if you have any questions or require additional information, please …...

^

pnv

Received

DEC 21 2016Air Permits Section

Enviva Pellets Northampton LLC874 Lebanon Church Road

Garysburg, NC 27381

+1(252)5412631fax (252) 5412632

www. envivabiomass. com

December 20, 2016Via FedExOvernight Delivery

Ms. Yukiko Puram

NCDEQ-DAQAir Permitting Section1641 Mail Service Center

Raleigh, NC 27699-1641

Re: Additional Information Request for Permit Application Submitted August 9, 2016(Application ID No. 6600167. 14B)Enviva Pellets Northampton, LLCGarysburg, North Carolina

Ms. Puram:

Enviva Pellets Northampton, LLC (Enviva) would like to provide the requested infonnation thatwas requested on November 29, 2016. Please find Enviva's response below:

1. Please submit additional documents to support the control efficiencies for the cyclonesCD-DC, CD-HM-CYC 11 8, and CD-CLR 1 through 6.

a. For the CD-DC, th 98. 5° refers to the cyclone and WESP's combined controlefficiencies. In Form C4 for Dryer, the manufacturer is listed Lundberg E-Tube115719, which is the WESP manufacturer.

b. For CD-HM-CYC 1 through 8, the control efficiency document is attached. TheES-HM 1 through 8 cyclones are in series with a process ending ba filterequipment, Aircon 16 RAB 412-10. The control efficiency of the cyclones is notincluded in the emission calculations.

c. For CD-CYC- 1 through 6, the control efficiency document is attached.2. Please update Table B-l of your application to include emissions from missing emission

sources. The sources that are missing from the table are: ES-DLB, ES-DLC-1, ES-BCS-2, ES-BCS-3, ES-BSB-1, and ES-BSB-2.

a. For ES-DLB, ES-DLC-1, ES-BSC-2, ES-BSC-3, ES-BSB-1, and ES-BSB-2, arerepresented in the emission calculations for ES-DWH in Table B-W: 13

b. All of the emission sources listed above are fugitive emissions from handling drywood.

c. From NCDEQ request for more information letter, ES-BCS-2 and ES-BCS-3were incorrectly named. They are ES-BSC-2 and ES-BSC-3 in the application,which are included in the ES-DWH emissions calculations.

in d ) i-P] ce;titte<f

..i. 'i. *ir>!>'>ri| iCSeyW INi''IATiV£

3. There are inconsistent source ID numbers throughout the application, with ES-DLH, DryLine Hopper and ES-DLB, Dry Line Feed Bin are the same source. ES-PFB-1, PelletFines Bin and ES-FB are the same source.

a. Attached are the edited forms and tables that will list ES-DLB, Dry Line Bin and ES-PFB-1, Pellet Fines Bin.

If you have any questions or require additional information, please contact me at 252-370-3181.

Sincerely,

M^

Joe Harrell

Corporate EHS Manager

ec: Royal Smith, Enviva Executive VP-Operations

Enviva LP | 7200 Wisconsin Ave., Suite 1000 | Bethesda, MD 20814 USA+1 (301) 657 5560 | fax (301) 657 5567 ) www. envivabiomass. com

mr^nrTic<>mr^<?>, o\

Si

i ^sss (^1 ^-, ^-|^^^|^, t.^(^(^CMOOCGOOOGOOÔ^'-i

0 0 OOOOOOOr^M-'*I.?

« I\o \o \Q oo f .^..^-.^..^..^..^. u-ir-r-F^r^c'. m >/S OGo6ooo6o6aoodmr<i's6 <o \£> o o in u-i u-> vS u-i in .^f' »n v; rt

'-< CT\

s

T?^"'^' ^ (N rr)rr)r^r<~]«im^^^»ninu-k Q ^ rrirrirTimrrir^^tnin

fa) _^ ^J ^ ^ o o ^r^^ ^-; ^-; ^' ^' m me;

\Q \C> \0 00 if r^M<NC^f^lMfnCT^<y\r~r-r- r^ V) ^'^'^-'^. ^.^^0000\0 \0 ^^ 0 0 \Dk<0\DÔ*^>^£>v-i'0\0

^" ^" '? 't <?' M t^.^t- -^- .^h Tf T .<T r^)

^ ^-; " ". o o ^ ^ ^^. ^:- ̂ - m m N

.

B^§§§ § i ^^^^^g^^^3oo S S !n><??nljr)!/?lnlr>ur><r>

.^ft;^;^^: ^ ^ triiîv-iiîu-iu-i^^vS

^gsg§gg isssssssssE 222 2 2 CT\o^CT*. CT*. CT*ioô;*i5*i5*i

§ *C 2 00° .<5 o ® o <?<? o5^5S1kol£lo l5 S5 <5'5>5SSc ^ ^ ^ t^- t^ t^. ?: r^t^ c^ ?:_t^ c^ r^ c^ (^

ôooooo so ao cooooooooooooooooo

li?§§§§ goooooogoS^g^000 0 0 000000300p^

fSSSs s§88§ssssst§§i s- s ^??S?^-;

^^^^^N cn,. ^^^^^,, ;^^

I

I

sill iB ssssslii686

^ssS3 uo

-r<r-iûi-<pW C?Cp

^sss^^^s'su ^ -. --

coco co co w 1^ r'l rSMM WW M M

p? oa

|S3s

^ .S.S ?- ^ ^

^Ê^^^^'^'S'SUU U U.S . £ .S&&&&'&'&'£

Source

Description

Dryer SystemEmergency Generator

Fire Water Pump

HammermiIls/Nuisance Dust System

Pellet Mill Feed SiloPellet Fines BinPellet Presses and Coolers

Finished Product Handling & LoadoutFinished Product Bagging ScreeningDried Wood HandlingDiesel Storage Tanks

Fugitive (Non-PSD Sources)Bark-HogChippingGreen Hammermills

Green Wood HandlingGreen Wood Piles

TABLE B-lFACILITV-WIDE CRITERIA POLLUTANT SUMMARY

ENVIVA PELLETS NORTHAMPTON

UnitID

ES-DRYERES-EGES-FWP

ES-HM-lthrug/ES-NDS(see note)

ES-PMFSES-PFB-1ES-CLR1 thru -6ES-FPH, PL1,2PB1-12ES-BSC-l, ES-BSS-1,2ES-DWH (see note), ES-PPTK1 & TK2

Total PSD Emissions

ES-BARKES-EPWCES-RCHIP-land2ES-GWH SES-GWSP1 5 ?

Total Facility Emissions:

co

(tpy)

60.950.500.43

NOx

(tpy)

125.500. 580. 49

TSP

(tpy)

29. 840. 030.02

20.27

PM-10 PM-2.5

(tpy) (*py)

CO;,, biomassS02 Total VOC ^ferral(tpy) (tpy) (tpy)

29.84 29.84 19.20 209.880.03 0.03 0.0010 0.00150.02 0.02 0.0008 0.0013

20.27 20.27

61.S8 126.57

61.88 126. 57

24.71

0.380. 54

38. 525. 33

33.790. 12

128. 84

0.032. 65

131.52

0. 380.54

35.054. 85

30.750. 06

121.79

0.011.33

123.13

0.380.54

21. 192. 9318.580.01

93.79

0.000.2093.99

19.20

19.20

142.86

9. 10E-04

377.46

0.301.251.25

2.93382.89

3,341. 4393. 3580.02

co,,(tpy)

162, 118.8393. 3580. 02

3,514. 80 162,292. 20

3 514.80 162 292.20

Note: DWH (Dried wood handling) includes several miscellaneous dried wood transfer sources as detailed in Table B-14 (including ES-DLB, ES-BSC-1, ES-BSC-2, ES-BSB-1, and ES-BSB-2).Note: NDS represents the Nuisance Dust System source which includes the transfer of materials from the Dry Line Conveyor (ES-DLC-1).

Page 1 of 1Facility Totals

161219 ENV NOR T5 App Emiss Calcs Rev2

TABLE B-14DRIED WOOD HANDLING DROP POINTEMISSIONS

ENVIVA PELLETS NORTHAMPTON

Annual Dryer Output Throughput (ODT/yr) 537,625Maxium Diy Line Annual Throughput (ODT/yr) 87,600

D[yer Throughput Plus D^'-line Throughput (ODT/yr) 625,225Amount ofFines Diverted from Hammermills 15.0%

Annual Hararaermill Throughput (ODT/)T) 531,441Pellet Press ThrouBhput (ODT/yr) 625,225

Max Dryer Short-Term Throughput (ODT/hr) 71.710Diy-line Feed Throughput (ODT/hr) 10.000

Max Hammermill and Pellet Press Throughput (ODT/hr) 81.710Max Bagging System Throughput (ODT/hr) 60. 000

Dryer Output Moisture Content: ] 7%

Pellet Mill Output Moisture Content: 7%

Emission Source Group

ES-DWH

Description

Dryer Discharger to DryerCollection Conveyor Belt

Pre-screen Feeder Fines Overs toHammermills Infeed and

Distribution

Hammennills Cyclone DiverterGates lo Hammermills System

Discharge Collection Conveyor Belt

C.ntr»I "co"..rolDescription

Enclosed Auction to 2 mphmean wina si

Enclosed Auction to 2 mphmean wind si

Reduction to 2 mphmean wind speed

Throu h ut

Mail.

Hourl

th

86.40

Annual

t

647. 741

Potential Unconlralled

Emissions for PM'IIb/hr

3. 1E-03

Pntcntial Uncontrolled PotenUal Uncontrolltd

Emissions for PMio' Emissions for FM;/Ib/hr t

1. 2E-02 1. 3E. 03 5. 5E-03

Ib/hr

2.2E-04

t

8. 3E-U4

112, 992 5. 3E-04 2. 0E-03 2. 5E-04 9. 6E-04 3. 8E-05 1. 5E-04

640,291 3. 0E-03 1.2E-02 1.4E-03 5.4E-03 2.2E-04 8.2E-04

Hammermitls System DischargeCollection Conveyor Bdt'loPeTlel Enclosed Redu<;"°n >° 2 "'Ph

Mill Feed Silo tofeed Screw mean wind speed753. 283 3. 5E-03 2. 5E-U4 9. 7E-04

Drop Point for Diy Line Transferfrom Dry Line Bin to Dry Line

ConveyorEnclosed K"l"<="°"-° 2 "P"

mean wind st105, 542 4. 3E-04 1. 9E-03 2. 0E-04

ES-PP Drop Emissions from Pellet Presses c""|". ^ Reduction lo 2 mphto Pellet Press Collection Conveyois E"'-'°se° niean wind speed 672, 285 1. 1E-02 4. 2E-02 5. 2E-03 2. 0E-02 7. 8E.04 3. 0E-03

Drop Emissions from BaggingSystem Coneyors 10 BaggingSystem Enclosed I<°d"clioRK)2mph ^^

Bins"''"'"''"'""'" -"~~~ mem wind speed

.& \^^^cz

625, 225 1. 8E-02 5. 8E-04

TOTAL 3.0E-02 1.2E-OI 1.4E-U2 5. 7E^)2 2. 1E-03 8.7E-03Note:

Fugitive emissions are not included in facilily-vvide PTE because the Northampton Pellet Mill does not belong to one of the listed 28 source categories.2 Max hourly rates based upon maximum calculated throughput rates provided in mass balance provided by Mid-South Engineering Company, June 17, 2011; updated for 13% moislure content on December 29, 20113 Based emission factors calculated per AP-42 Section 13.2.4, September 2006.

where: E = emission factor (Ib/ton)0. 740.35

0.053

2.00

Dryer Exit Pellet Press Exit

k = particle size multiplier (dimension I ess) for PMk =particle size multiplier(dimensionless)forPMiok ̂ particle size multiplier (dimensionless) for PM^;

U = mean wind speed (mph) ^H 0.00

M - material moisture conlenl (%) 17 7EforFM(lb/lon)= 3. 6E-05 1. 2E-04

EforPMio(lb/ton)= 1.7E-05 5. 9E-05E for PM;.; (Ib/lon) = 2.6E-Of. 8. 9E-06

^^[^0^^)^'^^

(b/'ro^

t^jO 0/7<3 2. l!'/+- <^

0^1 \^1^

FOR B9EMISSION SOURCE (OTHER)

REVISED: 12/01/01 NCDENR/Division of Air Quality - Application for Air Permit to Construct/OperateEMISSION SOURCE DESCRIPTION: Dry Line Bin EMISSION SOURCE ID NO: ES-DLB

B9

OPERATING SCENARIO: OF

DESCRIBE IN DETAIL THE PROCESS (ATTACH FLOW DIAGRAM):Dried wood materials are transferred to Dry Line Conveyor (ES-DLC).

