SI Units CarbonDioxide Methane Propane Ethane Ethylene Propylene Nitrogen Iso-Pentane n-Butane n-Pentane Iso-Butane Water

English Units Enthalpy—PureComponents CarbonDioxide Methane Propane Ethane Ethylene Propylene Nitrogen Iso-Pentane

Pressure-Enthalpy Diagrams

Appendices 273

Appendices 275

Appendices 277

Appendices 279

Appendices 281

Appendices 283

Appendices 285

Appendices 287

Appendices 289

290

Com


Appendices

291

292

Com


Appendices

293

294

Com


Appendices

295

296

Com


Appendices

297

298

Com


Appendices

299

300

Com


Appendices

301

302

Com


Appendices

303

304

Com


Appendices

305

306

Com


Appendices

307

308

Com


Appendices

309


Appendix D

Classification ofHazardous Atmospheres

Area classifications are divided into CLASS, DIVISION, andGROUPS.As shown in the following table:

TheaboveinformationhasbeenextractedfromNFPA70-1990,NationalElectric Code. National Electric Code andNEC are Registered Trade-marksof theNationalFireProtectionAssociation, Inc.Quincy,MA.

Appendices 311

Appendix E

Air/Oil CoolerSpecifications Check List

Appendices 313


Appendix F

Cylinder Displacement Curves

The following charts are provided as a shortcutmethod of deter-miningcylinderdisplacementasafunctionofpistonstrokeandcylinderdiameter. Curvesareprovided inbothSIunits andEnglishunits. The charts have been provided courtesy of Petroleum LearningPrograms,305WellsFargoDrive,Suite4,Houston,Texas77090.Phone:(281) 444-7632,Fax: (281) 586-9876. E-mail:[email protected].

Appendices 315

Appendices 317

Appendices 319

Appendix G

CompressorCylinder Lubrication

Thefollowingchartsareprovidedtodeterminecylinderlubricatingoilflowasafunctionofpistonstroke,cylinderdiameter,speedandthenumberofoilinjectionports. CurvesareprovidedinbothSIunitsandEnglishunits. The charts have been provided courtesy of Petroleum LearningPrograms,305WellsFargoDrive,Suite4,Houston,Texas77090.Phone:(281) 444-7632,Fax: (281) 586-9876. E-mail:[email protected].

Appendices 321


Appendix H

Troubleshooting ChartPositive Displacement Compressors

The following is a list of symptoms or problems that most fre-quently occur. The possible causes associated with each problem arelistedwith themost likely causeor causesfirst.

Appendices 323


Appendix I

Starting, Operating andMaintenance Procedures

CONTENTS• Preparation forStart-up• Operation• Maintenance

PREPARATIONFORSTART-UP

Compressor Onunits shippedwith the cylinders andpiston assemblies in-stalled,itisnecessarytocheckpistonendclearances,pistonrodrunout, crosshead shoe clearances, and piston pin through-bolt torqueand correct ifnecessary.Record these readingson the frame settingrecord sheet for future reference. Remove the suction valves from each end of the compressorcylinderstopermittheunittorunatnoloadforbreakinpurposes.Reinstall and tighten thevalve covers. Removethecompressorcrankcasecoverandcrossheadinspec-tion doors on the compressor. Inspect the compressor crankcase toinsure that it is free of all dirt, water and long-term preservatives.Inspect the compressor oil filter cases for dirt, rust, etc. Clean thefilter cases ifneeded. Fill all cooling systemswith coolantandbleedall air from thesystem.Allow the unit to sit for 2-3 hours then check inside thecompressor cylinders for coolant leaks. Ifnotalreadyinstalled, install thecompressoroilfilterelementandfiltertopusingagasketifneeded.Consulttheinstructionmanu-alforthecorrecttypeandweightofoiltobeusedinthecompressorcrankcase.

Appendices 325

Installoilinthecompressorcrankcasetotheproperlevel.Usingtheprelubehandpump,prelubethecompressorframe.Bleedallairfrom the compressor oil filter. Continue to prelube the compressoruntiloilisdeliveredtoalllubricationpointsofthecompressorframe.Avisualcheckwilldetermineifoilisreachingthemainbearing,rodbearings, andcrossheadassemblies. After ithasbeenestablished that the frame is lubricatedprop-erly, reinstall the crankcase topcoversand the crosshead inspectiondoors. Inspect the compressor lubricator reservoir forwater thatmayhave entered during shipping and storage. Clean the reservoir ifneeded.Checkthelubricatordriveforproperalignmentandeaseofoperation.This isdonebymovingthe lubricatoroilpumpcouplingbackon theoilpumpshaft and rotating theoilpumpbyhand.Ifbindingisfoundtoexist,loosenthelubricatormountingboltsandmovethelubricatorfromsidetosideorupanddownuntilunitturnsfreely.Tightenlubricatormountingboltsandconnectthecompressoroilpumpdrive coupling to theoilpumpdrive shaft. If the lubricator isnotfedfromthecompressorframebesure it isconnectedtoasupplywiththepropertypeandweightofoil,asrecom-mended in the instructionmanual, for the break-in of the compressorcylinders and packings and for the type of operation underwhich thecompressorwillbe run. Using the hand primers on lubricator pump or a suitable handprimer gun, prime and bleed all lubricator lines until they are free ofair.Continuetohandprimethesystemuntil there isoil inthecylinderboreandpacking cases.

Motor Check See themotor section of the instructionmanual for themanufac-turer’s instructions for start-up.

Auxiliaries Fill all systems that require oil such as themist lubricator for thestarter andair/gasprelubepumpmotors. Grease the cooler fan shaft and idlerbearings, ifneeded.

Control Panel Makeapreliminaryinspectionoftheoperationofthecontrolpanel.Duringthebreakinperiodoftheunititmaybenecessaryto“lockout”


certain parts of the control system in order to keep the unit running.Thisdependsonthetypeofpanelusedontheunit.Donot,underanycircumstances, lock out the low oil pressure shutdowns or completelydisable thepanel and run theunit. Should amalfunction occur on thecompressorduringthebreak-inrun,seriousdamagecouldoccurbeforetheunit couldbe shutdownby theplantoperator. Seethepanelsectionoftheinstructionmanualforcompletedetails.

Instrumentation A point-to-point continuity check must be made for all wiring.Conduitsealfittings(ifapplicable)mustbefilledpriortostart-up.Anyinstrumentationwhichwasremovedforshippingshouldbere-installedat this time.

Start-up Procedure Starttheunitandrunfor5minutes.Donotexceedthemaximumnumber of motor starts per hour. Remove the compressor crankcasecovers and check for any overheating of the crankcase main bearingsrodbearingandcrossheadassemblies.Reinstall the crankcase covers ifnoheating is found. Restart the unit and run for 30minutes.Note the compressor oilpressures,oiltemperaturesandmonitoringdeviceformainbearingtem-peraturesifprovided.Stoptheunitandrepeatthecheckforoverheatingofthecompressorbearings,(andmotorbearings)andotheritemslistedin step1. Removethesuctionvalvecoversandinspectthecylinderboreforoil, scuffing, anddebris.Check thepackingglands foroverheating. If there isno indicationof overheatingof the crankshaft, cylinderbores or packing glands, restart the unit and run for a second thirtyminuteperiod.Note theoilpressuresandoil&water temperatures. Attheendofthesecondthirtyminuteperiod,stoptheunit.Makeafinalinspectionofthecompressorcrankshaft,mainbearings,rodbear-ings,motor bearings, crosshead assemblies, rod packings and cylinderbores for overheating and proper lubrication. Inspect the bores of allcylinders for a glazing effect. If everything is found to be satisfactory,theunitcannowbeloaded.Installthesuctionvalvesandvalvecovers,and the crankcase cover. After all valve covers, piping, etc., have been installed, open theunit suctionvalve andallowpressure toflow through theunit out the

Appendices 327

blowdownvalveuntiltheunitisadequatelypurged.Onceallairisex-pelled,allowthepressuretobuildto50psiornormalsuction·pressurewhicheverisless.Itissuggestedthatairbepurgedfromtheunitusingnitrogen gas. Once all the air is adequately purged, then the nitrogenmaybepurged from theunitusing theprocessgas.

