simulation of robotic tig-welding - product development ... · simulation of robotic tig-welding...

Simulation of robotic TIG-welding

Mikael Ericsson

Department of Technology University of Trollhättan/Uddevalla

P.O. Box 957 SE-461 29 Trollhättan, Sweden

Division of Robotics Department of Mechanical Engineering

Lund Institute of Technology Lund University, P.O. Box 118, SE-221 00 Lund, Sweden

CODEN: LUTMDN/(TMMV-5170)/1-96/(2003) ISBN 91-628-5702-9 2003 by Mikael Ericsson and Department of Technology University of Trollhättan/Uddevalla. All rights reserved Printed in Sweden KFS I Lund AB, Lund

Abstract

Robotised welding is one of the most important robot tasks used in manufacturing industry. The operator usually performs the programming of the robot manually, i.e. by jogging the robot arm to each coordinate pose in space. Programming can, however, be made more accurate by the use of simulation, using so called Computer Aided Robotics. Simulation can also be a powerful tool to evaluate and control welding heat effects, such as unwanted stresses and deformation.

The objective of this thesis was to develop a simulation tool and a method by which robot trajectories, temperature histories, residual stresses and distortion can be analysed and optimised off-line. This was performed by integrating robot simulation software with finite element analysis software. A special interface was created allowing information exchange between the two software programs.

The method was used to program welding trajectories both for planar plates and for a part of an aerospace component. The trajectories were downloaded to the finite element analysis software where temperature and residual stress prediction were performed. Good agreement was found between the programmed robot trajectory, and the actual trajectory and only small adjustments were necessary. Temperature measurements were performed using both thermocouples and infrared imaging. Good agreement was also found between the results using these two methods.

The method developed provides a powerful tool to construct and optimise robot trajectories and welding process parameters off-line.

Acknowledgments

First, I would like thank to Professor Gunnar Bolmsjö, Lund Technical University, for his support and help during this study. I would also like to express my indebtedness to my second supervisor, Dr. Per Nylén, University of Trollhättan/Uddevalla, for his support, contributions and his enthusiasm for this project. Without his help, none of this had been possible.

I would like to express my appreciation to several people at Volvo Aero Corporation. Mr. Per Henrikson, for his help and support with temperature measurements, Mr. Börje Nordin for his willingness to share his knowledge about robotised TIG welding and Techn. Lic. Daniel Berglund for helping me to discover the wonderful world of finite element analysis.

I would also like to express my gratitude to the members in the VIP research group at the University of Trollhättan/ Uddevalla for all their help, especially Mr. Xavier Guterbaum for all our valuable discussions in the robot laboratory. I thank too Mr. Alastair Henry and Dr. Anita Hansbo of University of Trollhättan/Uddevalla their careful linguistic revision. Thanks also to the research team of the robotic division at Lund Technical University for all their help and assistance.

The project was funded by the Foundation for Knowledge and Competence Development and EC Structural Founds.

Finally, I would like to take this opportunity to express my gratitude to my parents, Sten and Ingrid, my brother Stefan and my fiancée Anna, for all their support and understanding during this time

Mikael Ericsson March 2003 Trollhättan

Contents

Abstract III

Acknowledgments V

Contents VII

List of figures IX

List of tables X

List of acronyms X

1 Introduction 1 1.1 Background and motivation ............................................................... 1 1.2 Objectives........................................................................................... 2 1.3 Scope and limitations ......................................................................... 2 1.4 Experimental equipment .................................................................... 3 1.5 Outline of thesis ................................................................................. 4

2 TIG welding theory 5 2.1 Principle of TIG welding.................................................................... 5 2.2 Process parameters in TIG welding..................................................... 7 2.3 Heat effects of welding ....................................................................... 8 2.3.1 Temperature fields.................................................................. 9 2.3.2 Residual stresses and distortion ............................................... 15

3 Modelling Techniques 17 3.1 General principles of off-line programming of robots ......................... 17 3.2 Off-line programming in the present study......................................... 19 3.3 General principles of finite element modelling of welding................... 20 3.3.1 Boundary conditions .............................................................. 21 3.3.2 Material properties.................................................................. 21 3.4 FEM-modelling in the present study .................................................. 22 3.4.1 Boundary conditions .............................................................. 23 3.4.2 Material properties.................................................................. 25 3.4.3 Properties for the thermal-mechanical modelling .................... 27 3.5 Principle of the integration between the off-line programming

model and the finite element analysis model ....................................... 27