9^ -^

CONTROL DEVICE ID NO(S): N/A- Fugitive

EMISSION POINT (STACK) ID NO(S): N/A - Fugitive

MATERIALS ENTERING PROCESS - CONTINUOUS PROCESS

TYPE UNITS

Dried Wood Materials ODT

MAX. DESIGN

CAPACITY (UNIT/HR)

10tph

REQUESTED CAPACITY

LIMITATION(UNIT/HR)

MATERIALS ENTERING PROCESS - BATCH OPERATION

TYPE UNITS

MAX. DESIGN

CAPACITY (UNIT/BATCH)

REQUESTED CAPACITi'

LIMITATION (UNIT/BATCH)

MAXIMUM DESIGN (BATCHES / HOUR):

REQUESTED LIMITATION (BATCHES / HOUR):

FUEL USED: N/A

MAX. CAPACITY HOURLY FUEL USE: N/ACOMMENTS:

(BATCHES/YR):

TOTAL MAXIMUM FIRING RATE (MILLION BTU/HR): N/AREQUESTED CAPACIPi^ ANNUAL FUEL USE: N/A

Attach Additional Sheets as Necessary

FOR BSPECIFIC EMISSIONS SOURCE INFORMATION (REQUIRED FOR ALL SOURCES)

REVISED 12/01/01 NCDENR/Division of Air Quality . Application for Air Permit to Construct/OperateEMISSION SOURCE DESCRIPTION: Dry Line Feed Conveyor EMISSION SOURCE ID NO: IES-DLC

CONTROL DEVICE ID NO(S): CD-HM-BF-3OPERATING SCENARIO _1 OF 1 EMISSION POINT (STACK) ID NO(S): EP-2DESCRIBE IN DETAILTHE EMISSION SOURCE PROCESS (ATTACH FLOW DIAGRAM):Dried wood materials are transferred from the dry line feed hopper (ES-DLB) to the exisiting hammermill pre-screens in-feed conveyor.

Bin

B

TYPE OF EMISSION SOURCE (CHECK AND COMPLETE APPROPRIATE FORM B1-B9 ON THE FOLLOWING PAGES):d Coal, wood, oil, gas, other burner (Form B1) D Woodworking (Form B4) D Manufact. of chemicals/coatings/inks (Form B7)D Int.combustion engine/generator (Form B2) D Coating/finishing/printing (Form B5) Q Incineration (Form B8)D Liquid storage tanks (Form B3) Q Storage silos/bins (Form B6) Other (Form 89START CONSTRUCTION DATE: 2014 OPERATION DATE: 2014 DATE MANUFACTURED: 2014MANUFACTURER / MODEL NO. : Enviva Built EXPECTED OP. SCHEDULE: 24 HR/DAY 7 DAY/WK 52 WK/YRIS THIS SOURCE SUBJECT TO? NSPS (SUBPART?): NESHAP (SUBPART?):_ MACT (SUBPART?):PERCENTAGE ANNUAL THROUGHPUT (%): DEC-FEB 25 MAR-MAY 25 JUN-AUG 25 SEP-NOV 25EXPECTED ANNUAL HOURS OF OPERATIO^ 8,760 VISIBLE STACK EMISSIONS UNDER NORMAL OPERATION: <20% % OPACITY

CRITERIA AIR POLLUTANT EMISSIONS INFORMATION FOR THIS SOURCESOURCE 0 EXPECTED ACTUAL POTENTIAL EMSSIONS

EMISSION (AFTER CONTROLS/LIMITS) (BEFORE CONTROLS/LIMITS) (AFTER CONTROLS/LIMITS)AIR POLLUTANT EMIFTED FACTOR Ib/hr tons/yr Ib/hr tons/yr Ib/hr tons/yrPARTICULATE MATTER (PM) See Emission Calculations in Appendix BPARTICULATE MATTER<10 MICRONS (PM, o)PARTICULATE MATTER<2. 5 MICRONS (PM^. s)

SULFUR DIOXIDE (S02)NITROGEN OXIDES (NOx)CARBON MONOXIDE (CO)VOLATILE ORGANIC COMPOUNDS (VOC)LEAD

OTHER

HAZARDOUS AIR POLLUTANT EMISSIONS INFORMATION FOR THIS SOURCESOURCE 0 EXPECTED ACTUAL POTENTIAL EMSSIONSEMISSION (AFTER CONTROLS/LIMITS) (BEFORE CONTROLS/LIMITS) (AFTER CONTROLS / LII/ITS)

HAZARDOUS AIR POLLUTANT AND CAS NO. FACTOR Ib/hr tons/yr Ib/hr tons/yr Ib/hr tons/yrN/A

fi

AIR POLLUTANT AND GAS NO.

^sagT

SOURCE

*H3S-

Ib/hr Ib/day Ib r

Attachments: (1) emissions calculations and supporting documentation; (2) indicate all requested state and federal enforceable permit limits (e. g. hours of operation, emission rates) anddescribe how these are monitored and with what frequency; and (3) describe any monitoring devices, gauges, or test ports for this source.

COMPLETE THIS FORM AND COMPLETE AND A TTACH APPROPRIATE B1 THROUGH B9 FORM FOR EACH SOURCEAttach Additional Sheets As Necessary

B

SPECIFIC EMISSIONS SOURCE INFORMATION (REQUIRED FOR ALL SOURCES)REVISED 12/01/01 NCDENR/Division of Air Quality - Application for Air Permit to Construct/OperateEMISSION SOURCE DESCRIPTION: Dry Line Muppyi g^ EMISSION SOURCE ID NO: ES-DLB

CONTROL DEVICE ID NO(S): N/AOPERATING SCENARIO _1 OF 1 EMISSION POINT (STACK) ID NO(S): N/A-FugitiveDESCRIBE IN DETAILTHE EMISSION SOURCE PROCESS (ATTACH FLOW DIAGRAM):Dried wood materials aretransferred to the Dry Line Conveyor (ES-DLC).

B

TYPE OF EMISSION SOURCE (CHECK AND COMPLETE APPROPRIATE FORM B1-B9 ON THE FOLLOWING PAGES):D Coal, wood, oil, gas, other burner (Form B1) Q Woodworking (Form B4) d Manufact. of chemicals/coatings/inks (Form B7)D Int.combustion engine/generator (Form B2) Q Coating/finishing/printing (Form B5) d Incineration (Form B8)D Liquid storage tanks (Form B3) D Storage silos/bins (Form B6) Other (Form B9START CONSTRUCTION DATE: 2014 OPERATION DATE: 2014 DATE MANUFACTURED: 2014

MANUFACTURER / MODEL NO. : Enviva Built EXPECTED OP. SCHEDULE: 24 HR/DAY 7 DAY/WK 52 WKA'RIS THIS SOURCE SUBJECT TO? NSPS (SUBPART?): NESHAP (SUBPART?):_ MACT (SUBPART?):PERCENTAGE ANNUAL THROUGHPUT (%): DEC-FEB 25 MAR-MAY 25 JUN-AUG 25 SEP-NOV 25

EXPECTED ANNUAL HOURS OF OPERATIOh 8, 760 VISIBLE STACK EMISSIONS UNDER NORMAL OPERATION: <20% % OPACITYCRITERIA AIR POLLUTANT EMISSIONS INFORMATION FOR THIS SOURCE

SOURCE 0 EXPECTED ACTUAL POTENTIAL EMSSIONSEMISSION (AFTER CONTROLS/LIMITS) (BEFORE CONTROLS/LIMITS) (AFTER CONTROLS / LIMITS)

AIR POLLUTANT EMITTED FACTOR Ib/hr tons/yr Ib/hr tons/yr Ib/hr tons/yrPARTICULATE MATTER (PM) See Emission Calculations in Appendix BPARTICULATE MATTER<10 MICRONS (PM, o)

PARTICULATE MATTER<2. 5 MICRONS (PMzs)SULFUR DIOXIDE (S02)NITROGEN OXIDES (NOx)CARBON MONOXIDE (CO)VOLATILE ORGANIC COMPOUNDS (VOC)LEAD

OTHER

HAZARDOUS AIR POLLUTANT EMISSIONS INFORMATION FOR THIS SOURCESOURCE 0 EXPECTED ACTUAL POTENTIAL EMSSIONS

EMISSION (AFTER CONTROLS/LIMITS) (BEFORE CONTROLS / LIMITS) (AFTER CONTROLS/LIMITS)HAZARDOUS AIR POLLUTANT AND CAS NO. FACTOR Ib/hr tons/yr Ib/hr tons/yr Ib/hr tons/yrN/A

TOXIC AIR POLLUTANT EMISSIONS INFORMATION FOR THIS SOURCEINDICATE EXPECTED ACTUAL EMISSIONS AFTER CONTROLS / LIMITATIONS

TOXIC AIR POLLUTANT AND CAS NO. EF SOURCE Ib/hr Ib/dayN/A

Ib/yr

Attachments: (1) emissions calculations and supporting documentat'on; (2) indicate all requested state and federal enforceable permit limits (e. g. hours of operation, emission rates) anddescribe how these are monitored and with what frequency; and (3) describe any monitoring devices, gauges, or test ports for this source.

COMPLETE THIS FORM AND COMPLETE AND A TTACH APPROPRIATE B1 THROUGH B9 FORM FOR EACH SOURCEAttach Additional Sheets As Necessary

OPERATION, INSTALLATION, & MAINTENANCE MANUAL

for

Aircon HE (High-Efficiency)

Pellet Cooler Cyclone Units

^}^t

Aircon CorporationP.O. Box 80446

2873 Chelsea AvenueMemphis, TN 38108-0446Telephone: (901) 452-0230FAX: (901)452-0264

TABLE of CONTENTS

/4tMMt ffwfaWMUiSK,

Operating Principle .

Einissions

Operating InstructionsReceivingInspectionInstallation

Start-Up Checklist

Troubleshooting

Recommended Maintenance

Safety

Efficiency Chart

3

4

5

6

7

8

9

10

f4<WHt S«l^tOWCt01t

OPERATING PRINCIPLE

A.

B.

c.

D.

E.

Dust-laden air enters the cyclone at the inlet section. Typically, it is directlyfrom the pellet cooler.

As air is collected in the body section of the unit, the weight of this rotatingmass of air begins to drop into the cone section. The deceleration ofparticles sinking the inside surface of the cone section causes the heavierparticulate matter to drop from the air stream into the cone discharge.Hence, the cone section conti^butes most to the action of mechanicalseparation inside the cyclone.

The swirl of air in this cone section creates a low pressure area in the centerarea of the cyclone body. Consequently, the lighter (cleaner) air in thecenter of this vortex rises into the tube and into the plenum section.

A smaller cone called a "vortex cone" or "vortex breaker" sits directly underthe bottom of the main cone. Its purpose is to lower the final discharge ofthe cyclone below the profile of the vortex, so that material can be allowedto flow out of the unit without being pulled back up into the unit.

A pressure drop occurs as air swirls through the unit. Every rotation of airwithin the unit requires energy to cause it to constantly change direction.The average amount of pressure drop through the cyclone will typicallyrange from 5 inches to 7 inches [water gage]. Given different air velocitiesand temperatires, a chart in the Appendix is available for more preciseestimates.