Lubrication & Cooling Thecompletelubricatingsystemofthecompressormaybeconve-niently divided into three parts of equal importance.Complete protec-tion of all frame running parts is provided by lube oil from the framesump.Aseparate,independent,forcefeedlubricatorandtubingsystemprovides lubricationforcylinderwallsandpistonrodpacking.Ashell-and-tube cooler isprovided for frameoil cooling.

Lubricating Oil Requirements A good mineral oil which provides resistance to oxidation andcorrosionisgenerallysatisfactoryforlubricationinareciprocatingcom-pressorwhichhasitscrankcasesealedofffromthecylinders.However,there is no objection to the use of a detergent type oil if this is morereadilyavailable.Thebestassuranceofobtainingasuitableoilistouseonlyproductsofwell knownmerit,producedby responsible concerns,and used in accordance with their recommendations. Do not permityourcompressortobeusedasanexperimentalunit fortryingoutneworquestionable lubricants. In some cases it may be convenient or practical to use the sametype oil in the compressor as is used in the compressor drive engine.This ispermissibleas longas theengineoil isofproperviscosity. Ifstart-upsaretotakeplacewhenambienttemperaturesarebelowfreezing, thepour point of the oilmust be low enough to insure flowto theoilpump.Heavieroil shouldbeheatedbefore starting. Ifacompoundedoilisused,thenon-corrosivenessofthisoilmustbelookedintoverycarefully.Theoilmustnotcontainsubstanceswhichmightbeinjurioustotinorleadbasebabbittsanditisalsohighlydesir-able that itbenon-corrosive to copper-leadalloys.

Mechanical Check List 1. Compressor jack screwsarebackedoff aftergrouthas setup 2. Compressoranchorbolts areproperly torqued. 3. CheckCouplingAlignment. 4. Motoranchorbolts areproperly torqued.


5. Check tightnessofadapterplatebolts toflywheelandsheave (ifapplicable).

6. After cylinder ismountedon compressor. Jamnuts areproperlytightened.

7. Checkpackinggland for clearance. (RecordClearance) 8. Packingbolts are tight. 9. Torque crossheadpinbolts.Procedure to torque thesebolts are: a. Torquethebolts b. Taptheboltwithahammer c. Re-torquethebolt. 10. Checkand record crossheadclearance. 11. Recordpistonendclearance.

—————————————

OPERATION

Preparation for Initial Start-up

CAUTIONREAD THIS PROCEDURE AND FAMILIARIZE YOURSELF WITH THECOMPRESSORAND ITSAUXILIARY EQUIPMENT PLUS YOUR COMPANYSAFETYPROCEDURESBEFOREATTEMPTINGTOSTARTTHISMACHINERY.

The followingprocedure is suggestedbefore starting theunit· forthefirst time,or afteroverhaul,or after a longperiodof idleness.

1. Removethetopcoverofthebaseandthecoversforthecrossheadsanddistancepiecesoneachcrossheadguide.Checkforcleanlinessbefore adding lube oil to the base. Be sure that no dirt, cuttings,water, etc., areallowed to remain.

2. Removeavalvefromeachendofeachcompressorcylinder.Rotatethe compressor slowly and check for piston-to-head clearances.Makesureallpartsmovefreely.Checkthatallvalvesareinstalledproperty.

CAUTIONBEFORE REMOVINGANY GAS CONTAINING PART OF THE COMPRESSOR ORASSOCIATED PIPING SYSTEM, VENT THE COMPRESSORAND SYSTEM TOAT-MOSPHERE.

Appendices 329

3. Add lubricatingoil to thebaseand to the lubeoilfilter. 4. Check the force feed lubricator for cleanliness and fill to the

proper levelwithoil. 5. Adjust all force feed lubricatorpumps to full stroke for cylinder

andpackingbreak-in. 6. Disconnectendsofforcefeedlubricatorlinesascloseaspossible

to cylinders and cross head guides. Hand pump the lubricatorstofill linesandeliminateair.

7. Connect the force feed lubricator lines and operate pumps tenmore strokes to forceoil into cylindersand rodpacking.

CAUTIONHIGH PRESSURE OIL STREAM MAY PUNCTURE SKIN. USE PROPERWRENCHAND KEEP HANDSAWAY FROM IMMEDIATE POINTWHERECONNECTION ISPURGINGAIR. 8. Prime the systemwith the lubeoilprimingpump. 9. Hand lubricate thepiston rodnext to thepacking. 10. Replaceallcoverswiththeirrespectivegasketsandtightenscrews

according to torque chart. Distance piece coversmay be left offto check forpacking leakson startup ifgas isnot sourgas.

11. Check to see that all crosshead guides or distance pieces andpackingsareventedindividuallyandwithpropersizeventlines.

12. Unload the compressor for start-up. 13. Check tobe sureallguardsare inplace.

Initial Start-up1. Openthevalvessupplyingwatertothecompressorcoolingsystem

(when required).2. Startupandoperatetheunitunderno-loadconditionsatreduced

speedwherepossible(i.e.,enginedrivenunits).Checktheoilpres-sure. When the compressor is started, an oil pressure of 20 psimustbeexperiencedwithin5 secondsor the compressormustbeimmediatelyshutdown.Donotrestartuntiladequateoilpressurecanbeassured.

3. After running theunit15 to20minutes, shutdownandcheckallbearingsandpackings forhigh temperature.

4. Checkpiping foroilorwater leaks.5. Start up again and run for approximately 20 to 30minutes.Add

oiltothecrankcasetobringtheoillevel(whilerunning)uptothe


middleof the sightglass. Shutdownand recheckasabove.6. Start theunit andbring itup to full rated speed.Apply the load.7. During the initial period of operation, pay close attention to the

machineforanyunusualhightemperature,pressure,orvibration.In the eventof equipmentmalfunctionwhere excessivevibration,noise, high temperature, or any other dangerous condition exists,the compressor shouldbe stoppedimmediately.

WARNINGIF COMPRESSOR HAS BEEN STOPPED, DO NOT IMMEDIATELY RE-MOVE EQUIPMENT COVERS. ALLOW THE UNIT TO COOL DOWN TO PREVENT POSSIBLE EXPLOSION DUE TO IN RUSH OF AIR, AND TO PREVENT INJURY WHICH MAY BE CAUSED BY HOT SURFACE.

Normal Start-up Not all of the instructions provided for initial start-ups arerequired for routine starting. The followingnotes comprise the normalstartingprocedure:1. Unload the compressor.2. Operate the force feed lubricatorpumps,byhand, for ten strokes.

(Be sure the lubricator tank iskept full).3. Handprime the frame lubeoil system.4. Turnon coolingwater supply.5. Start theunit.Check frame lubeoilpressure.6. Operate at low speed (where possible) and no load for several

minutes.Check force feed lubricator sight glasses for feed.Checklubeoil forproper level at sightgauge.