VIII

4 Model validation techniques 31 4.1 OLP calibration.................................................................................. 31 4.1.1 Signature calibration............................................................... 31 4.1.2 Tool calibration ...................................................................... 32 4.1.3 Work cell calibration .............................................................. 32 4.2 Temperature measurements techniques............................................... 33 4.2.1 Thermocouple instrumentation on plates................................ 33 4.2.2 Infrared imaging measurements techniques............................. 34 4.3 Residual stress measurements techniques ............................................ 35 4.4 Distortion measurements.................................................................... 36

5 Results 37

6 Summary and conclusion 39

7 Proposals for future work 41

References 43

Included papers 47

List of figures

Figure 1 Principle for Tungsten Inert Gas (TIG) welding........................... 6 Figure 2 Schematic diagram of the TIG process with its three parameter

groups .......................................................................................... 7 Figure 3 Weld characteristics for two butt-welded plates with thickness t ... 8 Figure 4 A typical cross section of a weld.................................................... 8 Figure 5 Schematic of the welding thermal model ...................................... 11 Figure 6 Temperature contour plots with different welding parameters.

Upper left: welding speed 2.0 mm/s, Upper right: welding speed 3.0 mm/s. Lower left: welding current 100 A, Lower right 80 A ............................................................................................. 13

Figure 7 Important temperature characteristics........................................... 13 Figure 8 Example of distortion that can occur during welding [2] .............. 16 Figure 9 IGRIP model of the experimental setup........................................ 19 Figure 10 Aerospace component, whole part left, 1/13 of the part to the

right ............................................................................................. 20 Figure 11 The non-uniform mesh used in paper one. Note the higher

densities along the weld path ........................................................ 22 Figure 12 Shell model of aerospace component. Note the higher densities

along the weld path ...................................................................... 22 Figure 13 a) Cross section of a plate mounted in a welding fixture. b)

Applied boundary conditions in a heat transfer simulation ........... 23 Figure 14 Heat flux Gaussian distribution with 5% cut off limit .................. 25 Figure 15 Specific heat for Stainless Steel 361L ............................................ 26 Figure 16 Conductivity for Stainless Steel 316L. Conductivity without

considering convection (a) and conductivity when weld pool convection is considered by increasing the conductivity value above the melting point (b) .......................................................... 26

Figure 17 Block chart showing the integration between OLP and FEM ....... 28 Figure 18 Robot pose description for a path ................................................. 28 Figure 19 Input file to the FEA simulation generated in the Robot

simulation program ...................................................................... 29 Figure 20 Overview of a cross section from a FEA simulation showing the

penetration. The color represents a temperature interval close to the melting point.......................................................................... 29

X

Figure 21 Schematic of a plate with thermocouples together with selected measurement line for the IR camera measurements....................... 34

Figure 22 Principle overview of the VarioScan 3021 high resolution ............ 35 Figure 23 Distorted welded plate measured in a CMM Machine. Results

post processed in UniGraphics...................................................... 36

List of tables

Table 1 Boundary conditions in the heat transfer analysis in paper I.......... 24 Table 2 Material properties for Stainless Steel 316L and Greek Ascaloy..... 25

List of acronyms

Acronyms Explanation

AC Alternating Current

CAR Computer Aided Robotics

DC Direct Current

FEA Finite Element Analysis

FEM Finite Element Method

GTAW Gas Tungsten Arc Welding

HAZ Heat Affected Zone

OLP Off-Line Programming

TCP Tool Centre Point

TIG Tungsten Inert Gas

1

Introduction

1.1 Background and motivation

Arc welding equipment was originally designed to be used manually but during the industrial evolution and through the introduction of robots in industry in the 1970s, automatic welding was developed. It is today one of the most common tasks for an industrial robot. Examples of some of the driving forces for this automation are higher productivity, higher quality demands, and an increased demand of higher flexibility. Automatic TIG (Tungsten Inert Gas) welding is, however, still rather rare since it puts high demands on equipment and on part geometry accuracy. Manual TIG–welding is, on the contrary, one of the most common welding processes in the aircraft industry. This is due to high product requirements for materials with high heat and corrosion resistance, with good fatigue properties and with low weight. Examples of materials are Inconel 718 and Greek Ascaloy, which can be successfully joined by TIG welding resulting in joints with few defects and, comparative to other welding processes, low distortion.