There is a practical upper and lower design limit to the capacity of anycyclone unit or cyclone set. The best range is based on an inlet velocity ofbetween 3, 500 and 4, 000 feet per minute. However, the upper range can beextended to 4, 300 feet per mmute without any measurable loss of efficiency.

/46w»t ffwfaaWtiMt

EFFICIENCY

The efficiency of any cyclone is nearly proportional to the mean size of suspendedparticles entering the miit. Because an Aircon pellet cooler cyclone has a relativelylarge height-to-widfh ratio, it is the most efficient type of mechanical separator usedin the feed industry.

According to an EPA publication AP-42, a high-efGciency pellet cooler cyclone ofthis type will emit a maximum of 0. 15 Ibs. of PM-10 particulate per one ton ofmaterial entering the pellet cooler. {Source: EPA publication AP-42, Chapter 9(for the food and agricultural industries), at their website:h ://www.e a. ov/ttn/chiefa 42/ch09/fmal/c9s0909-l. df, page 24.}

For example, a 60-ton per hour counter-flow pellet cooler system will have thefollowing maximum emissions rate:

60 rocesstons x 0. 15 Ibs emissions =

hr process ton91bs/hr

The chart in the Appendix can help detemiine the capture percent efficiency of acorrectly sized pellet cooler cyclone for any one particular size of dust. A correctlysized cyclone typically will be sized for an inlet velocity of between 3,500 and 4, 300(achial) feet per minute.

The new EPA emission standard concerning manganese and chromium (40 CFR Part63, Subpart DDDDDDD) states that air pollution conta-ol equipment must achieve95% or greater (by weight) reduction in particulate matter (PMio) emissions. Thisefficiency is guaranteed by Aircon provided that the cyclones are adequately sizedfor the system. The ideal range design inlet range for any HE cyclone is from 3, 500to 4,000 fpm. However, a system can be adequately sized and still allow air velocityto be as high 4,300 feet per minute.

The lower recommended range for a cyclone of 3,500 feet per minute is based onmaintenance concerns and not necessarily on emission concerns. If the velocity in acyclone mlet falls below 3,500 Q)m, there may be buildup of material in the inletduct. However, this will not adversely affect emissions at the point of the cyclone.Nevertheless, it is possible that the lack of adequate ventilation on the pellet coolerexhaust may allow the unvented emissions to show up somewhere else in theprocess.

OPERATING INSTRUCTIONS

RECEDING

Depending upon the request of the installer, the cyclone may be shipped in severalsections. Therefore, a quick inspection should be perfonned on each section fordamage that may have occurred m ti-ansit. Also, both the quantity and quality of anyparts that may have been shipped loosely should be checked. Boxes containing theseparts should be mspected for signs of improper handling that may have causeddamage. Any missing or damaged parts should be noted with the shipper beforeaccqiting the shipment. Aircon is not responsible for any damage that occurs duringshipping. The purchaser should bring all damage claims against the carrier.

INSPECTION

Upon accepting the shipment a closer inspection of the cyclone is necessary. Careshould be taken to thoroughly inspect each section of the cyclone for dents or cracks.Aircon should be notified of any inconsistencies between the unit and any certifieddrawings containing Aircon specifications. No changes should be made without theconsent ofAircon.

INSTALLATION

Most pellet cooler cyclones are installed inside a mill. The available space willdetennine how the cyclones and related ductwork are installed.

y4iWMi. ̂ l»^faw<ttK>f»

START-UP CHECKLIST

A. Unit body sections, supports, and compressed air piping secured with allbolts adequately tightened.

B. Any unused optional or auxiliary NPT connections plugged and sealedairtight.

C. Clean-out door secured.

D. Any access doors in place and properly secured.

E. At the conclusion of an operating period, turn off the process equipment andother related process equipment twenty (20) minutes before the system fan.This routine will allow the system to be purged after each use. Remember todischarge all related auxiliary equipment, such as the screw conveyor androtary airlock.

^inwtt ^wiftowdait

TROUBLESHOOTING

A. Observation: Visible dust leakage (not steam)

Problem: Fan oversized, too much airflow, or daniper open too much.

B. Observation: Flow rate of air fb-ough system too low

Problem: System blower or fan (fan undersized; fan running backwards;fan belt slippage)

System blockage (blockage m duct leading up to filter)

/46MM ̂ 1»^&W<US5»(»

RECOMMENDED MAINTENANCE

INSPECTION

Daily

A. Check exhaust from the system fan to make sure there are no visible dustemissions.

B. Check solid discharge from the cyclones.C. Check to see if there if there is any visible water (condensate) leaking from

the ductwork.

D. Check system fan amperage draw, calculated brake horsepower, andcompare to estimated brake horsepower.

Quarterly (every three months)

A. Remove any access doors (if equipped wifh them) to observe if there is anydust accumulation in the plenum.

B,, Do a thorough examination of the cyclones and ductwork to see if there areany air leaks, mst, or corrosion. In some cases, insulation may need to beremoved to check affected areas.

SAFETY

Before removing any access doors, please observe the following safety precautions:

A. Turn off the system fan or blower and lock out all electiical disconnects for allassociated and auxiliary equipment.

B. Depending upon the size of the access door, two operators may be required toremove it. Access doors are to be removed completely.

C. Do not enter a cyclone without a confined space pennit or without observingrequired safety protocol at the site.

CDRPDRATIDNP. O. BOX 844B

MEMPHIS, TENNESSEE 3B108(901 ) 452-0230

EFFICIENCY CHART FORHIGH EFFICIENCY CYCLONES

100

90

sw^ 70^ BO

50g

40

i 308SS 20

10

g

EFFICIENCY CHART POINTS if

«2 «3 #4 #5 «B

«\

EFFICIENCY e MICRON (SIZE OF PARTICLE)«l«2«3#4#5«6

90%e9B'/.eaoy.e99, 5%

101520e 25

99, 85% e 3099. 9Z e 35

AIRCON MODEL

^

0. ____. _5____ _/0. ____.., 5 20 25 30PARTICLE SIZE IN MIWDNS

PRESSURE DROP CHART

35 40 p

INLET AS TBPBUTURE.20?* F

- 7^v73»' F

400' FBOO' F

10

8

COLLECTOR

0_ /_ 2 345578FPH x 1000 Z7F INLET VELOCITY

SUITABLE IMPLEMENTATION OF AIRCDN H-STYLE COLLECTORS CAN ACHIEVE UP TO 99. 97.COLLECTOR EFFICIENCY ON PARTICLES GREATER THAN 34 MICRONS, PARTICLES SMALLERTHAN 34 HICRDNS CAN BE COLLECTED WITH EFFICIENCIES RANGING FROM 90°,. JO 99, 85°,..

//;"' NPTDELUGEHEAD t//SS PL J

63- CL TO CL

?5 X 125}

6

^J 16 1/2 L-

PLAN VIEW

NOTE: CAP HAY BEROTATED TO ADJUSTOUTLET TO MATCHFAN ALIGNMENT

30" x 30" PRESSUREPELIEV VENT OPENINGVI TH PVC BOLTS SHOLDING CHAIN

25

25

r>-R ^

.^

^

5 I

I I I

32"

^

54

?1^

BINDICATORPLATE FARSIDE

8" ACCESSDOORFAR SIDE

22

0

30

^

^<\ [n

K <;s

L;,JELEVATION VIEV

^ Corrected fop view to shop exp vent-

ND. REVISION - DESCRIPTION

2/13/12 JT

DATE BY

'̂^lCC^iCORPORATION

P. O. BOX 80446MEMPHIS. TN 38108-0446PHONE;(001) 452-0230FAX: [901) 452-0264

DRAWN

dp/J^SCALE

3/16" = r- iDATE

E-MAIL:doron @oirv con-cor oration. corn

PROJ. NO.

2012-5620

DWG. NO.

5620-HE54

This droving is the property of Aircon Corporofion. If is Mf to be prifite^ phofogroph^, copied,Sooned of used fo the disadvontoge of the OVMF. All rights of design and ifivenfionore reserved

;/?. COUPLIH; »;7BPLATE R» PBCUtTmfSVIHdERpumisa

CfCLOIC LIVED Wm1/4- w n. m w

TW tZ) «Et<£F >E175ie' x 4(1' vimcHimHCPW BOLTS

^kl/Z' EOfLIW VS THPLATE FOR PBBmrm? siviMtLBiPAINT ISO

PLAN VIEV

iyie

I/^

r\^l L/

a- OM.amn

BOLTEIItllCCOfECTIOI

«GL;£F VBITS[SSCETAIi)

v-o-

L _ _ JBCtTEBOWKTIOH

SOL7E7COWCCTION

BIWlmCR PLATC*f/7H / t/4' COW. INBIFAR SIX1

NOTE: 1) TWO (2] SPRINKLER TAPS DELUGEPER CYCLONE 1/2" FULL COUPLINGWITH PLATE AND PENDANT SPRINKLER(DO NOT USE UPRIGHT) PAINT RED

REVISION - DESCRIPTION DATE BYA

ELEVATION VIEV

^^ -PHONE :[fl01 ) 452-0230FAX: [BOD 452-02B4

E-MAIL ; Johnaaalrcon-corpora+lon.com f/2'= l'-0'A

WW1!

AC96 CYCLONE

Bf! *nfi» iiHifnpirtf il jlfnu ftiyiflto. 11 h »f fa t> prlsM, ^trfiy^taf, <vMhid irmth fli Onhify if UIT. <H rl)»fa rf Mf, arf f.mtfn m niind

SK-SWOw . .

ACX-enratf

Aircon AC Cyclone Efficiency Chart

100%

90%

80%

70%

^ 60%.N£ 50%PL)^ 40%

30%

20%

10%

0%

0 10 20 30 40 50 60 70 80 90 100Particle size in Microns

Puram, Yukiko

From:

Sent:To:Cc:Subject:

Joe Harrell <[email protected]>Friday, June 30, 2017 3:44 PMPuram, YukikoVoelker, Joseph; Wike, Will; Christopher Seifert; Michael H CarbonRE: Enviva Northampton WESP and Compliance Assurance Monitoring (CAM) Plan

Hi Yuki,

Thanks for preparing the summary of our recent discussions I found it very helpful. As a follow-up to your questionsplease find the below responses and associated attachment for your review.

Best regards, 'Joe

envtva'Joe Harrell


Enviva Pellets Ahoskie, LLC142 NC Route 561 EastAhoskie, NC 27910 USAwww.envivabiomass.com

+1 (252) 209 6032 x(2202)cell (252) 370 3181fax(252) 364 3428"oe. harrelt envivabiomass. com

From: Puram, Yukiko [mailto:[email protected]]Sent: Tuesday, June 13, 2017 9:33 AMTo: Joe Harrell <joe. harrell@envivabiomass. com>Cc: Voelker, Joseph <[email protected]>; Wike, Will <[email protected]>; Christopher Seifert<Christopher. Seifert@envivabiomass. com>Subject: RE: Enviva Northampton WESP and Compliance Assurance Monitoring (CAM) Plan

Hi Joe,

Do you think three weeks be sufficient? How about by July 5th?

Yuki

Yuki PuramEnvironmental EngineerDivision of Air QualityNorth Carolina Department of Environmental Quality

9197078470 officeyuki. Duram@ncdenr. gov

217 West Jones Street1641 Mail Service CenterRaleigh, NC 27699

-"^Nothing Compares .-^---

Email correspondence to and from this address is subject to theNorth Carolina Public Records Law and may be disclosed to third parties.