7. Bringup to rated speedandapply load.

Normal Shutdown Before shutting the unit down, it is good practice to unload thecompressorand reduce speed (wherepossible).

Emergency Shutdown

WARNINGIF COMPRESSOR HAS BEEN STOPPED, DO NOT IMMEDIATELY RE-MOVE EQUIPMENT COVERS. ALLOW THE UNIT TO COOL DOWN TO PREVENT POSSIBLE EXPLOSION DUE TO IN RUSH OF AIR, AND TO PREVENT INJURY WHICH MAY BE CAUSED BY HOT SURFACE.

Appendices 331

Recommended Operating Conditions For compressor operating speed ranges, refer toCompressor Per-formanceDataSheet.

WARNINGIMPROPER SETTING OF VARIABLE VOLUME POCKETS, FIXED VOLUME POCKETS, VALVE UNLOADERS OR OTHER UNLOADING DEVICES CAN RE-SULT IN DANGEROUS OPERATION AND POSSIBLE DAMAGE AND/OR INJURY TO EQUIPMENT OR PERSONNEL OPERATING THIS UNIT WITHOUT CLEAR-ANCE AND LOADING INFORMATION CAN RESULT IN FAILURE DUE TO. OVERLOADING, EXCESSIVE ROD LOADS OR HIGH TEMPERATURE.

—————————————

MAINTENANCE

General The diligent observation of the inspection andmaintenance pro-cedure, given in this section, will go a longway toward insuring sat-isfactory operation of the compressor. Good preventative maintenancepractice includes aperiodic checkof critical bolt torques, suchas com-pressormainandconnectingrodboltsandflywheel&sheavebolts.Therecommendedschedule shouldbe followed:

1. One (1)monthafter theunit isplaced into service.2. Six (6)monthsafter theunit isplaced into service.3. At leastonceevery twelve (12)months thereafter.

Thisscheduleshouldberepeatedaftereachcompressoroverhaul.During the first 300 hours of operation, the compressor should bechecked frequently. Check bearings and packings for excessively hightemperatures, and lookover the completeunit foroilorwater leaks. Constant care inmaintaining themachine in clean conditionwillpayoff in time, labor, and repair costs. Inspect and clean the unit atregular intervals. When repair or inspection inside the covers of the frame or cyl-inders is required, use only clean, lint-free rags. Lint or loose threadsmay cause clogging of lubrication lines or filters and result inmajortroubles.Alltoolsandworkplacesshouldbekeptfreefromgrit,dustordirt.


WARNINGWHEN WORK IS BEING DONE ON THE COMPRESSOR, THEELECTRIC MOTOR MUST BE BLOCKED IN SUCH A WAY THAT THE COMPRESSOR CANNOT TURN OVER. VALVES MUST BE CLOSED ON THE SUCTION AND DISCHARGE LINES. AIR OR GAS MUST BE BLED OFF FROM THE CYLINDERS. PRECAUTION MUST BE TAKEN TO PREVENT THE OPEN-ING OF ANY VALVE WHICH WOULD RELEASE PRESSURE AGAINST A PISTON, CAUSING IT TO ROTATE THE UNIT AT A CRITICAL MOMENT.

The following paragraphs contain information about the variousassemblies that go tomakeup the complete compressor assembly.Theconscientious operator will obtain valuable knowledge by a thoroughreadingof thismaterial from time to time.

BASE (Crankcase) Thebaseismadeofhighstrengthironandisheavilyreinforcedformaximumrigidity.Removable topandendcoversprovideaccess toallmovingpartsand theopen topdesignallowseasycrankshaft removal.Drilled lubrication passages in the base should be carefully cleaned atoverhaul.

Crankshaft and Main Bearings The complete crankshaft assembly includes the oil slinger. Thecrankshaft is rifle-drilled to carry lubrication from themain bearing tothe connecting rodbearing.Plainand thrustmainbearing shellsareofthe split,non-adjustable,precision type. Newupper and lower plainmain bearing shells are interchange-able.However, after the compressor has been run, it is preferable thattheshellsbeplacedbackintheiroriginalpositionafterremovalforanycause.Therefore, thebearing shells shouldbe somarked.

NOTEUsepencilmarkingsonlyon theparting line facesof

bearing shellsor in thebearinggrooves.

Carefully clean the crankshaft and bearing shells and saddles be-fore attempting to replace the bearing shells. Under no circumstancesshould any filing, scraping, or other fitting be done on either bearingshells or saddles. The bearing cap nuts should be tightened uniformly(criss-crossmethod) to theproper torque.

Appendices 333

Connecting Rods Theconnectingrodisasteelforging,rifle-drilledtoprovidelubri-cationtothecrossheadpinbushing.Thecapandrodarematch-marked,anda completeassemblymustbeordered if replacement isnecessary. Theupperendoftheconnectingrodcarriesapressed-inbushing.Thisbushinghasa largeoilgrooveboth insideandoutsideat itsmid-point,with communicating holes to admit oil from the connecting rodto the crosshead pin. The inside diameter of the bushing has equallyspaced, helical oil spreader grooves to insure adequate lubrication be-tweenbushingandcrossheadpin.Thisbushingisprecisionboredafterassemblyintherodtoinsurepropersizeandlocation.Ifthesebushingsare replaced in the field, extreme care should be used in maintainingtheboreof thenewbushingsparallel to andproperly spaced from thecrankpinbore.

THE CAPSAND RODSAREMATCH-MARKED. NUMBERED BY THROWANDHAVETHEIRRESPECTIVEWEIGHTSSTAMPEDON.ALWAYSINSTALLRODSWITH THIS INFORMATION UP. The rod cap is removed and re-placedwhile thecrank throw is in theouterpositionand the rod itselfcanbeeasedintooroutofpositionwiththecrankthrowslightlybelowthe innerposition.

Crosshead Guides Thecrossheadguideshavelubricationholeswithmeteringfittingsfor both top and bottom slides. BLOCKING OF THESE FITTINGSMAYCAUSETHECROSSHEADSHOESTOHEATUPANDSEIZE. The nuts and capscrews holding the guide to the base must betorqued evenly (criss-cross method) to prevent cocking of the guiderelative to the base and crankshaft. Large side covers on the guide al-loweasyaccess to thecrosshead,connectingrod,androdpacking.Thecrosshead can be removed through these openings without disturbingthe cylindermounting.

Crosshead—Removal and Installation Thecrosshead ismadeofductile ironandhas removable topandbottomsteelbackedshoeswithdurablebearingmaterialon theslidingsurface. Flat head screws and locknuts hold the shoes firmly in place,andthesemustbetorquedevenly.Likethemainbearingsmaintenance,cleanlinessisanimportantfactorduringtheassemblyofshoestocross-headandcrosshead toguide.


Cylinder Body Thecylinderbodyisprovidedwithdrilledwaterpassages,topandbottom,whichconnectthewaterinletandoutletwiththecoolingsource.Whenever thewater jacketcoverson thecylindersidesareremovedtoclean out deposits, the drilled passages should also be cleaned out. Ifthe pipe plugs in the crank end of the drilled passages are removed,they should be coatedwith goodwaterproof sealer and replaced. Thiswillpreventwater seepage into theatmosphericvent space. Lube oil, from the force feed lubricator tubing system, passesthrough a check valve and into a fitting on the outside of the upperflange of the cylinder. From here, the flow is through the wall of thedischarge flange, into a tube and fitting assemblywhich connects to adrilled passage near or at the top of the cylinder bore.Cylinders havetwo point lubrication, one in the upper flange and one in the lowerflange.Duringgeneraloverhaultheoilholeforcylinderborelubricationshould be cleaned out and all steel tubing checked for soundness andtightness.