However, any welding process induces changes in the base material and generates unwanted stresses and deformation due to the heat input. The most common way to avoid this deformation is to use fixtures to clamp the part to be welded. Unfortunately these fixtures are difficult to design, time consuming to construct and very expensive. Another method to reduce deformation is to optimise the welding sequence to allow a more uniform heat distribution into the part. An optimal welding sequence can be hard to find and requires a very skilled operator. Therefore, a simulation tool that can be used to evaluate fixture solutions and to plan welding sequences early in the product development stage would be desirable. Such a simulation tool would reduce both the number of welding experiments and the need for welding operator experience. The tool should preferably be able to simulate the welding torch path, be capable of detecting collisions between the torch and workpiece, and of optimising the welding parameters with respect to penetration and component deformation.

2 Introduction

Research in arc welding simulation can be divided into two main fields, namely robot simulation, often refereed to as CAR (Computer Aided Robotics), and thermal-mechanical modelling [4, 8]. CAR concerns simulation and programming of a robot task using a virtual model of the workcell and the part to be welded. Examples of research in this area are the integration and development of virtual sensors and the optimisation of welding sequences and torch trajectories to avoid collisions and to increase productivity [5, 6].

Thermal-mechanical modelling concerns the modelling of the influence of the process on the component. It includes the prediction of temperature histories, microstructure phase transformations, residual stresses and distortion.

This thesis addresses both these areas, i.e., both CAR and thermal-mechanical simulations, through the integration of an off-line programming system with a Finite Element Analysis (FEA) system. CAR is used to simulate TIG welding torch paths and to detect collisions between the torch and workpiece. FEA is used for the prediction of temperature histories and residual stresses and the optimisation of welding parameters as regards penetration.

The industrial interest in manufacturing simulation tools has increased significantly in recent years, which is why simulation has become an increasingly common tool to test and verify different approaches prior to manufacturing. A simulation tool such as described would therefore be of great benefit to the industry.

1.2 Objectives

The objective of this research is to develop and validate a simulation tool for the TIG welding process. The tool shall be capable of simulating torch paths, predicting temperature histories, residual stresses, and deformation, thus making it possible to optimise welding sequences and fixture solution prior to manufacture. Of particular interest is whether or not sufficiently complex models can be developed, that can be used industrially in the design and production engineering phases.

1.3 Scope and limitations

The simulation of welding is a very wide field, which incorporates several techniques and disciplines. Models have been developed for the simulation of the robot path, the arc, the liquid pool, and for the solid part. The different models involve disciplines such as plasma physics, electromagnetics, fluid mechanics, material science and production technology. The different models also impose different demands concerning time and space resolutions. Time scales can range from microseconds to minutes and length scales from micrometers up to decimetres and meters. Limitations are therefore necessary when a simulation model is to be developed. The limitations in this work are:

1.4 Experimental equipment 3

• No model of the arc or the molten pool. Convective heat transfer within the molten pool is considered by increasing the heat conductivity when the temperature increases beyond the melting point.

• No models of fixtures. Reaction forces from these are considered in the boundary conditions.

• No development of material models, such as phase transformations.

• The study is limited to the TIG welding process, although results are expected to be generic and applicable to other welding processes.

• The study is limited to two materials, namely Greek Ascoloy and stainless steel 316L. The methodology developed is, however, not material dependent.

Focus is placed on CAR and the integration of CAR with FEA. The thermo-mechanical simulations have been made in collaboration with Luleå Technical University, Sweden [14]. In the validation work focus is placed on temperature measurements.