From: Joe Harrell mailto:'oe.harrell envivabiomass.com]Sent: Tuesday, June 13, 2017 9:29 AMTo: Puram, Yuktko <vuki. Duram(a)ncdenr.gov>Cc: Voelker, Joseph <'ose h.voelker ncdenr. ov>; Wike, Will <will.wike ncdenr. ov>; Christopher Seifert<Christo her. Seifert envivabiomass. com>

Subject: RE: Enviva Northampton WESP and Compliance Assurance Monitoring (CAM) Plan

Hi Yuki,

When wilt you need this by? I'm speaking with my support team on Friday, so we can address your comments andquestions below.

Thank you,Joe

envivaJoe Harrell



+1(252) 209 6032 x(2202)cell (252) 370 3181fax(252) 364 3428"oe.harrell envivabiomass.com

From: Puram, Yukiko fmailto:[email protected]]Sent: Monday, June 12, 2017 2:43 PMTo: Joe Harrelt <'oe. harrell envivabiomass. com>

Cc: Voelker, Joseph <"ose h.voelker ncdenr. ov>; Wike, Will <will.wike ncdenr. ov>Subject: Enviva Northampton WESP and Compliance Assurance Monitoring (CAM) Plan

Hi Joe,

So here's a summary of what we discussed...

Based on the stack test conducted on October 3, 2013, the dryer was controlled by the WESP with 1000 amp ofsecondary current and 62 kV of secondary voltage. Even though the margin of compliance from the stack test result waspretty large (3. 07 Ib/hr for the 48. 0 Ib/hr limit), I'm not sure how you can demonstrate the compliance with the 2D .0515standard at 200 amp and 20kV, which is drastically different than what it was tested. I don't think you can use theemission factor from the stack test (3. 07 Ib/hr) unless you operate the WESP with the same parameters.

Enviva has performed a detailed review of the October 3, 2013 stack test results and vendor performance specificationsof the WESP and have concluded that the emissions from the dryer comply with the process weight rate rule particulatematter (PM) emissions limit of 2D .0515 without use of the WESP. Therefore, CAM does not apply and the proposedoperating parameters above are not required. Please see below for a detailed discussion of the emissions review anddocumentation of CAM applicability.

Based on the vender's letter dated on May 10, 2017, the proposed minimum voltage was 20kV. The equation that wasused for the calculation was based on the input rate of 54. 91 Ib/hrl am a little concerned about this input rate as thebefore control emission rate was never tested during the stack testing. The permit application indicated that the beforecontrol emission rate was 150 Ib/hr, which is not consistent with the vendor's letter. The equation that the vender usedto calculate the PM emissions from the WESP is clearly dependent of the input rate. Therefore, I cannot approve to usethe vender's calculation unless I can confirm the validity of the input rate.

Based on the October 3, 2013 stack test results and vendor performance specifications for WESP control efficiency, theinlet PM emission rate to the WESP during the test is calculated to be 35.47 Ib/hr. Therefore, as stated above, CAM doesnot apply and the WESP is not required to comply with the 2D.0515 standard. Please see below for a detailed discussionof the emissions review and documentation of CAM applicability.

Another issue that I'm concerned is the applicability of CAM. The following are the three criteria to be subject to CAM:1. be subject to an emission limitation or standard, and2. use a control device to achieve compliance, and3. have potential pre-control emissions that exceed 100 tpy.

It's pretty clear that the dryer satisfies the first two conditions. It appears to me that the dryer also meets the thirdcriteria. Based on the application, the befor e control emission rate was 150 Ib/hr, which is equivalent to 657 tpy. If weuse the input rate indicated on vender's letter (54. 92 Ib/hr), it would be 236 tpy. Either way, the pre-control emissionsexceeds 100 tpy. Therefore, the dryer is subject to CAM.

As discussed above, Enviva reviewed the October 3, 2013 stack test results and vendor specifications ofWESP controlefficiency to determine the WESP inlet PM emission rate (uncontrolled drver emission rate). The October 3, 2013 stacktest resulted in an outlet WESP PM emission rate of 3.07 Ib/hr. Based on the vendor specifications of the WESP andactual flow rate measured during the test (109, 700 acfm) the control efficiency corresponding to the test was91. 35%. The resulting uncontrolled dryer PM emission rate during the test was calculated to be 35. 47 Ib/hr.

As stated by Lundberg in the May 10, 2017 letter, WESP control efficiency is based on WESP collection area, effectivemigration velocity, and actual gas flow rate. Page 2 of the attached document includes the control efficiency of theWESP based on the vendor's calculation.

Enviva also evaluated potential uncontrolled PM emissions from the dryer at the maximum design throughput rate of71. 71 odt/hr. The process throughput during the October 3, 2013 stack test was 60 odt/hr. The resulting uncontrolleddryer PM emission factor is 0. 59 Ib/odt (35. 47 Ib/hr divided by 60 odt/hr). This factor multiplied by the maximum designthroughput rate of 71. 71 odt/hr results in an uncontrolled dryer PM emission rate of 42.40 Ib/hr, which is less than thePM limit calculated in accordance with 2D .0515. Therefore, condition 2 above is not met and the dryer is not subject to

CAM. Page 1 of the attached document includes the process throughput information, PM emission limit, and potentialPM emissions.

We then need to determine if the WESP is considered to be a large PSEU. If the potential post control PM10 emission islarger than 100 tpy, the WESP will be a large PSEU. Even though the stack test from 2013 shows PM emission factor was1.54 Ib/hr, which is equivalent of 6.74 tpy, this emission factor is not relevant unless you want to operate the WESP atthe same parameters used at the stack test. If we use the emission rate of 42 Ib/hr at 20 kV from the vender's letter, theemission would be about 42.00 Ib/hr, which is equivalent to 183.96 tpy, and would make it as a large PSEU.

As documented above and in the attached, CAM is not applicable.

The difference between small and large PSEU is the deadline for submitting a CAM plan. If the WESP is a small PSEU, wecan wait to for the next. If it's large, we have to address CAM for this permit.

It is up to you which minimum voltage and current you want to use, but it will also affect CAM depending on how youoperate the WESP. Based on the information I have (and I don't have), I think we need to include a test requirement onthe permit. I think we need to test for emissions at 20 kV and 200 amp to show the compliance. We need to measurethe before and after the control so that we know the control efficiency of the WESP (this was on the permit when thetesting was conducted, but they did not test for the efficiency for some reason.) Also, we need to incorporate CAM atthis time as I think this would be considered as a large PSEU. For more information regarding CAM, please review thispage:htt s: ncdenr.s3.amazonaws.com s3fs- ubtic Air%20 ualit enf cam CAM Overview, df

As documented above and in the attached, CAM is not applicable. Enviva does not believe testing of the WESP inlet andoutlet for control efficiency is warranted. As stated above and in the May 10, 2017 letter provided by Lundberg, WESPcontrol efficiency is dependent on WESP collection area, effective migration velocity, and gas flow rate. The WESPcollection area is fixed, effective migration velocity is proportional to the secondary voltage, and gas flow rate wasmeasured during the October 3, 2013 stack test. Therefore, alt values required to calculate WESP control efficiency areknown and WESP inlet testing should not be required.

Please fill out this Form E6 to submit a CAM plan.htt s: ncdenr. s3. amazonaws. com s3fs- ublic Air%20

If you have further questions, please let me know.

Thanks.

Yuki Puram

ualit ermits files 2016 Permit A lication Forms E6. df


9197078470 officeyuki. Duram@ncdenr. gov


a <>^htothin9 Compare&^^


Disclaimer

The information contained in this communication from the sender is confidential. It is intended solely for use by the recipient andothers authorized to receive it. If you are not the recipient, you are hereby notified that any disclosure, copying, distribution ortaking action in relation of the contents of this information is strictly prohibited and may be unlawful.

This email has-been scanned for viruses and malware, and may have been automatically archived by Mimecast Ltd, an innovator inSoftware as a Service (SaaS) for business. Providing a safer and more useful place for your human generated data. Specializing in;Security, archiving and compliance. To find out more Click Here.

Disclaimer


This email has been scanned for viruses and malware, and may have been automatically archived by Mimecast Ltd, an innovator inSoftware as a Service (SaaS) for business. Providing a safer and more useful place for your human generated data. Specializing in;Security, archiving and compliance. To find out more Click Here.

Puram, Yukiko

From:

Sent:To:Cc:Subject:

Joe Harrell <[email protected]>Tuesday, July 11, 2017 7:47 AMPu ram, Yukiko

Willets, William; Voelker, Joseph; Christopher Seifert; Roland Burnett; Heath LucyRE: Enviva Northampton First Time Title V Permit

Hi Yuki,

We agree with the department that stack test needs to be performed, which will be a condition of the Title V air permit.

Thank you,Joe

enviva'Joe Harrell


Enviva Pellets Ahoskie, LLC142 NC Route 561 EastAhoskie, NC 27910 USA

www. envivabiomass. com

+1 (252) 209 6032 x(2202)cell (252) 370 3181fax (252) 364 3428'oe. harrell envivabiomass. com

From: Puram, Yukiko [mailto:[email protected]]Sent: Friday, July 07, 2017 2:29 PMTo: Joe Harrell <joe. harrell@envivabiomass. com>

Cc: Willets, William <william. willets@ncdenr. gov>; Voelker, Joseph <joseph. voelker@ncdenr. gov>Subject: Enviva Northampton First Time Title V Permit

Hi Joe,

Hope you had a great 4th of July!

I've discussed the issues that were raised in the draft Enviva Northampton First Time Title V permit with my supervisors.We came to a conclusion that it is necessary to request a stack test in order to establish the WESP operationparameters. Since this is a First Time Title V permit, it is critical to ensure your operation is in compliance with all theapplicable regulations. Although theoretical calculations are useful designing pollution controls, they do notdemonstrate compliance with actual emissions. The only way to ensure compliance is through a stack test in this case.

Unfortunately, the stack test that was conducted in October 2013 does not represent Enviva Northampton's currentoperation since the voltage of the WESP during the testing was much higher than your typical operations. Other than

1

the stack test, we don't have any emission data we can rely on at the voltage that you are requesting. Therefore, I amgoing to insert a performance test requirement in the permit. I recommend testing the emissions with various operationScenarios in order to have flexibility in the WESP operations. Some examples of alternative scenarios could be operatingwith one or two fields only or operating at different voltage/current values. The more you can demonstrate compliancein different ways, the more flexibility we can give with operation parameters. In addition, these data can be used forother Enviva facilities. I believe additional stack test data would be beneficial to your facilities in order to maximize thecontrol efficiency.

You stated that Enviva Northapmton is not subject to CAM because the dryer emissions will be able to meet the 02D. 0515 PM emissions standard without a control. However, you still need to demonstrate the inapplicability of CAMthrough a stack test. In addition to the 02D .0515 standard, you also need to show that the dryer can meet the opacitystandard without a control in order to avoid the CAM applicability. Otherwise, DAQ will assume the WESP is subject toCAM. After you conduct a stack test to measure the outlet with certain WESP parameters, we can then determinewhether the WESP is either large or small PSEU. If it's a small PSEU (post control emissions are less than 100 tpy), thenwe can address CAM during the next permit cycle. If it's a large PSEU, then the permit needs to be modified to includeCAM.

I hope this explained why we need to include a stack test condition in the permit. I will send you a copy of the draft assoon as I update it. If you need further questions or concerns, please contact me or William Willets at 919-707-8726.

Thank you. Have a great weekend.

Yuki


9197078470 officevuki. [email protected]


-^"Nothing Compares


Disclaimer



env»v3Enirlva Peilats Nortliampton LLC

874 fauidti Chinch l»wdowyrtHiig. NcaTaw

*i<aaa)Mi.2«ufax (2M) Ml 2832

www.enviv@t?Ioma^<;wn

June 30, 2017

WillWikeNCDEQ-DAQ-Compliance OfficerRaleigh Regional Office217 West Jones St.Raleigh, NC 27603

RE: Request for Extension of Deadline for Emission Testing, Enviva Pellets Northampton, ILC (Permit #10386R05)

Dear Mr. Wike:

Enviva Pellets Northampton (the "Facility") filed a request to process up to 45% softwood through the dryhammermill (ES-HM-1 through 8) dated December 21, 2016. DAQ approved the request with emission testing in180 days, which the facility received on January 19. 2017. In the last six months the facility has been unable toprocess above 20% pine through the system and meet our customer's requirements. At this time, we request toperform emissions testing, when we achieve 45% softwood instead of being subject to a specific due date, becauseat this time we are unsure when we will achieve 45% softwood. I've attached the facility 12 month rolling averagefor your review.