Cylinder Head Thegasket,whichisusedtosealthecylinderheadtothecylinderbody,maybere-usedonlyifanewoneisnotavailable.However,ifpos-sible,anewgasketshouldbefitted if theheadhasbeenremovedafteraperiodof running.Thewater seal grommets should also be replacedat the same time.

Pistons, Piston Rings and Piston Rod Thepiston to rodnut is torquedafter locking twocrossheadnutson the crossheadendof the rodand clamping them in a largevise. (Ifthepistonisunusuallyheavy,providesomekindofsupportduringthisoperation.). DONOT clamp onto the piston rod. The piston nut mayalsobetorquedwhenrodandpistonareinplaceandtherodis lockedin the crosshead. The piston end clearance is adjusted by mounting the head andmeasuring theclearanceonbothendswitha feelergauge.Check theseclearanceswiththecrossheadnutsnuggeddown.Ifadjustmentisneces-sary, loosen crosshead nut, remove cylinder head and turn piston androdassembly.Once againDONOTusewrencheson rod shank;do allturningofpiston rod through thepiston to rodnut. Piston rings are designedwith proper end clearance.However, it

Appendices 335

is a good idea to check this in the cylinder.Rings should be square intheborewhenchecking.

Piston Rod Packing Thepurposeofpiston rodpacking is toprevent the loss of gasesfromthecylinderalongthepistonrod.Duringinitialoperationapack-ingmay leak or tend to overheat. This is a temporary conditionwhiletheringsareproperlymatingtotherodandcase.Asageneralguide,atemperatureinexcessofthatwhichcanbetoleratedbyrestingthehandon the packing flange will indicate too fast a rate of wearing in. Thenominalrateoflubricationforthepistonrodpackingisspecifiedbythemanufacturer.However,definite lubricationratesandtimeintervalsforwearing-inofthepackingaredifficulttoprescribehere.Experiencehasindicated that these factorsmay varywidely on different applications.Ifthereisconcernforproperlubricationratecontactthemanufacturer’srepresentative. Rodpackingmustberemovedandreplacedasacompleteunit.Thisispossibleafterfirstremovingthecylinderhead,unlockingthecrossheadnutandscrewingthepistonrodoutofthecrosshead.Leavethecrossheadintheguideandbarcompressoroveruntilcrossheadmovestoinnerdeadcenterpositiontoallowremovalofthepacking.Pullrod(withsplitsleeveoverthreads)throughthepackinguntil itclears.Disconnecttubefittingsandremovescrewsholdingpacking tocylinder. When disassembling a packing, always carefully note and recordthepositionof eachpacking cupandeach ring, and thedirection eachringfaces.Ordinarily,packingcupsandglandsarenotsubjecttoseverewear. It is, therefore,possible tomakerepairsbyorderingonlyasetofnewrings.Itisalwaysgoodpracticetokeepcompletesparesetsofnewrings for thepackingassembly. Before installing a newpacking assembly, it is important that thepistonrodbecarefullychecked.Iftherodisworn,rough,pitted,orhasataper,itmustbereplaced.Thepackingcupsandgland,andallpartsthatarenotreplacedbynewparts,mustbesoakedandthoroughlycleanedin a non-acid solvent. They should then be blown dry and examinedclosely forunusualnicksorburrswhichmight interferewith the ringsfreefloatingorcontactwiththerod.Particularcaremustbetakenwithringsmade of softmetals and it is very important thatwiper rings behandledandinstalledcarefullytopreventdamagetothescrapingedges. Theringsmustbeplacedinthepackingcupsinthesameposition


(facingoriginaldirection)as theoriginal set. The stuffing-box bore for the rod packing must be cleaned andexamined forburrs. If found,burrs shouldbecarefullyandcompletelyremoved.Anewmetallic gasket shouldbeplaced in the groove of thepacking cupwhich seats in the stuffingbox. When installing a new or rebuilt packing assembly, the attachingscrewsmustbe tighteneduniformly.Thegasketmustnot cockorbindand theassemblymust seat squarely in thebore.

NOTEThepacking caseflangemustnot contact the cylinderbody.

Clearancemust exist at thispoint to ensurepressure is applied tothe seal ringwhenpacking screwsare tightened.

Beforeconnectingthetubingtothepackingflange,handpumptheforce feed lubricatoruntil oil runs fromone of thedisconnected tubes.Connect this tube to the respectivehole in thepackingflangeandcon-tinue to pump lubricator 12 to 15 more strokes. Connect vent tubing.With split sleeve still on the rod threads, insert the piston rod (withpistonstillassembledthereon)backthroughthecylinderheadopeningand through the packing assembly. Remove the split sleeve, add thecrossheadnut, and thread thepiston rodbackonto the crosshead.

Valve Installation Suctionanddischargevalvesmustbeinstalledintheproperdirec-tion.This canbedeterminedbyfirst inspecting thevalve to seewhichdirection the springspush thevalveplates closed. Thegaswillflowinthesamedirectionthatthevalveplatemoveswhen it compresses the spring toward the center of the cylinder. Thedischargevalvesare installedwith thespringsaway fromthecenterofthe cylinder.

Valve Replacement

WARNINGBEFORE REMOVING ANY GAS CONTAINING PART OF THE COMPRES-SOR OR ASSOCIATED GAS PIPING SYSTEM, VENT COMPRESSOR AND SYSTEM TO ATMOSPHERIC PRESSURE.

Appendices 337

Proceedwithvalve removal in the followingmanner:

1. Loosen bolts or nuts holding valve cap. DONOT remove com-pletely until cap is pulled out far enough to vent any pressuretrappedunder cap.

2. Removevalve cap. InspectO-ring; replace ifdefective.

3. Loosen setscrew in valve retainer (bottom valves only); insertthreadedpuller intovalve retainerand remove.

4. Using threadedvalvepuller, removevalve fromseat incylinder.

5. Removegasket, inspect and replaceasneeded.

6. Clean thegasket surfaceonvalveand invalvepocket.Useneworagoodgasket.

7. Placevalve inpocket,FACINGPROPERDIRECTION.

8. LocatetheretainerontopofthevalveassemblyandgreaseoroiltheO-ringofthevalvecapforcingthecapintoplacebytorquingthenutsor screwsevenly.

NOTEThevalve capflangemustnot contact the cylinderbody.

Clearancemust exist at thispoint to insure thatpressure is applied to thevalveand retainergasketswhen thevalve capboltsornutsare tightened.

To replace a valve into a bottom port (this is a discharge port),proceedas follows: 9. Invert retainer. Place valve on top of retainer with valve guard

facingout (away fromcylinder). Slipgasketonvalveassembly. 10. Lift the complete works up into the bottom port, making sure

that thevalve seat entersfirst. 11. Tighten the retainer setscrew just enough to hold everything in

place. 12. Lubricatevalve capO-ring.

NOTE:Atthecompletionofanymaintenance,itisimportanttobarthecompressor crankshaft over, at least one complete revolution, to checkforadequate runningclearances.