1.4 Experimental equipment

Different experimental equipment and software has been used in this research. The TIG welding experiments were carried out with a robotised welding cell consisting of a six-axis robot (ABB IRB 1400) supplied by ABB Automation Technology Products AB Robotics, Västerås Sweden, linked with a torch produced by Binzel AB (thoriated tungsten electrodes) and supplied by Abicor Binzel AB, Karlskrona, Sweden. The power source is a TIG Commander 400 AC/DC produced by Migatronic AB.

The robot simulations were performed using IGRIP, commercial software produced by Dassault Systemes, Suresnes Cedex, France. The thermomechanical simulations were performed using the Marc system developed MSC Software Corporation, Santa Ana, USA. The calculations were performed using an in-house developed Linux cluster consisting of ten 1.0 GHz Pentium III processors.

For temperature measurements, both thermocouples (type K) and infrared imaging were used. A PC based data acquisition system was used to sample the signals from the thermocouples and write the data to disk. The infrared camera is a VarioScan 3021-ST high resolution 16 bit stirling cooled camera produced by Jenoptic GMbH, Jena, Germany.

1.5 Outline of thesis

In this thesis, a method for the off-line optimisation of welding by the use of simulation is proposed. The thesis commences in chapter 2 with an introduction to

4 Introduction

the theory of TIG welding and its heat effects, such as residual stresses and distortion. Chapter 3 describes methods for robot simulation as well as modelling techniques for the prediction of temperature histories, residual stresses and distortion. In chapter 4 different methods for the validation of temperature, residual stresses and distortion are discussed. Chapter 4 is followed by the results and discussions in chapter 5, and by the conclusions in chapter 6. Finally, the thesis ends with a proposal for future research. Papers published by the author are listed in the appendix.

2

TIG welding theory

This chapter provides a brief presentation of the theory of TIG welding and its heat effects on the base material.

2.1 Principle of TIG welding

The TIG welding process was invented during the Second World War due to the need of the American aircraft industry for a method of joining magnesium and aluminium. Russell Meredith [7] demonstrated the first TIG process for the welding of magnesium using a Tungsten electrode and helium gas in the late 1930´s. TIG welding or GTAW (Gas Tungsten Arc Welding which is the common name in North America) uses a non-consumable tungsten electrode protected by an inert gas. The electrode is either made of pure tungsten or tungsten, mixed with small amounts of oxides (thoriumoxide, zirconiumoxide) improving the stability of the arc and makes it easier to strike. Since the process uses a non-consumable electrode, extra filler material is usually added. The principle of the process is schematically presented in Figure 1.

The electrical discharge generates a plasma arc between the electrode tip and the work piece to be welded. The arc is normally initialised with a power source with a high frequency generator, which produces a small spark that provides the initial conducting path through the air for the low voltage welding current. The frequency of this ignition pulse is large, up to several MHz. This frequency, together with a high voltage (several kV), produces strong electrical interference around the welding cell, which is a disadvantage when sensors and measuring equipment are used. The arc consists of a high-temperature conducting plasma that produces the thermal energy needed to melt the base and the filler material. The arc temperature spans between 12000 K and 15000 K above the pool surface and

6 TIG Welding Theory

the temperature of the melted surface spans from 1700 K to 2500 K, dependent on the material.

Weld Center Line

Arc

Gas Shield

Tungsten Electrod

Shieldinggas

CurrentConductor

Filler Metal

SolidifiedWeld Metal

Weld Pool

Figure 1: Principle of Tungsten Inert Gas (TIG) welding.

Three different alternatives of current can be used namely; direct current (DC) with a positive electrode, DC with a negative electrode or alternative current (AC). AC is mainly used for the welding of aluminium and magnesium since cleaning of the oxide layer on the surface can in this way be achieved. DC with a negative electrode is used for most other materials, including thick plates of aluminium. Pulsed and non-pulsed currents can be used. A non-pulsed current is most common. The use of a pulsed current has some advantages, such as increased penetration.

Depending on the thickness of the base material, type of joint and certain other factors, extra filler material might be needed. In automatic TIG welding hot or cold wire can then be used. Cold wire is fed in the front of the melted pool and hot wire fed in the back. The filler material is usually the same as the base material.

An inert gas is used to sustain the arc and to protect the melted pool and the electrode from atmospheric contamination. Depending on the welding parameters and welding materials, either argon, helium or a mix of the two gases can be used.