Once we achieve the 45% softwood content through the dry hammermills the facility will submit an emissionstestine protocol immediately. When the emission testing protocol has been approved by NCDEQ-DAQ, the facilitywill perform emissions testing to establish VOC emission factors for the dry hammermills at the 45% softwoodcontent.

Please contact me if you have any questions regarding this matter, and thank you for your consideration of thisrequest.

S' cerely

Harrelt

Environmental, Health, and Safety Manager

Cc: Chris Seifert, EHS Director

Yukiko Puram, NCDEQ-DAQ, air permit section

^U^tfif. ^J' l-'^-t

MmillrWhl-IS

-ISIS

Qct-U

NOV. 1S15

itn-iaWMtMw-16

-Itu

*m-MM. 16

»

M

Oct-UNiw-ItD«c-t6

l»irl?

B*-UM»M7

.u17

laimMinontbsHmontht

ToflSWBDT

790B. 1789M. 27

(N6. 1X4M6A363S7. 688339J9

nai-w«?0.»

71U.Uno^-w7S19SSituao. 60I03W. EO

ueirn O.E8eil sa

fsaiM.1

S 3»

wisa

1M2S. 6S

»^47.W

13I. SOA1

Dry SbttVftiesBBT

3 ; 3U

asM.ya1669. BISOSS5

tS5159sosa. ss

: i

»ia.703 -KlSfffatste?.®

.vs

MS .11944na;n

fSIIM»n. iz

iiisz.WMW. tO7t 95

«»Mt&Sl

.as

isn.iB

27,948. 13O. SEIAS

0

SWWl

on

<19»S<2S9H.S7SKsasSSUM<a<&o9

4i2Sl.CT*sw .42St<0

»as3*aiAMaesiKtsOOSISS

<]sis.nI0»4^»SS33S1

fSStMWfcSli93S.ir9SU-0ifttus1SW37wnsmsssa

67. SW.87

wyi.w

wwwn BntOn

40B6,WOW249wwasvsasfM40811151fWsfai.WW6.W4UOM8Minsi411S2. SS4 ., OTa »»

l.OS.ttast. so

l!!S2»,atBSMM

4»W(

I

4aw

aMi«.i»

4M, «7^288B.951.98

BOTTdUl

siMB.mSOTOS.U

-li&U4UUUIISOSSIM48<)7U»W62Sil»s.sas-w4U51;asato&a

4BU6Ma<B6«.tS

30.teaiisfl<7WM5

saaui47S123?

wesm4431UJS

4856M3(Bis.n4»WW

570,504.12820/tei. 4l

ToidB Dr<»an

4w»<awmsseweaeaa

WW.TIwew.a

4aiioa.m4K33flS45SCUSS1BV.U<anaAa46l78S»44ni9.«4M1MS

MMSS. wsaM

MSS1.

socai.stfssnaf

WsaastiiaOBjM

tl

43S01iHU137.W

M25SSJ9m^SAS

%fHBiyw

cmailtli11»

tttt10Xim11Naw,1BK»

ftK

atISK

15KUtt

ux«<UKw

»»

4*19KfflK

Bfyw

WC.MBMhly

T»n>18

.s

.. t916iaw

1B

It»u

tl

. 1»t7M

»

18a>r

t

ft-I?

; 18

OHM WCMwtflfy

TOM2

1

1

a

2

2

2

2

a

i

.I2

.t

ita

1

:tI

2

4

I

'.21

Coohr

vocMamW

Tom

It

11u-9

111211:u

uu

II11M

u

ti)ttu

tl

Itw

tlit

in

TntilVOC.

W UNl

Toht

MSMS

sta

S62Iffi'

»7dsra3Wm

s»

sati»MBSf6WB301»l

art1Wa

anM9

co

Torn

.s6

t

;s;-s6

s

5

?

..?«

's's

(

i'ss

6

iSs

s

s

s

.s

eMtiiffingUn

torna

62

6iStSt»

6a»3«

,Mss.65sstS

M .*sS6

:6S:

65-»M

63

63

0 Mt

tlBtMao

20:51

19J»1S5Sffi 17. 89

t?.(Bises

7.rauifr

B. 88

OW

2B. 51S

i eo19.tM3I

t9.

^5

lae

D rMC

RaBian

191S4SQitt

. 19MI&39

193t

1949iaa»

UfS1SW18.77

18.7*u,UM

ia.9419.U

19J2T19JS

AS71A»Wfl20.U

»AZ

HW

Punhinnt

32ioo.oa

34SM.WSt W54 . Ot>aegss. Bo

3SIM.OOSOOS0. 41

M1U30

717BS. 65oaiToM M«s*.m

2

ai-

3rt59. (W .

oaa».7; ,<M35.950600AO»

IMtS. lil)

aao?.u

3MSS. 70

M», M».m

log HWMC

a

44.00M. GB

44.0044AI

44.COu.oo44.0BMM

-MiOB44. (K>

44

4400

JOO

4«.BO

44

44.

44 JOG

44

44. 00

44.0044

44. 00

sw

Puithaswt

ufssm

i»».miuaw.mison.!»». tl8

34S«.CO

re t.S&SS. TQ

ssaa. 771

.903098.

SSMM

wit,n

lOfr .684S9<.?0

.Ott

ISW

22033. 7B

tl, 33S. S?

togowMC

w

w.44.00*t.(m'

usa44.0«44. 00

"44JtO

MM

ww44.TO

4fl.OO

M

44.00

MJWl«.flfl44. 0»

4dOB

' JOQ44 JM44 OS

14jtW

*t.l

I HW W

«

26. 8SSI<W,®)(

»WtodW(6.8»!»

a.-yni3,<W(

ll.Ottit

ll.nis,z.

O.ITWMWir.w^as.w^fi&a<

B-SSS

ii&.ggflt»^»ffjf^u

.sjeid^

ws/sA;29<

W41»

14.10

IBiV«-SW)C»«nt'Ul ll«iiah*it. tStBK(lt<M/f»siS<Wlunil<UMonthA«g. lt.7ttl

'v^ss^^tjsstes^sssfM^ ^WSRi»u»^wa<rtHt<«Nt^M«a(^,^

Otfktol Sttrt OttB- Annl 22, 3013TnulTOCUmK *S3 tpy

Fmhiion ftomslOKPIne

Ib/OOTVOClOFVBFfc/OtSTOO; Dry*r

MX Pin*

b/ODTVOC: DtyrrIb/ODTVOC DHM

<S»(P«ieIb/ODTVOC: OHMIb/ODTVQCaR

ft» OtUhBf 18. »ia nai* usi results0.3 Oadxr »a. B)13 stack tut naulb

0.781 Ahcsklt Stack Test Jum MM, u»d t» tilt .bmn tatolaltensOA93 Ahoskh Stack Ttst Junt MM. liml In tin >binBcitaili(k>n>

0.10 <SX soltwwi) content thrnugbthc DHMs »5 of Bmiuv 19, 2017. rwt used rn ttw tbom nfculnlniu, will uit imce aclitew above 30K sofnmml cmttirt0.457 AhosUc Stack Test June 2014. used In the abom calailatiom

Puram, Yukiko

From:

Sent:To:Cc:Subject:

Joe Harrell <joe. harrell@envivabiomass. com>Friday, June 30, 2017 3:44 PMPuram, Yukiko

Voelker, Joseph; Wike, Will; Christopher Seifert; Michael H CarbonRE: Enviva Northampton WESP and Compliance Assurance Monitoring (CAM) Plan

Hi Yuki,

Thanks for preparing the summary of our recent discussions I found it very helpful. As a follow-up to your questionsplease find the below responses and associated attachment for your review.

Best regards,Joe

enviva'Joe Harrelt


Enviva Pellets Ahoskie, LLC142 NC Route 561 EastAhoskie, NC 27910 USAwww. enviyabiomass. com

+1 (252) 209 6032 x(2202)cell (252) 370 3181fax (252) 364 3428"oe. harrell envivabiomass. com

From: Puram, Yukiko [mailto:[email protected]]Sent: Tuesday, June 13, 2017 9:33 AMTo: Joe Harrell <joe. harrell@envivabiomass. com>

Cc: Voelker, Joseph <joseph. voelker@ncdenr. gov>; Wike, Will <will.wike@ncdenr. gov>; Christopher Seifert<Christopher. Seifert@envivabiomass. com>


Hi Joe,

Do you think three weeks be sufficient? How about by July 5th?

Yuki

Yuki Pu ramEnvironmental EngineerDivision of Air QualityNorth Carolina Department of Environmental Quality

9197078470 officevuki. Duram(ia)ncdenr.gQy


t -^>fcNothing Compares.


From: Joe Harrell mailto:'oe. harrell envivabiomass. com]Sent: Tuesday, June 13, 2017 9:29 AMTo: Puram, Yukiko <yuki. puram@ncdenr. gov>

Cc: Voelker, Joseph <'ose h.voelker ncdenr. ov>; Wike, Will <will.wike ncdenr. ov>; Christopher Seifert<Christo her.Seifert envivabiomass.com>


Hi Yuki,When will you need this by? I'm speaking with my support team on Friday, so we can address your comments andquestions below.

Thank you,Joe

envtva'Joe Harrell



+1 (252) 209 6032 x(2202)cell (252) 370 3181fax (252) 364 3428"oe.harrell envivabiomass.com

From: Puram, Yukiko mailto: uki. uram ncdenr. ov]Sent: Monday, June 12, 2017 2:43 PMTo: Joe Harrell <'oe. harrell envivabiomass. com>

Cc: Voelker, Joseph <"ose h.voelker ncdenr. ov>; Wike, Will <will.wike ncdenr. ov>Subject: Enviva Northampton WESP and Compliance Assurance Monitoring (CAM) Plan

Hi Joe,

So here's a summary of what we discussed...

Based on the stack test conducted on October 3, 2013, the dryer was controlled by the WESP with 1000 amp ofsecondary current and 62 kV of secondary voltage. Even though the margin of compliance from the stack test result waspretty large (3. 07 Ib/hr for the 48.0 Ib/hr limit), I'm not sure how you can demonstrate the compliance with the 2D .0515standard at 200 amp and 20kV, which is drastically different than what it was tested. I don't think you can use theemission factor from the stack test (3. 07 Ib/hr) unless you operate the WESP with the same parameters.

Enviva has performed a detailed review of the October 3, 2013 stack test results and vendor performance specificationsof the WESP and have concluded that the emissions from the dryer comply with the process weight rate rule particulatematter (PM) emissions limit of 2D .0515 without use of the WESP. Therefore, CAM does not apply and the proposedoperating parameters above are not required. Please see below fora detailed discussion of the emissions review anddocumentation of CAM applicability.

Based on the vender's tetter dated on May 10, 2017, the proposed minimum voltage was 20kV. The equation that wasused for the calculation was based on the input rate of 54. 91 Ib/hrl am a little concerned about this input rate as thebefore control emission rate was never tested during the stack testing. The permit application indicated that the beforecontrol emission rate was 150 Ib/hr, which is not consistent with the vendor's letter. The equation that the vender usedto calculate the PM emissions from the WESP is clearly dependent of the input rate. Therefore, I cannot approve to usethe vender's calculation unless I can confirm the validity of the input rate.