RECOMMENDED COMPRESSOR MAINTENANCE SCHEDULE

Appendices 339


Appendix J

Basic Motor FormulasAnd Calculations

The following formulas are provided courtesy of Baldor ElectricCompany to assist the reader in selecting the propermotor driver forhis compressorapplication. The Baldor Electric Company is located at 5711 R.S. Boreham, Jr.St.,FortSmith,AR72901.Phone: (479)646-4711ext.5123.Fax(479)648-5281.WebSitewww.baldor.com

BASICMOTORFORMULASANDCALCULATIONS

Theformulasandcalculationswhichappearbelowshouldbeusedfor estimatingpurposes only. It is the responsibility of the customer tospecify the required motor Hp, Torque, and accelerating time for hisapplication. The salesmanmay wish to check the customers specifiedvalues with the formulas in this section, however, if there is seriousdoubt concerning the customersapplicationor if the customer requiresguaranteed motor/application performance, the Product DepartmentCustomerServicegroupshouldbe contacted.

Rules of Thumb (Approximation)At1800 rpm,amotordevelopsa3 lb.ft.perhpAt1200 rpm,amotordevelopsa4.5 lb.ft.perhpAt575volts, a 3-phasemotordraws1ampperhpAt460volts, a 3-phasemotordraws1.25ampperhpAt230volts a 3-phasemotordraws2.5ampperhpAt230volts, a single-phasemotordraws5ampperhpAt115volts, a single-phasemotordraws10ampperhp

Appendices 341

Mechanical Formulas

HPx5250 Torquex rpm 120xFrequencyTorque in lb.ft.= ———— HP= —————— rpm= —————— Rpm 5250 No.ofPoles

Temperature Conversion

DegC= (DegF - 32)x5/9

DegF= (DegCx9/5)+32

High Inertia Loads

WK2x rpmt= ————— 308xTav. WK2= inertia in lb.ft.2 t=accelerating time in sec. WK2x rpm T=Av. accelerating torque lb.ft..T= ————— 308x t

Load rpminertia reflected tomotor=Load Inertia(——————)2

Motor rpm

Synchronous Speed, Frequency and Number of Poles of AC Motors

ns=120x f f=Pxns P=120x f ————— ———— ———— P 120 ns

Relation between Horsepower, Torque, and Speed

Txn 5250HP 5250HPHP= ———— T= ———— n= ————— 5250 n T

Motor Slip

ns -n%Slip= ———— x100 ns


Code KVA/HP Code KVA/HP Code KVA/HP Code KVA/HP A 0-3.14 F 5.0 -5.59 L 9.0-9.99 S 16.0-17.99 B 3.15-3.54 G 5.6 -6.29 M 10.0-11.19 T 18.0-19.99 C 3.55-3.99 H 6.3 -7.09 N 11.2-12.49 U 20.0-22.39 D 4.0 -4.49 I 7.1 -7.99 P 12.5-13.99 V 22.4&Up E 4.5 -4.99 K 8.0 -8.99 R 14.0-15.99

SymbolsI = current inamperesE = voltage involtskW = power inkilowattskVA = apparentpower inkilo-volt-amperesHP = outputpower inhorsepowern = motor speed in revolutionsperminute (RPM)ns = synchronous speed in revolutionsperminute (RPM)P = numberofpolesf = frequency in cyclesper second (CPS)T = torque inpound-feetEFF = efficiencyasadecimalPF = power factorasadecimal

Equivalent Inertia Inmechanicalsystems,allrotatingpartsdonotusuallyoperateatthesamespeed.Thus,weneedtodeterminethe“equivalentinertia”ofeachmovingpart at aparticular speedof theprimemover. The total equivalentWK2 for a system is the sum of theWK2 ofeachpart, referenced toprimemover speed.Theequation says:

NpartWK2EQ=WK2

part(—————)2

Nprimemover

This equation becomes a common denominator on which othercalculationscanbebased.Forvariable-speeddevices, inertia shouldbecalculatedfirst at lowspeed. Let’s look at a simple systemwhich has a primemover (PM), areduceranda load.

Appendices 343

WK2=900 lb.ft.2

WK2=100 lb.ft.2 (as seenatoutput shaft) WK2=27,000 lb.ft.2———————————————————————————————— PRIMEMOVER 3:1GEARREDUCER LOAD

Theformulastates that thesystemWK2equivalent isequal to thesumofWK2

parts at theprimemover’sRPM,or in this case:

Red.RPM LoadRPMWK2

EQ=WK2pm+WK2

Red. (—————)22+WK2Load(—————)2

PMRPM PMRPM

Note: reducerRPM=LoadRPM

1 1WK2

EQ=WK2pm+WK2

Red.(———)2 2+WK2Load (———)2

3 3

TheWK2equivalent isequaltotheWK2oftheprimemover,plustheWK2oftheload.ThisisequaltotheWK2oftheprimemover,plustheWK2 of the reducer times (1/3)2, plus theWK2 of the load times(1/3)2. Thisrelationshipofthereducertothedrivenloadisexpressedbythe formulagivenearlier:

NpartWK2EQ=WK2

part(——————)2

Nprimemover

In other words, when a part is rotating at a speed (N) differentfrom the prime mover, theWK2

EQ is equal to theWK2 of the part’sspeed ratio squared. In theexample, the result canbeobtainedas follows: TheWK2equivalent is equal to:

1 1WK2

EQ=100 lb.ft.2+900 lb.ft.2(——)2 2+27,000 lb.ft.2 (——)2

3 3

Finally: WK2

EQ= lb.ft.2pm+100 lb.ft.2Red+3,000 lb.ft2Load WK2

EQ=3200 lb.ft.2


ThetotalWK2equivalent is thatWK2seenbytheprimemoveratits speed.

Electrical Formulas

I =Amperes; E = Volts; Eff = Efficiency; pf = Power Factor; KVA = Kilovolt-amperes;kW=Kilowatts

Locked Rotor Current (IL) From Nameplate Data

577xHPxKVA/HPThreePhase: IL=—————————— ESee:KVA/HPChart

1000xHPxKVA/HPSinglePhase: IL= —————————— E

Appendices 345

EXAMPLE: Motornameplate indicates10HP,3Phase, 460Volts,CodeF.

577x10x (5.6or6.29) IL= —————————— 460

IL=70.25or78.9Amperes (possible range)

Effect of Line Voltage on Locked Rotor Current (IL) (Approx.)

ELINE IL@ELINE= IL@EN/Px ——— EN/P

EXAMPLE: Motorhasalockedrotorcurrent(inrushof100Amperes(IL)attheratednameplatevoltage (EN/P)of 230volts.

What is ILwith245volts (ELINE) applied to thismotor? IL@245V.=100x254V/230V IL@245V.=107Amperes

Basic Horsepower Calculations Horsepower isworkdoneperunitof time.OneHPequals33,000ft-lbofworkperminute.Whenwork isdonebyasourceof torque (T)toproduce (M) rotationsaboutanaxis, theworkdone is:

radiusx2πx rpmx lb.or 2πTM

When rotation is at the rateN rpm, theHPdelivered is:

radiusx2πx rpmx lb. TN HP= ——————————— = ——— 33,000 5,250

Forverticalorhoistingmotion:


WxS HP= ————— 33,000xE

Where: W = totalweight in lbs. tobe raisedbymotor S = hoisting speed in feetperminute E = overall mechanical efficiency of hoist and gearing. For

purposesof estimating E = .65 for eff. ofhoist andconnectedgear.

For fansandblowers:

Volume (cfm)xHead (inchesofwater)HP= —————————————————— 6356xMechanicalEfficiencyofFan

Or

Volume (cfm)xPressure (lb.Per sq. ft.)HP= —————————————————— 3300xMechanicalEfficiencyofFan

Or

Volume (cfm)xPressure (lb.Per sq. in.)HP= —————————————————— 229xMechanicalEfficiencyofFan

For purpose of estimating, the eff. of a fan or blowermay be as-sumed tobe0.65.