2.2 Process parameters in TIG welding 7

Argon is commonly used in welding unalloyed, low alloyed and stainless steels. However, a mixture of argon and hydrogen or helium can be used for mechanical welding. For duplex stainless steel, it is common to mix argon with nitrogen to ensure a correct ferrite/austenite balance. Aluminium and aluminium alloys are usually welded using argon. However, the addition of helium can be used to improve the heat transfer and is therefore sometimes used for the welding of thicker parts. Argon is suitable for welding copper and its alloys, and gives excellent results for thicknesses up to 6 mm. Helium, or a mixture of helium and argon (up to 35 %), are suitable for thicknesses greater then 6 mm. Titanium requires an extremely high purity of the shielding gas, usually not less then 99.99 %. Either argon or helium can be used here. Argon is the more common shielding gas for thicknesses less than 3 mm while helium is more commonly used for thicknesses in excess of 3 mm. In stainless steel and other easily oxidised materials, applications of a root gas can be used to protect the root side of the weld from oxidation. The root gas can be a mixture of nitrogen and hydrogen, or pure argon.

2.2 Process parameters in TIG welding

Three main parameter groups can be defined in the TIG welding process. The first group (group 1 in figure 2) is the group of controllable process parameters. The second group consists of sensor variables for the supervision and control of the process, and the third comprises final weld characteristics. Group one can be divided in three sub groups; those that can be varied on-line during the process (such as arc-current, torch travel speed and wire feed speed), those that are set prior to the process (for example composition and flow rate of the shielding gas) and finally the last subgroup that consists of variables that cannot be modified, such as part geometry [15].

Filler material W-electrode

Weld pool

Inert gas

- Bead geometry- Indications (NDT)- Distortion

1.PARAMETERS:

- Current- Travel speed- Arc length- Wire feed speed

2.SENSOR VARIABLES:- Pool geometry- Temperature- NDT 3.

WELD CHARACTERISTICS:

Figure 2: Schematic diagram of the TIG process with its three parameter groups.


Examples of sensor variables of group two are weld bead width and weld pool geometry. Another example is seam tracking which is commonly used in robotised welding.

The last parameter group, weld characteristics, is strongly dependent on the other two parameter groups. To this category belong weld geometry, metallurgy phase composition, residual stresses and distortion. Figures 3 and 4 show a schematic and a real cross-section for two butt-welded plates. The most important geometry characteristics are: weld width at topside ( )B , weld width at root side( )C , weld height at topside hB (so called reinforcement) and weld height at root side hC (usually called drop through) [17].

Bh

ChC

B

t

Figure 3: Weld characteristics for two butt-welded plates with thickness t.

Figure 4: A typical cross section of a weld.

2.3 Heat effects of welding

Heat transfer phenomena play an important role in welding. Heat effects of welding refer to temperature fields, residual stresses and distortion that occur

2.3.1 Temperature fields 9

during or after welding. Since the focus of this thesis is placed on temperature prediction, a thorough description of the temperature fields is given below.

2.3.1 Temperature fields

One objective with heat transfer analysis in welding applications is to determine the temperature fields in an object resulting from conditions imposed on its boundaries [18]. The quantity that is sought is the temperature distribution, which represents how temperature varies within positions in the object. When this distribution is known, the conduction heat flux calculated at any point in the medium or at the surface may be computed from Fourier’s law.

The temperature fields during welding are highly heterogeneous and transient. The temperature of a component can vary from below zero to 3000 centigrade, i.e. the evaporation temperature of the metal. Within this range phase changes, micro structural transformations and thermal strains take place, all of which determine residual stresses and distortion. Fourier’s law of heat conduction describes the heat propagation in mechanism in the solid material. The law states that the heat flow

density q

2J mm is proportional to the negative temperature gradient ∂∂Tn

[ ]K mm by equation

λ ∂= −∂T

qn

(2.1)

where λ [ ]J mmsK denotes the thermal conductivity and T [ ]K the temperature. Consider a homogenous medium expressed in one dimension x with a temperature distribution ( )T x expressed in Cartesian coordinates with infinitesimally small control volume dx . The condition heat rate at the control area can then be expressed as xq . The condition heat rate at the opposite surface can be expressed as a Taylor series expansion neglecting higher order terms as

xx dx

qq q dx

x+∂

= +∂

(2.2)