Based on the October 3, 2013 stack test results and vendor performance specifications for WESP control efficiency, theinlet PM emission rate to the WESP during the test is calculated to be 35.47 Ib/hr. Therefore, as stated above, CAM doesnot apply and the WESP is not required to comply with the 2D. 0515 standard. Please see below for a detailed discussionof the emissions review and documentation of CAM applicability.

Another issue that I'm concerned is the applicability of CAM. The following are the three criteria to be subject to CAM:1. be subject to an emission limitation or standard, and2. use a control device to achieve compliance, and3. have potential pre-controt emissions that exceed 100 tpy.

It's pretty clear that the dryer satisfies the first two conditions. It appears to me that the dryer also meets the thirdcriteria. Based on the application, the befor e control emission rate was 150 Ib/hr, which is equivalent to 657 tpy. If weuse the input rate indicated on vender's letter (54. 92 Ib/hr), it would be 236 tpy. Either way, the pre-control emissionsexceeds 100 tpy. Therefore, the dryer is subject to CAM.

As discussed above, Enviva reviewed the October 3, 2013 stack test results and vendor specifications of WESP controlefficiency to determine the WESP inlet PM emission rate (uncontrolled dryer emission rate). The October 3, 2013 stacktest resulted in an outlet WESP PM emission rate of 3.07 Ib/hr. Based on the vendor specifications of the WESP andactual flow rate measured during the test (109, 700 acfm) the control efficiency corresponding to the test was91.35%. The resulting uncontrolled dryer PM emission rate during the test was calculated to be 35.47 Ib/hr.

As stated by Lundberg in the May 10, 2017 letter, WESP control efficiency is based on WESP collection area, effectivemigration velocity, and actual gas flow rate. Page 2 of the attached document includes the control efficiency of theWESP based on the vendor's calculation.

Enviva also evaluated potential uncontrolled PM emissions from the dryer at the maximum design throughput rate of71.71 odt/hr. The process throughput during the October 3, 2013 stack test was 60 odt/hr. The resulting uncontrolleddryer PM emission factor is 0. 59 Ib/odt (35. 47 Ib/hr divided by 60 odt/hr). This factor multiplied by the maximum designthroughput rate of 71.71 odt/hr results in an uncontrolled dryer PM emission rate of 42.40 Ib/hr, which is less than thePM limit calculated in accordance with 2D .0515. Therefore, condition 2 above is not met and the dryer is not subject to

CAM. Page 1 of the attached document includes the process throughput information, PM emission limit, and potentialPM emissions.

We then need to determine if the WESP is considered to be a large PSEU. If the potential post control PM10 emission islarger than 100 tpy, the WESP will be a large PSEU. Even though the stack test from 2013 shows PM emission factor was1.54 Ib/hr, which is equivalent of 6.74 tpy, this emission factor is not relevant unless you want to operate the WESP atthe same parameters used at the stack test. If we use the emission rate of 42 Ib/hr at 20 kV from the vender's letter, theemission would be about 42. 00 Ib/hr, which is equivalent to 183.96 tpy, and would make it as a large PSEU.

As documented above and in the attached, CAM is not applicable.

The difference between small and large PSEU is the deadline for submitting a CAM plan. If the WESP is a small PSEU, wecan wait to for the next. If it's large, we have to address CAM for this permit.

It is up to you which minimum voltage and current you want to use, but it will also affect CAM depending on how youoperate the WESP. Based on the information I have (and I don't have), I think we need to include a test requirement onthe permit. I think we need to test for emissions at 20 kV and 200 amp to show the compliance. We need to measurethe before and after the control so that we know the control efficiency of the WESP (this was on the permit when thetesting was conducted, but they did not test for the efficiency for some reason.) Also, we need to incorporate CAM atthis time as I think this would be considered as a large PSEU. For more information regarding CAM, please review thispage:htt s: ncdenr. s3. amazonaws. com s3fs- ublic Air%20 ualit enf cam CAM Overview, df

As documented above and in the attached, CAM is not applicable. Enviva does not believe testing of the WESP inlet andoutlet for control efficiency is warranted. As stated above and in the May 10, 2017 letter provided by Lundberg, WESPcontrol efficiency is dependent on WESP collection area, effective migration velocity, and gas flow rate. The WESPcollection area is fixed, effective migration velocity is proportional to the secondary voltage, and gas flow rate wasmeasured during the October 3, 2013 stack test. Therefore, all values required to calculate WESP control efficiency areknown and WESP inlet testing should not be required.

Please fill out this Form E6 to submit a CAM plan.htt s: ncdenr. s3. amazonaws. com s3fs- ublic Air%20 ualit ermits files 2016 Permit A lication Forms E6. df

If you have further questions, please let me know.

Thanks.

Yuki Puram


9197078470 officevuki. [email protected]


Enviva Northampton Dryer Particulate Matter Emissions Estimate

Process Information(1)Dryer ThroughputMoisture Content

Wet Process Weight

ISA NCAC02D.0515Paniculate Limit:(2)

October 2013 Test Data

WESP Outlet PM Emissions RateSecondary Voltage

WESP Efficiency(3)Uncontrolled PM Emission Rate

Dryer ThroughputUncontrolled PM Emissions per ODT

Uncontrolled Emissions @ Maximum DryerThroughput(4)Uncontrolled Emissions Less Than ApplicableLimit?

71. 71

17%83.90

ODT/hr

tons/hr

49.53

42.40

Ib/hr

3.0763.5

91.3535.47

600.59

Ib/hrkV%

Ib/hrODT/hrIb/ODT

Ib/hr

Y/N

Notes:

111 Dryer throughput represents the design maximum process rate of oven dried tons (ODT) perhour. Moisture content represents the average outlet moisture content of dried wood. Wetprocess weight is based on the maximum process rate of 71.71 ODT/hr and the average outletmoisture content.

121 Particulate Limit is based on the process weight rate equation from ISA NCAC 02D. 0515.

131 WESP efficiency during the October 2013 based on vendor efficiency calculations.141 Uncontrolled emissions at the maximum dryer throughput is calculated by multiplying theOctober 2013 uncontrolled PM emission factor of 0.59 Ib/ODT and the maximum dryerthroughput of 71. 71 ODT/hr.

000001-* Ni U* 4i» 01

pco

!-''-' N

1->

U1t-»

00N

N

N

U1N

IDu

tuLUco

4^M

^>.^1

inNJ

U1OB

014»

^J.~101

COU)

1.0u>'«! s K S G G

.^1 U>t>M

.& In01 h->

w s'U1 "t

l° P G K 1-> ^ N W.~) ID >-> U

M M U».>J U> 1->

"<U)

u>U1

u>.^1

4>10

«> *.1-> W

Û1

4!>~>1

4^<D

tn ai CTIU> t-> Ul 2 (D

1/1 a 3

su) lo °° <»UJ 0 (Tl tU

ION 0 (D <D

~i ~sl

S B s000*> ^

0 0UJ W NJ

000M N IS)

^-^. E-ooii-'^wooiNiooiwooocnSNou>ui^t->Otuo6wi^j-iuiNu0âSF.>i^5i->Cn

p pp p p00000co » q .5m.sj Uk) -^1 1-> 0^

CUtAJOJtAiUJtAJhJNJrsJM*>AtJUUJI-»plOpO >IU1

NcnQU>oo'^J>>lMooUIUI-. IIONOSWUlOlm

N N) N ^.4^ N j-> U>i-» op u co h>0 0^h-> 0^ 10

U) LUU1 U1.f. :t><^ <^

U)U1

w w u uy ui ui ui

tU U)Oi Oi

s <n in ^ NU1 N 00w 10 e

s

§ I

43. *. 4Si 4^ <>.^J ~>1 . ~1 . »J -^

CD

8 § § g §

U) W U Utn Ln t/i ui

^ ^ ^ ^!tu tu u>in ui ui^ *> <>.^1 . ^1 <J

0

yu>po <>i m uitUN N 4!> .̂ 1w inio ui o s y

UJ U)

s sW M NJ MMQID ->J U1 tU.~l In ui M ui

UJ tU tU tUOJU1 U1 UILh 1/1*;.t> .&:*:*><>.^J -»J ~~1 '»! .~1 .<!

s yi s sU1*. 0 <»

UlUUJUJWUlUUltUtUuiuiintnCn&iunOiuiOi4!>4»4a. 4ai *>.(>*>*>. B>t>.^'^. sj'^-si-j^l'^î^

.10 10rsj N»>1 ~ôo co

WID ^D ^D <>D

M N N M M^ . h*J .^'sj -^10006 06 00 06

s

§

-10

's

s

§

-10

's

s

s

.10

s

.s8

.u>

§

8

u>

s

8

10

s

s

§

10

s

s

§

s

§

s

§

5

's

s

8

K

's

s

§

s

s

s

§

U)

's

.s8

>0

§

s

s

<0

s

s

8

10

§

s

8

.5

s

s

8

.10

's

s

§

w

s

s

§

10

's

s

§

-u>

§

s

s

ID

's

s

§

s

s

s

8

us

's

s

8

ID

s

s

§

s

s

s

§

s

's

s

§

s

's

s

8

s

s

Ii§Q

lu

ftin

K<T>s

ss

KK

sK

sg

K

t;yK

ys

ss

îs

s?

sy

Bs

sK

!Ss

8StS

?s

s

:jag

°<s

ss

E2s

sss

ss

S3g

sfe

^§

^s

s§3

sy

ss

Ss

'>"Nothing Compares


Disclaimer



Disclaimer



Enviva Northampton Dryer Paniculate Matter Emissions Estimate

Process Information<1)Dryer ThroughputMoisture Content

Wet Process Weight

ISA NCAC02D.0515Paniculate Limit:(2>

Uncontrolled Emissions @ Maximum DryerThroughput(4)Uncontrolled Emissions Less Than ApplicableLimit?

71. 71 ODT/hr

17%83.90 tons/hr

49. 53

42. 40

Ib/hr

October 2013 Test Data

WESP Outlet PM Emissions Rate

Secondary Voltage

WESP Efficiency(3>Uncontrolled PM Emission Rate

Dryer ThroughputUncontrolled PM Emissions per ODT

3.0763.5

91.3535.47

0. 59

Ib/hrkV%

Ib/hrODT/hrIb/ODT

Ib/hr

Y/N

Notes:

111 Dryer throughput represents the design maximum process rate of oven dried tons (ODT) perhour. Moisture content represents the average outlet moisture content of dried wood. Wetprocess weight is based on the maximum process rate of 71.71 ODT/hr and the average outletmoisture content.

p> Particulate Limit is based on the process weight rate equation from ISA NCAC 02D. 0515.

131 WESP efficiency during the October 2013 based on vendor efficiency calculations.(4) Uncontrolled emissions at the maximum dryer throughput is calculated by multiplying theOctober 2013 uncontrolled PM emission factor of 0.59 Ib/ODT and the maximum dryerthroughput of 71. 71 ODT/hr.