Note:AirCapacity(cfm)variesdirectlywithfanspeed.DevelopedPres-surevarieswithsquareoffanspeed.Hpvarieswithcubeoffanspeed.

Forpumps:

GPMxPressure in lb.Per sq. in. xSpecificGrav.HP= —————————————————————— 1713xMechanicalEfficiencyofPump

Appendices 347

Or

GPMxTotalDynamicHead inFeetxS.G.HP= ——————————————————— 3960xMechanicalEfficiencyofPump

whereTotalDynamicHead=StaticHead+FrictionHead

Forestimating,pumpefficiencymaybeassumedat 0.70.

Accelerating Torque The equivalent inertia of an adjustable speed drive indicates theenergy required to keep the system running. However, starting or ac-celerating the systemrequires extra energy. The torque required to accelerate a body is equal to theWK2 ofthe body, times the change in RPM, divided by 308 times the interval(in seconds) inwhich this acceleration takesplace:

WK2N (in lb.ft.)ACCELERATINGTORQUE= ———————— 308t

Where: N = Change inRPM W = Weight inLbs. K = Radiusofgyration t = Timeof acceleration (secs.) WK2 = Equivalent Inertia 308 = Constantofproportionality

Or WK2NTAcc= ——— 308t

The constant (308) is derived by transferring linearmotion to angularmotion,andconsideringaccelerationduetogravity.If, forexample,wehave simplyaprimemoveranda loadwithno speedadjustment:


Example 1 PRIMELOADER LOAD———————————————————————————————— WK2=200 lb.ft.2 WK2=800 lb.ft.2

TheWK2EQ isdeterminedasbefore:

WK2EQ=WK2

pm+WK2Load

WK2EQ=200+800

WK2EQ=1000 ft.lb.2

Ifwewanttoacceleratethisloadto1800RPMin1minute,enoughinformation is available to find the amount of torque necessary to ac-celerate the load. The formula states:

WK2EQN 1000x1800 1800000

TAcc= ————— or ————— or ————— 308t 308x60 18480

TAcc=97.4 lb.ft.

Inotherwords,97.4lb.ft.oftorquemustbeappliedtogetthisloadturningat 1800RPM, in60 seconds. Note that TAcc is an average value of accelerating torque duringthespeedchangeunderconsideration. Ifamoreaccuratecalculation isdesired, the followingexamplemaybehelpful.

Example 2 The time that it takes to accelerate an inductionmotor from onespeed toanothermaybe found from the followingequation:

WR2x change in rpmt= —————————— 308xT

Where: T = Average value of accelerating torque during the speed

changeunder consideration.

Appendices 349

t = Time themotor takes to accelerate from the initial speedto thefinal speed.

WR2 = Flywheel effect, or moment of inertia, for the drivenmachinery plus the motor rotor in lb.ft.2 (WR2 of drivenmachinerymustbe referred to themotor shaft).

The application of the above formulawill now be considered bymeans of an example. FigureA shows the speed-torque curves of asquirrel-cage induction motor and a blower which it drives. At anyspeed of the blower, the difference between the torquewhich themo-torcandeliverat itsshaftandthe torquerequiredby theblower is thetorque available for acceleration. Reference to FigureA shows that theacceleratingtorquemayvarygreatlywithspeed.Whenthespeed-torquecurves for themotor and blower intersect there is no torque availableforacceleration.Themotorthendrivesthebloweratconstantspeedandjustdelivers the torque requiredby the load. Inordertofindthetotaltimerequiredtoacceleratethemotorandblower, theareabetween themotorspeed-torquecurveandtheblowerspeed-torquecurveisdividedintostrips,theendsofwhichapproximatestraight lines.Eachstripcorresponds toa speed incrementwhich takesplacewithinadefinitetimeinterval.ThesolidhorizontallinesinFigureArepresenttheboundariesofstrips; thelengthsofthebrokenlinestheaverageacceleratingtorquesfortheselectedspeedintervals.Inordertocalculatethetotalaccelerationtimeforthemotorandthedirect-coupledblower it isnecessary tofind the time required to accelerate themotorfrom the beginning of one speed interval to the beginning of the nextinterval and addup the incremental times for all intervals to arrive atthetotalaccelerationtime.If theWR2ofthemotorwhosespeed-torquecurve is given inFigureA is 3.26 ft.lb.2 and theWR2of theblower re-ferred to themotor shaft is 15 ft.lb.2, the totalWR2 is:

15+3.26=18.26 ft.lb.2,

And the total timeof acceleration is:

308

WR2T1

rpm1+ T2rpm2

+ T3rpm3

+ . . . . . . .+ T9rpm1; E

Or


t= 30818.26

46150+ 48150+ 47300+ 43.8

300+ 39.8

200+ 36.4

200+ 32.8

300+ 29.6

100+ 11408 B

t=2.75 sec.

Figure A Curves used to determine time required to accelerate inductionmotorandblower

Accelerating Torques T1=46 lb.ft. T4=43.8 lb.ft. T7=32.8 lb.ft. T2=48 lb.ft. T5=39.8 lb.ft. T8=29.6 lb.ft. T3=47 lb.ft. T6=36.4 lb.ft. T9=11 lb.ft.

Duty Cycles Sales orders are often enteredwith a note under special featuressuchas: “Suitable for10 startsperhour”Or “Suitable for3 reversesperminute”

Appendices 351

Or “Motor tobe capableof accelerating350 lb.ft.2”Or “Suitable for5 starts and stopsperhour”

Orderswithnotessuchasthesecannotbeprocessedfor tworea-sons.1. The appropriateproductgroupmustfirst be consulted to see if a

design is available thatwill perform the requiredduty cycle and,ifnot, todetermine if the typeofdesign required fallswithinourpresentproduct line.

2. Noneoftheabovenotescontainsenoughinformationtomakethenecessary duty cycle calculation. In order for a duty cycle to becheckedout,thedutycycleinformationmustincludethefollowing:a. Inertiareflectedtothemotorshaft.b. Torqueloadonthemotorduringallportionsofthedutycycle

includingstarts,runningtime,stopsorreversals.c. Accuratetimingofeachportionofthecycle.d. Informationonhoweachstepofthecycleisaccomplished.For

example, a stop can be by coasting, mechanical braking, DCdynamicbrakingorplugging.Areversalcanbeaccomplishedbyplugging,orthemotormaybestoppedbysomemeansthenre-startedintheoppositedirection.

e. Whenthemotorismulti-speed,thecycleforeachspeedmustbecompletelydefined,includingthemethodofchangingfromonespeedtoanother.

f. Anyspecialmechanicalproblems,featuresorlimitations.

Obtaining this information and checkingwith the product groupbeforetheorderisenteredcansavemuchtime,expenseandcorrespon-dence. Duty cycle refers to the detailed description of awork cycle thatrepeatsinaspecifictimeperiod.Thiscyclemayincludefrequentstarts,plugging stops, reversals or stalls. These characteristics are usually in-volvedinbatch-typeprocessesandmayincludetumblingbarrels,certaincranes,shovelsanddraglines,dampers,gate-orplow-positioningdrives,drawbridges, freight and personnel elevators, press-type extractors,somefeeders,pressesofcertaintypes,hoists,indexers,boringmachines,cinder block machines, key seating, kneading, car-pulling, shakers


(foundry or car), swaging and washing machines, and certain freightandpassengervehicles.Thelistisnotall-inclusive.Thedrivesfortheseloadsmustbecapableofabsorbingtheheatgeneratedduringthedutycycles.Adequate thermal capacitywouldbe required in slip couplings,clutchesormotorstoaccelerateorplug-stopthesedrivesortowithstandstalls.Itistheproductoftheslipspeedandthetorqueabsorbedbytheloadperunit of timewhich generates heat in thesedrive components.All the eventswhich occur during the duty cycle generate heatwhichthedrive componentsmustdissipate. Because of the complexity of the duty cycle calculations and theextensiveengineeringdataperspecificmotordesignandratingrequiredfor the calculations, it isnecessary for the salesengineer to refer to theproductdepartment formotor sizingwithaduty cycleapplication.

Last Updated September 1, 1998

Copyright©2007,BaldorElectricCompany.AllRightsReserved.

353

Index

Aabsolute velocity 19, 23acceleration 100, 101, 102 probes 100accelerometers 100, 112acoustic resonances 194actual surge 88adiabatic discharge temperature 54adiabatic efficiency 12altitude 54ambient pressure 65, 66ambient temperature 54, 65, 66American Petroleum Institute

Standard 618 98amplitude 100analyzers 111angle of attack 16angle of deviation 21annubar 112antisurge protection 69approach factor 93Avogadro’s Law 8axial 24, 81 component 17 compressor 15, 24, 25, 64, 82,

97, 118, 119, 120

Bbalanced opposed compressor 51 reciprocating 51 separable 77balance drum 99Barber, John 3base plate 53 resonance 101, 102bearing 51, 53, 70, 84, 99, 117, 126, 128

housing design 98 journals 33 looseness 122 shoe 122 spans 98belt driven 28blades 117, 119 vibration 19bleed valves 69blower 27blowers 28, 81, 98, 125blow-off 108boosters 27borocilicate microfiber 58bottle unloader 45brake horsepower (BHP) 12, 39

Ccamber line 21camber angle 21capacity control 73casing 99 pressure rating 98centrifugal 24, 81, 131, 133, 135 barrel compressor 99 compressor 15, 22, 23, 24, 25,

64, 69, 77, 98, 132, 137 forces 18, 19characteristic speed curves 64choke 70 tube 199, 203chord 17 length 20clearance pockets 178clearance volume 39, 56, 59, 70 pockets 147


coalescing 125 elements 58compressibility 9, 69, 70 factor 10, 12, 20, 68, 93compression load 33compression ratio (CR) 12, 56, 59,

63, 68, 176 per stage 55compression rings 39compressor 7, 29, 56, 63, 81, 115 characteristic curve 17, 24, 25,

26, 63, 65, 70, 73, 74 cylinders 53, 146 discharge pressure 119 efficiency 24, 83, 117 fouling 84 map 83 piston 53, 122 pressure 7 ratio 24, 83 stall line 15 throughput 39, 41, 44 valves 125concave inlets 57concave surface 16concentric ring valve 211, 213, 214,

217connecting rod 32, 33, 51, 53contamination 119, 120converter 107convex inlets 57counterweight 37coupling selection 98crankcase 33crank-end 33, 35, 37, 39, 121, 164 collar 36 valves 122crankpin 33 bearings 44

throws 51crankshaft 32, 33, 36, 51, 53 bearings 44 journal 33CR, compression ratio 12, 55, 56,

59, 63, 68, 176critical pressure 10critical speed 60, 97critical temperature 10, 94crosshead 32, 33, 37, 51, 183, 184,

185, 188, 189, 190, 191 cylinder interface 121 end 33 guide 33, 35, 44, 184, 191 pin 44 shoes 187Ctesibius of Alexandria xi, 1, 2Cv 106cyclone separator 125cylinder 32, 37, 46, 51, 126, 147,

148, 149, 150, 151, 153, 154, 155, 158, 161, 162, 163, 164, 165, 166, 168, 170, 172, 174, 177, 183, 184, 186, 187, 188, 189, 191, 192, 194, 197, 199, 202, 206, 207, 216

bore 37, 220 centerline 41 clearance 218 diameter 55, 56 displacement curves 56 gas temperature 41 head 56 liner 39 loading 41 ports 216 size 55 toe 195 valve cavity 41

Index 355

DDalton’s Law 11density 93design compression ratios 64design conditions 54diaphragms 117differential pressure 68, 83diffuser 22, 23digital acoustic analysis 195discharge 106 bottles 49, 197, 206 coefficient 93 port 58 pressure 33, 54, 70, 83, 86, 93 pulsation bottles 47, 194 temperature 55 throttling 73 control 75 valve 39, 108, 121, 124, 213 cover temperatures 117 seats 56 throttling 59, 60distance piece 34, 35, 120, 164, 184double-acting pistons 37drag 16, 17dual variable control 86Dumball, John 3, 4DVA 214dynamic 63, 81, 115 compressor 15, 64, 70, 73, 74,

108, 112, 119, 125 surge 125 displacement 85 pressure drop 195 rotor stability analysis 98 valve analysis 213

Eeffective flow area (EFA) 215

efficiency 17, 84electrical 116electric control 105electric motor 28, 52, 115electric motor driven pump 46enclosed 22engine 53entropy 11equal percentage 105, 106exit blade angle 21exit flow angle 21

Ffactor Z 10finger type unloader 42, 43, 44first law of thermodynamics 11fixed blade row 20fixed volume clearance pocket 174,

179flexible shaft 97, 98floats 112flow 24, 67, 68, 77, 111, 112 and horse power 79 coefficient 23, 24 orifice 68 squared 87flutter 19, 214, 216, 218forced-feed method 44fractional frequency whirl 98frame 32, 33, 51frequency 100fuel gas booster 77

Ggas compressibility 67 factor 12gas density 19gas discharge pressure 54gas engine 52, 115


gas expansion factor 94gas horsepower 12gas path analysis 117, 124gas pressure 70gas properties 54gas suction pressure 54gas suction temperature 54gas turbine 52, 77, 115gate valve 105, 106gear 98 noise 101, 102globe valve 105, 106guide vanes 23, 73

Hharmonic resonance analysis 49harmonics 193head 10, 20, 24, 67, 68, 184 adiabatic 12head-end 35, 37, 39, 77, 79, 80,

121, 174 bypass 41 suction valve 178 unloader 42, 43Hero of Alexandria 1Hertz 102horsepower 15, 77 calculations 12hydrates 70hydrodynamic bearing 46, 98, 122,

129 whirl orbits 123hydrodynamic journal bearing oil

whirl 102hydrodynamic sleeve-type

bearings 59hydrostatic testing 161, 165

IIdeal Gas Law 7, 8ideal head 23impeller 22, 23, 24, 84, 98, 99, 117, 119 inlet (eye) 22inlet blade angle 21inlet flow angel 21inlet pressure 67inlet temperature 119instruments 111integral engine compressor 3, 31,

52, 53, 150integral engine compressors 157,

176 reciprocating 98intercooler temperature 54I/P 107isolation valve 105isometric piping 49

Jjournal 122 bearing 33, 59

Kkinetic energy 23knife valve 105, 106knock-out vessel 125k value of gas 56

Llateral 98 bending 98 modes 99level 112lift 16linear 105, 106line booster 107liquid level 111

Index 357

liquids 70, 125load 51, 70lobe 27locknut 37looseness 101lubrication 44, 217, 219 oil 126, 129, 219, 220 rate 218lubricity 126