5^^^ Ib^ y7^-^^Yr

T-i2 --ec? (A

^ /S'T ^^ Tp\/

ENVIVA NORTHAMPTON WESP CONTROL EFFICIENCY

70-

15.5

15.1

14.6

14.2

13.9

13.7

12.8

12

11.2

10.4

9.7

9

8.3

7.6

7

6.4

5.8

5.2

4.7

4.2

3.8

3.3

2.9

2.5

2.2

1.8

1.5

r?~

1

0.8

0.6

0.4

0.3

0.2

0.1

67

66

65

64

63.5

63

61

59

57

55

53

51

49

47

45

43

41

39

37

35

33

31

29

27

25

23

21

19

17

15

13

11

9

7

5

0. 066

0. 071

0. 077

0.083

0. 087

0. 090

0. 105

0. 121

0. 140

0. 161

0. 182

0. 206

0. 233

0. 263

0. 292

0.325

0.361

0.401

0. 438

0.478

0. 513

0. 560

0. 601

0. 644

0. 679

0.729

0.768

0. 810

0.839

0. 869

0.900

0. 932

0. 949

0. 965

0. 983

iesaate

b/hr)

2.33

2.52

2. 73

2. 95

3.07

3. 19

3. 74

4. 31

4. 96

5.70

6.45

7. 29

8. 25

9. 33

10. 37

11. 52

12. 80

14. 22

15.53

16. 96

18. 19

19. 86

21. 31

22. 86

24. 10

27. 25

28. 73

29. 76

30. 82

31. 92

33. 07

33. 65

34. 25

34, 86

^

35.47

35. 47

35. 47

35. 47

35.47

35. 47

35. 47

35. 47

35. 47

35. 47

35. 47

35.47

35. 47

35. 47

35. 47

35. 47

35. 47

35. 47

35. 47

35.47

35. 47

35. 47

35. 47

35.47

35. 47

35. 47

35. 47

35. 47

35. 47

35. 47

35.47

35. 47

35. 47

35. 47

35. 47

J109, 700

109,700

109, 700

109,700

109, 700

109, 700

109,700

109, 700

109, 700

109,700

109,700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

109,700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

109, 700

L_..19, 278

19, 278

19,278

19,278

19, 278

19,278

19, 278

19, 278

19, 278

19,278

19, 278

19,278

19. 278

19,278

19, 278

19, 278

19,278

19, 278

19, 278

19,278

19, 278

19, 278

19,278

19,278

19,278

19, 278

19,278

19, 278

19,278

19, 278

19, 278

19,278

19, 278

19, 278

19, 278

93. 44

92. 90

92. 31

91.68

91. 35

91.00

89. 45

87. 86

86. 03

83.92

81. 82

79. 44

76.74

73. 70

70. 77

67. 52

63. 91

59. 90

56. 22

52.20

48.72

44.01

39. 93

35. 55

32. 06

27. 12

23. 17

19.01

16. 12

13. 12

10. 01

6. 79

5. 14

3. 45

1.74

Control Efficiency calculated using LUNDBERG's equation. E = i.e-(AW/QlA = Collecting Area (ft2)Q = Gas Flow Rate (acfm) = 109,700 arfm as measured during October 2013 Stack Testw = effective migration velocity (ft/min), as provided by LUNDBERG in letter dated 5/10/2017

£

I

w

Is(B

hHffl I2 S

."»-

lis 5:

I-tt ?s ^

Isv

<-l 2

I(0

I0

?o?o?OWMSOWSOSOSOSOSOS?S?SOSOSOOOOO°oooooooa'w«'oooooooooooopooopooooooopopo&r^. !^!?:!^:[^:^^^î^'is~i>s-fs-f<i<i<i<i"î';i:îs ;-^F^t;î;^t:î:î:^F:^F^F^F:îî::^

r^ (^ r^ M r^ ^ M N f^rNN^N^N^N^N^MîN rsl (N IN (N N CN r^ (N <N fN <N r>i r^ ri ri ri r^ FN ^ fN; fN o3a; WO^WCT moiCTt a; w en <n <n en en en en: CTt: w (n en o^ CTI' en en en en en <n <n o<' en 01" en en en en en en en

iH^T^^THT^rtT^^^rH^^^rt»Hrtt-ii--iT-it-i»-i?-ir-iT -i,-irt?-i?-i?^?-i?-i?-i?-i?-i?-i

|S§SSSS§§S§§§§S§§S§§§§§§§S§§§S§§§§§§S§S81^' '^ r?. 1^- r^- r>*.. '^ 1^. r^ 1^ '^ rs,. l'\ ls:;,. '^ '^ '^ '^ i^ r^ ^ ^ r^. i^ i^ i^ i^ r::- ^ ^ r:::- i^- i^-^^ ^ i^ i^ i^ ^ggggggg g g § g § g g g g g § § g g g g § g g g g g § g g g g g g g g g g

r^t-lt-l»-lT-lT-li-li-lrti-lrHi-1r-l»-<<-lr-lt-lT:Hr-irttH?i?:iT =i?:i?:i?i?HT??????5???i?(?i?i?i

cnaioiCT}0}o^<T»aiaiaiaiaôioi(n<T>cno)aia^CTiaioioioi<ncnaicncnoio^CT>cnoiCTtOioioioi'S.'5t';t'!'t'i't'?'!f'^'it'^'^'^''^^'?:'!t'!?''!^'!f'^'a:^:'<i:'t'*'«t<d:'^:'^:<y:î:<si:'t<t<a:'T<T^):iniînuiininL OLni^tninininini/iini/iuîniflininintoininiAininLntOLninLntnuStntnLnLn

rsimtnor^toi^-r-irîôir\it0oimr~. r<iu3

mro^t-'srTinLntOiDrs. r^. oooocn

u3porM<napipi<<»S?Q!5Sr-i to FM [<m

r< <* o roCT^'-'SommL naiinmminoorHipT -ir^Qipmpoqu?<!3-Mr-ioiO)oooooooooo<nOi-ir<imifl^booCTiQQi-!rMrMrr)îflu3u3r~oocr>Or-ir>iminiorôoaiO<-ir^^-T-it-ir-li-it-i<-ii-ii-ir-ir-ir-i^^-i. (Mr>ir^r<ir^r<jrMfM(Nmm(nm

lor>1003r'io9L£?m<Nt-iQQQr(roL oi~.. o^t-oomoo<itor~. <srrôcorî><tOu3r^rôocnOT-ifM^r^rôog?CT>pi HfNm^tn^rôo(J)Or^cn<tior>. dSr:iiN<û3oooSr-imLnr::. ijiT-imi?iooc>NppQpQOr^^r^^rH^THrHîHr^cNrîr^rîîrNmrnmmrnm^T-^'^r'^-^l-'LnuSuSîBio0000000000000000000000000000000000000000

r~. iû-i<d-mrMT-iQCT>oor~. y3in<Tmr<it-iooioor~. t£iin^tmr<i<-iu3uDU)t£)ioiou3U3Lntninir>inu'»ininuiLn^r^^^-^t^-<d-<!t<j-

to mmm m m m ^ m FM r-i o cnoo

m m mm r^ M

I n l/30to^^MO°?ou3<NOS30r~-^QUimcnu3mp^, <tôpLnfMQrsin^Qooinm^cT>rs.3SS333333?!^lSSSC TiCT'cnodG°r<:F<^rs;uiu5U3L/>l^lûi':t^:^:'15l:<v>mmmoJN

2.5

2.32.22.01.81.71.5

1.41.21.11.00.90.80.70.6

27262524

23222120191817

15,

13

0.643

0.6640.6850.7060.7260.7460.7660.7850.8040.8220.8400.8560.873oiss,'0.903

35. 2836.4337.5738.7139.8240.9242.0043.0644.0845.0846.0546.9847.8748.71,49. 52

54.954.954.9

54.954.954.954.9

54.954.954.954.954.954.954.9?54.9

109,700109,700109,700109,700109,700109,700109,700109,700109,700109,700109,700109,700109,700109,700109,700

19,27819,27819, 278

19,27819,27819,27819,27819,27819,27819, 278

19,27819,27819,27819,27819, 278

Based on Deutsch Equation; LnP = -Au/Q where A = Collection area ofESP, Q. = flow through ESP, P=Penetration (1-efficiency) and u is theeffective migration velocity (EMV)

^.1\ ^%7f,7| OPT/hr

^/^/ 7l. 7|''IL4-o"'^. '°|b/^-

J D E13201 Bel-Red Road

Bellevue, Washington 98005tel: 425. 283. 5070fax: d25. 283. 5081

May 10, 2017Attention: Mr. Joe Harrell


Subject: Use of Secondary Voltage (kV)as an Indicator of WetElectrostatic Precipitator(WESP) Performance

Enviva Pellets Ahoskie, LLC142 NC Route 561 East

Ahoskie, NC 27910 USA

Dear Mr. Harrell.

Thank you for your inquiry regarding the use of secondary voltage (kV) as an indicator of theperformance of WESPs manufactured by our company and installed at Enviva's Northampton, NorthCarolina pellet facility. As we have discussed, the most important factor in precipitator performance isthe electrostatic field strength (kV/inch). This value is maximized when secondary voltage ismaximized.

In short, the efficiency of a wet precipitator is defined by the following equation:E = 1 - e'(A<B/Q)

Where:

E = EfficiencyA = Collecting Area (ft2)Q = Gas Flow Rate (actual ft3/mjn)d) == effective migration velocity (ft/min)

(Stated alternativelyp ^ g-(AU/Q)

or

In P = -ACD/Q

Where:

P = Penetration = 1- Efficiency)

To increase WESP performance, the options are to either increase A, increase W, or decrease Q.Obviously, the WESP collecting area (A) and gas flow rate (Q) are constants in the Northampton WESPapplication. Therefore, the only option is to increase the migration velocity (W).

The migration velocity is directly proportional to the charging field strength times the collecting fieldstrength. In effect, this means that the migration velocity is proportional to the secondary voltage,squared. Thus, the secondary voltage of the WESP is clearly the best option as a parametric monitoringdata point to indicate performance.

JACKSONVILLE, FLORIDA MONROE, LOUISIANA NAPERVILLE, ILLINOIS

WWW. LUNDBERG-US. COM

OLD SAYBROOK, CONNECTICUT BILBAO, SPAIN

Our review of emissions test data from the subject plant shows that the Lundberg Model 567 WESP inoperation at Northampton is operating within the expected design parameters for the airflow and otherconditions observed. Based on this, we have developed the table below to describe WESP total gridaverage control efficiency over a range of secondary kV in the Northampton application. (Note: TheWESP inlet concentration used in the table below is assumed to be 0. 10 grains/scfd.)

Enviva Northampton Wet ESP Performance Model

^'©utlet^llnleit,''- »; ... ~ ' "'"^'.. '.s

?^. ^£'^^vy . s ^'M^yy%E^V,

"

Voltage jPeTOti^ '. w!"'c1' hli;J(ft/min); ^fltV) ^ (fractido) ?C^^i ^a

(=16w Art-a.

15.514.613.712.812.011.210.49.79.08.37.67.06.45.85.24.74.23.83.32.92.52.21.81.51.21.00.80.60.40.30.20.1

676563615957555351494745434139373533312927

25232119171513119

7

5

0.0660.0780.0910. 1050. 1220. 1400. 1600. 1830. 2070.2340.2630.2940.3270.3620.3980.4370.4760.5170.5590.6010.6430.6850.7260.7660.8040.8400.8730.9030.9290.9520.9710. 985

^ltN'^,3.624.254. 975. 776.677.688.7910.0211.3612.8314.4116. 1017. 9119.8321.8523.9526. 1328.3830. 6632.9735.2837.5739. 8242.00^0846.0547.8749.5250.9852.2353.2554. 03

flb/h):54.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.954.9

(acfm)r^ '{ft2).

109,700109,700109, 700109,700109,700109,700109,700109,700109, 700109, 700109, 700109,700109,700109, 700109,700109,700109,700109,700109,700109, 700109, 700109,700109,700109,700109,700109,700109, 700109,700109,700109, 700109,700109, 700

19,27819,27819,27819,27819, 27819,27819,27819,27819,27819,27819,27819, 27819,27819,27819,27819, 278

19,27819,27819,27819,27819,27819,27819, 27819,27819,27819,27819,27819,27819,27819,27819,27819,278

LUNDBERG Northampton letter

From this table Enviva can infer the WESP outlet emission rate at a given total average of all grids at akV value.

Thank you for contacting Lundberg regarding this important matter. If you have any questions orcomments regarding this matter please do not hesitate to contact me at (425) 283-5070.Sincerely,

v

Steve A. Jaasund, P. E.