Mmagnetic pickups 112manometers 112mass flow rate 9mass in lb 100maximum motor current 86mechanical 116Mils 102minimum suction pressure or

temperature 86misalignment 101molecular weight 8, 10, 63, 64, 67,

68, 69, 70, 84, 93, 151, 218mole fraction 11moles 8mole weight 20moving blade row 20multi-cylinder balanced opposed

compressor 31

Nnatural circular frequency 100natural frequency 82, 97, 98natural lateral frequency 99needle valve 106Newtons 32non-synchronous precession 98

Ooil-flooded compressor 58oil-free compressors 58oil heater 46oil reservoir 29, 33oil viscosity 217oil wedge 98oil whirl 101oil wiper packing 35open 22operating point 65, 82, 88 line 90orbits 98orifice 112 plate 112original equipment manufacturer

115overall compression ratio 54overspeed 69

Ppackager 115packing 158partial pressures 10passive gravity-feed method 44PEEK 212, 219percent clearance 56performance 116 problems 117piezometers 112pinch valve 106piping isometric 50piston 29, 32, 33, 35, 37, 39, 42, 44,

46, 51, 53, 56, 70, 98, 121, 122, 148, 153, 155, 158, 165, 168, 176, 184, 185, 186, 189, 190

balance 37 diameter 33 displacement 39, 56


rings 220 rod 32, 34, 35, 37,51, 70, 120,

121, 168, 176, 183 droop 35 nut 122 packing 220 valve 106pitch 21Pitot tubes 112plates 40plate valve 41plug 42 type suction valve 177 type unloader 42, 76 valve 105, 106pneumatic control 105pocket unloader 41polytropic efficiency 68polytropic head 68, 87poppet valves 40, 41, 165, 217ported plate 218 valve 211, 213, 214, 215, 216, 217positive displacement 29, 85, 115 compressors 63, 70, 81power loss 84power pistons 53pressure 8, 63, 64, 70, 94, 111 ratio control 65 ratio differential 26 relief valves 105preventive maintenance 115primary oil pump 46process control 73, 105proximity probes 100, 112pulsation bottle 27, 47, 157, 193,

195pulsation control 27, 48pulsation suppression devices 48

Qquick opening 105, 106

Rratio control 87, 89ratio of specific heat 10, 54, 68reciprocating 29, 131 compressor 31, 47, 54, 59, 70,

73, 76, 79, 120, 121, 126, 128, 129, 132, 133, 135, 136, 145, 146, 174, 183, 194, 206, 212

separable balanced opposed 52

valves 124, 211reciprocating engine 28reciprocating linear motion 33reciprocating throughput 107recycle 106, 108 control 73 valve 77, 86, 137, 138, 139, 140 control 76reduced temperature and pressure

10relative velocity 19required flow capacity 54required horsepower 60required HP 55resistance temperature detectors

112, 113resonance 19 avoidance 194 frequency 97, 98Reynolds Number 93rider rings 189rod 37, 70, 158, 165, 184, 187, 188,

189, 190 damage 121 diameter 33 load 32, 47, 155

Index 359

nut 359 packing 35 reversal 47 run-out 184, 185, 186, 189, 190,

191, 192 sag 189, 190, 191Roots blower 2Roots, Philander and Francis 2rotary lobes 26rotating motion 33rotating stall 19, 82rotometer 112rotor 60, 83, 98, 99 imbalance 101 lobe 59 shaft 100run-out 187, 192

Ssafety valve 105screw compressor 59, 73, 98, 125 control 59screws 57scrubber 27, 125sealing rings 35seals 59, 70, 84, 117second law of thermodynamics 11seismic 112 probes 100seizing pulsation bottles 48semi-enclosed 22sensors 111separable balanced opposed 116 compressor 52separable compressor 77separable reciprocating

compressors 98shaft 84, 99single-acting 37

site conditions 54slide valve 58, 59, 73smart transducers 113specific gas constant 8specific gravity 54, 94specific heat 63 ratio 12, 64, 70, 93specific volume 8spectrographic vibration plot 102speed 55, 64, 70, 98, 111speed control 59, 73split bearing design 46spring constant 100stage HP 55stagger angle 21stall 15, 17, 18, 19, 81 limit 82 line 82 margin 15, 82 rotating 19static pressure drop 195stators 83, 84steady state 82 operating 15steam engine 52steam turbine 52, 115sticktion 213, 215, 218, 219, 220stiff shaft 97, 98Stolze, Dr. Franz 3stone wall 15, 18stop plate 219stop surge algorithm 91stop surge line 95stroke 39, 55, 56, 192suction 27, 49, 106, 108, 117, 121 and discharge valves 199, 213 bottles 194, 197 port 58 pressure 33, 54, 68, 77, 79, 93


pulsation 47 ring 213 strainers 219 temperature 55, 68, 93 throttle valve 138 valve 39, 60, 213 guards 56 throttling 59, 60 unloaders 39, 41, 42, 73, 76,

77, 132, 133 plug type finger type 41supervisory control 131, 134, 140surge 15, 17, 26, 64, 65, 69, 81, 82,

85, 87, 91, 93, 112, 113, 117, 132, 138

avoidance algorithm 92 avoidance line 91, 95 control 107, 113 line 88, 90, 95 map 88, 89 limit 69, 70, 87 line 25, 64, 65, 66, 81, 88, 90, 95 point 68 margin 75 point 74 region 86 stop algorithm 92 stop line 90surging 83surve avoidance line 90synchronization gears 57synchronous motor 177system resistance 17, 63 curve 65

Ttail rod 162tangential component 17

Teflon rider rings 186temperature 8, 10, 63, 64, 67, 70,

111, 112tension load 33theoretical HP 55Thermal Expansion Factor 92thermocouples 112, 113thermodynamic analysis 115thermodynamic laws 24thermowells 113throughput 84 control 73thrust bearings 83tilt-pad bearing 123tilt-pad designs 47tilt-pad thrust bearings 59torque 36 requirements 98torsional 98 analysis 52 critical speed and response

analysis 52 natural frequencies 99 study 139total HP 55total volume 8transducers 112, 113transmitters 111turbine meter 112turbocharger 28turbulent flow 17

Uundamped critical speed 98universal gas constant 8unloaders 174unloading systems 146

Index 361

Vvalve 39, 42, 105, 212, 213, 219 ball-type check valves 29 ball valve 105, 106 butterfly valve 105, 106 cages 165, 216 caps 165 check valve 105, 106, 108 choke valve 106 covers 121 cover temperature 124 dynamics 212 failure 70 in bore 37 -in-head 157 life 212 lift 212, 216 out of bore 37 plate 42, 124 impact velocities 211 lift 211 type 211 pockets 151 volume bottles 158 ports 56 seats 150 sequence 77 trim 105, 106vanes 117variable control 75variable frequency drive 132variable guide vanes (VGVs) 25,

74

variable or fixed volume pockets 79

variable speed drive 60variable volume 174 pocket 126, 179, 180velocity 100, 101, 102 diagrams 22 triangles 19venturi meter 112vibration 111, 112, 122 signature 117volume booster 140volume bottles 39volume clearance pocket 155, 165,

176volume flow 98volume pockets 39, 44, 73volumetric clearance 39volumetric efficiency (Ev) 56volumetric flow 9, 64, 83 factor 93VVP 128

Wwater organ xi, 1, 2wear band 39weight flow of the gas 12whirl orbit 122work 11

ZZ, compressibility factor 55