Manager, Geoenergy Products

Lund berg

LUNDBER6 Northampton (ettet

^ ^l/ / 2-Go ^ hp / ^uAw!t ̂ ^ ^,

Review DSAFT

B.3 WET ELECTROSTATIC PRECIPITATORS^7 8'22;>3

B. 3. 1. Back round

A wet electrostatic precipitator (WESP) typically is used to control PM emissions inexhaust gas streams containing sticky, condensible hydrocarbon pollutants, or where thepotential for explosion is high. A WESP may be used to control a variety of emission points andpollutants, such as wood chip dryers; sulfuric acid mist; coke oven off-gas; blast furnaces;detarring operations; basic oxygen furnaces; cupolas; and aluminum potlines. In the woodproducts industry, WESPs often are used in combination with wet scrubbers or regenerativethermal oxidizers (RTOs) to control both PM and gaseous emissions. The general operatingprinciples and components ofESPs and the specific features ofdryESPs are discussed insection B.2; this section focuses on the components and operation ofWESPs that differ fromthose of dry ESPs.

The two primary differences between dry ESP and WESP design are the use of aprequench and the collector plate cleaning method. Unlike dry ESPs, WESP control systemstypically incorporate a prequench (water spray) to cool and saturate the gases prior to enteringthe electi-ical fields. As PM accumulates on the collector plates of a WESP, the plates arecleaned by a continuous or intermittent film or spray of water. Major differences in the types ofWESPs available include: the shape of the collector; orientation of the gas stream (vertical orhorizontal); use of preconditioning water sprays; and whether the entire ESP is operated wet.Configurations include circular plate, concentric plate, tubular, and flat plate WESPs.

In circular-plate WESPs, the cu-cular plates are irrigated continuously; this provides theelectrical ground for attracting the particles and also removes them from the plates. Concentric-plate WESPs have an integral, tangential prescmbbing inlet chamber, followed by a verticalwetted-wall concentric ring ESP chamber. The discharge electrode system is made of expandedmetal, with corona points on a mesh background.

Tube-type WESPs typically have vertical collecting pipes; electrodes are typically in theform of discs placed along the axis of each tube. The particles are charged by the high-intensityelectric field, and, as they travel farther down the tube, they are forced to the tube walls by theelectrostatic field. The tube walls remain wet because the fine mist entrained in the saturated gasis also collected on the tube surfaces and flows down along the tube walls. Flushing isperformed periodically to clean the tube surfaces. The water is collected in a settling tank, andthis water is used to quench the gaseous stream prior to its entering the WESP.

In rectangular plate WESPs (horizontal flow), water sprays precondition the incominggas and provide some initial PM removal. Because the water sprays are located over the top ofthe electrostatic fields, collection plates are also continuously irrigated. The collected water andPM flow downward into a sloped trough. The last section of this type ofWESP is sometimesoperated dry to remove entrained water droplets from the gas stream.

CAM TECHNICAL GUIDANCE DOCUMENTB.3 WET ELECTROSTATIC PRECIPITATORS

4/02 B-21

Review DRAFT

The conditioning of the incoming gas sti-eam and continual washing of the internalcomponents with water eliminate re-entrainment problems common to dry ESPs. Efficiency isaffected by particle size, gas flow rate, and gas temperature. Common problems with WESPsinclude: poor gas flow; high gas flow; poor water flow; low voltage; low current; and highdissolved solids in the flush or prequench water. Other common mechanical-type problemsinclude: poor alignment of electrodes; bowed or distorted collecting plates; full or overflowinghoppers; plugged water sprays; corrosion of electrodes; and air inleakage.

B.3.2 Indicators of WESP Performance

The primary indicators ofWESP performance are opacity, secondary corona power,secondary voltage, and secondary current. Other indicators ofWESP perfonnance are the sparkrate, primary current, primary voltage, inlet gas temperature, gas flow rate, inlet water flow rate,solids content of flush water (when recycled water is used), and field operation, section B-2describes each of these indicators with the exception of the inlet water flow rate and the flushwater solids content, which are described below. For some systems, mist may be entrained in theexhaust gas. In such cases, opacity measurements would be misleading. Table B-3 lists theseindicators and illustrates potential monitoring options for WESPs.

Inlet water flow rate. Because WESPs use water to clean collector plates, the water flowrate is an indicator that the cleaning mechanism is operating properly. If flow rates decrease,sections of the WESP may not be as effective. As a result, PM collection rates would decreaseas material built up on the collectors. In addition, low flow rates increase the likelihood ofineffective spraying and distribution of water, as well as nozzle plugging.

Flush water solids content. When recycled water is used, the solids content of the waterincreases with each recycling. If the solids content becomes excessive, the effectiveness of thecleaning mechanism is reduced. Increased solids content also can lead to plugging of spraynozzles.

B.3.3. Illustrations

WESPs:The following illustrations present examples of compliance assurance monitoring for

3a: Monitoring secondary current, secondary voltage, spark rate, and inlet water flowrate.

3b: Monitoring secondary current, secondary voltage, inlet water flow rate, and flushwater solids content.

B.3.4 Biblioeraphv

CAM TECHNICAL GUTOANCE DOCUMENTB.3 WET ELECTROSTATIC PRECIPITATORS

B-22 4/02

-t".

s Review DRAFT

§

TABLE B-3. SUMMARY OF PERFORMANCE INDICATORS FOR WESPs

Approach No. 1 2

Illustration No. 3a 3b

Example CAM Submittals

Parameter Performance indication Comment .

Primary Indicators of Performance

Opacity Increased opacity or VE denotes perfonnance degradation. COMS, opacity observations, orvisible/no visible emissions. If mist is entrained in exhaust gas or a condensed plume ispresent, opacity measurements may be misleading.

x

x

A9a A9b

"

K

g ̂ Secondary corona Performance usually increases as power input increases; indicates work done by WESP to^ § power remove PM. Product of voltage and current; can help identify any fields that are not

§ 9.% P Secondary current Partial indicator of power consumption; too low indicates malfunction. Can help identify^ i* any fields that are not operating properly.ti ?n § Secondary voltage Partial indicator of power consumption; too low indicates problem such as groundedg S electrodes. Can help identify any fields that are not operating properly.

Other Performance Indicators

Inlet water flow Indicates cleaning mechanism is working properly; if low, can indicate plugging. As anrate alternative to water flow, the water pressure can be monitored.

Flush water solids High solids may cause plugging, reduce collection efificiency. Applies to systems that usecontent recycled water.

Inlet/outlet gas Indicates water sprays and prequench (if applicable) are working. Also, temperature affectstemperature resistivity ofparticulate.

Comments:

. Approach No. 2 also corresponds to 40 CFR 60, subpart PPP (Wool Fiberglass).

. Approach No. 3 includes monitoring the voltage to indicate that the WESP is collecting paniculate, VE as an indicator ofPM emissions, water flow toindicate PM being removed, and outlet temperature to indicate sufBcient water.

a Monitoring both secondaiy current and voltage is essentially the same as monitoring secondary corona power. Monitoring of corona power is notappropriate for WESPs with a large number of fields.

No Part 63 rules refer to WESP.

x

x

x

x

x

x

x

x

x

x

x

x

Review DRAFT

CAM ILLUSTRATIONNo. 3a. WET ELECTROSTATIC PRECIPITATOR FOR PM

1. APPLICABILITY

1. 1 Control Technology: Wet electrostatic precipitator (WESP) [010, 011, 012]1.2 Pollutants

Primary: Particulate matter (PM)Other:

1.3 Process/Emission units: Wood products dryers

2. MONITORING APPROACH DESCRIPTION

2. 1 Parameters to be Monitored: Secondary current, secondary voltage, and inlet waterflow rate.

2.2 Rationale for Monitoring Approach. Secondary current: Current is generally constant and low; increase or drop in

current indicates a malfunction. The current directly affects collection efficiency.. Secondary voltage: Voltage is maintained at high level; drop in voltage indicates a

malfunction. When the voltage drops, less particulate is charged and collected. Thevoltage directly affects collection efficiency.

. Inlet water flow rate: Indicates sufficient water flow for proper removal ofparticulate from the collection plates.

2.3 Monitoring Location. Secondary current and secondary voltage: Measure after each transformer/rectifier

set.

. Inlet water flow rate: Water line.

2.4 Analytical Devices. Secondary current: Ammeter.. Secondary voltage: Voltmeter.. Inlet water flow rate: Liquid flow meter or other device for liquid flow; see

section 4 for more information on specific types of instruments.2.5 Data Acquisition and Measurement System Operation

. Frequency of measurement: Hourly, or continuously by strip chart or dataacquisition system.

. Reporting units:Current: Amps.

- Voltage: Volts.- Inlet water flow rate: Gallons per minute (gpm) or cubic feet per minute (ft3 /min)

. Recording process: Operators log data manually, or recorded automatically on stripchart or data acquisition system.

2. 6 Data Requirements. Baseline secondary current, secondary voltage, and inlet water flow rate

measurements concurrent with emission test.

B-24


4/02

Review DRAFT

. Historical plant records on secondary current, secondary voltage, and inlet waterflow rate measurements.

2. 7 Specific QA/QC Procedures: Calibrate, maintain, and operate instaiimentation usingprocedures that take into account manufacturer's specifications.

2.8 References: 7, 8, 9, 13.

3. COMMENTS

3. 1 Data Collection Frequency: For large emission units, a measurement frequency of onceper hour would not be adequate; collection of four or more data points each hour isrequired. (See Section 3.3. 1.2.)

4/02

CAM TECHNICAL'GUIDANCE DOCUMENTB.3 WET ELECTROSTATIC PRECIPITATORS

B-25

Review DRAFT

CAM ILLUSTRATIONNo. 3b. WET ELECTROSTATIC PRECIPITATOR FOR PM

1. APPLICABILITY

1. 1 Control Technology: Wet electrostatic precipitator (WESP) [010, 011, 012]1.2 Pollutants

Primary: Particulate matter (PM)Other:

1.3 Process/Emission units: Insulation manufacturing, dryers

2. MONITOMNG APPROACH DESCRIPTION

2. 1 Parameters to be Monitored: Secondary voltage and current, inlet water flow rate, andsolids content of flush water.

2.2 Rationale for Monitoring Approach. Secondary current:. Secondary voltage: Low voltage or current indicates a problem in the WESP.. Inlet water flow rate: Indicates suflficient water flow for proper removal of

particulate from the collection plates.. Flush water solids content: High solids content of recycled water reduces the

efficiency of cleaning.2.3 Monitoring Location

. Secondary current and secondary voltage: Measure after each transformer/rectifierset.

. Inlet water flow rate: Measure at inlet water inlet line or pump discharge.

. Flush water solids content: Measiire at inlet line or recycle water tank.2.4 Analytical Devices:

. Secondary current: Ammeter.

. Secondary voltage: Voltmeter.

. Inlet water flow rate: Liquid flow meter or other device for liquid flow; seesection 4 for more information on specific types of instruments.

. Flush water solids content: Manual sampling of water.2.5 Data Acquisition and Measurement System Operation

. Frequency of measurement: Hourly, or continuously on strip chart or dataacquisition system; flush water solids, weekly.

. Reporting units:Current: Amps.

- Voltage: Volts.- Inlet water flow rate: Gallons per minute (gpm) or cubic feet per minute

(ft3 /min).- Flush water solids content: Percent solids.

. Recording process: Operators log data manually, or recorded automatically on stripchart or data acquisition system.

B-26


4/02

Review DRAFT

2.6 Data Requirements. Baseline secondary current, secondary voltage, inlet water flow rate, and solids

content measurements concurrent with emission test.

. Historical plant records on secondary current, secondary voltage, inlet water flowrate, and solids content measurements.

2. 7 Specific QA/QC Procedures: Calibrate, maintain, and operate instrumentation usingprocedures that take into account manufacturer's specifications.

2. 8 References: 7, 8, 9, 11, 13.

3. COMMENTS

3. 1 Data Collection Frequency: For large emission units, a measurement frequency of onceper hour would not be adequate; collection of four or more data points each hour isrequired. (See Section 3. 3. 1.2.)


4/02 B-27

if you have any questions or require additional information, please …...

Documents