Inside the control medium an energy source term (2.3) and an energy storage term (2.4) can be expressed as

gE qdx= (2.3)

st pT

E C dxt

ρ ∂=∂

(2.4)

Using the law of energy conservation equations 2.1, 2.3 and 2.4 can be substituted to


x x dx pT

q qdx q C dxt

ρ−∂+ − =∂ (2.5)

Substitution from equation 2.2 and Fourier’s law the heat diffusion equation can be written in a general form, in three dimensional Cartesian coordinates as,

( ) ( ) ( ) ( )pT T T T

T T T C T Qx x y y z z t

λ λ λ ρ ∂ ∂ ∂ ∂ ∂ ∂ ∂ + + = − ∂ ∂ ∂ ∂ ∂ ∂ ∂

(2.6)

where the Cartesian coordinates x , y and z denote the welding direction, the transverse direction, and the normal direction to the weld melt surface,

respectively, see figure 5. Q 3W mm stands for internal heat generation rate

and the material properties, thermal conductivity, density, and specific heat, are denoted by λ , ρ and pC respectively.

Several possibilities for initial conditions exist. The most common is:

0T T= at 0t =

A general boundary condition can be written as:

( ) 0x y zT T T

l l l q h T Tx y z

λ λ λ ∞∂ ∂ ∂+ + − + − =∂ ∂ ∂

(2.7)

where h denotes surface heat loss coefficient, xl , yl and zl the direction cosines

to the boundary surface. The surface temperature and environment temperature are denoted T and T∞ respectively.

The heat diffusion equation (2.6) can be solved both analytically and numerically (in the latter case, the FEM is commonly used, which is further presented in section 3.2). The equation can be analytically solved assuming the following conditions [2]:

• The energy from the welding heat source is applied at a uniform rate.

• The heat source is moving with a constant speed.

• The cross section of the work piece is constant.

• Constant material properties are used.

• The end effects resulting from the initiation and termination of the arc weld are neglected (quasi-static solution).

Different analytical solutions exist depending on the plate thickness and welding positions. The plate can be assumed to be thick, thin and finite, respectively. The dimensionless τ is used to determine whether the plate is to be considered as thin


or thick. In a thick plate the heat flow is considered to be three dimensional, through the plate thickness and lateral from the weld. The thin plate equation can be applied where the heat flow is essentially lateral. This means that the difference in temperature between the bottom and top surface is small in comparison to the melting temperature.

( )0p c

net

C T Th

H

ρτ

−= (2.8)

where h denotes the plate thickness. netH is net energy input equal to υ

ηEI. 0T

stands for the initial plate temperature, cT denotes the temperature at which the cooling rate is calculated. The plate is considered to be thin if τ is less the 0.75 and to be thick if τ is larger than 0.75.

For an analytical quasi-static solution of the heat transfer model it is assumed that the material properties are independent of the temperature, that the metallurgy zones are homogenous and that the thermal model is linear in the welding direction. The solution gives the temperature in a specific point if the welding speed ( )υ , energy heat input ( E , I ,η ) and the material properties ( ρ , λ , pC )

are known. This point is defined by r (2.9):

2 2 2r y zξ= + + (2.9)

where ξ denotes a moving coordinate (2.10),

x tξ υ= − (2.10)

Here the origin of the moving coordinates (ξ , y , z , se figure 5) is fixed at the centre of the heat source see figure 5. This means that the coordinates move with the heat source at the same speed. Solutions are usually derived for the thin and thick plate separately.

Welding direction

X

YZ

Weld Centre Line

Figure 5: Schematic of the welding thermal model.


Analytical solutions were first presented by D. Rosenthal 1935 [1, 2 and 3].

( )0 2 2

v rqT T e

r

ξπλ κ

− + − =

(2.11)

for a thick plate and

20 02 2

dq vrT T e K

d d

νξ

πλ

− − =

(2.12)

for a plate considered to be thin, where κ denotes the thermal diffusivity of the base metal, 0T denotes the preheat temperature in the base metal, d stands for

plate thickness and 0K denotes the modified Bessel function of the second kind, zero order. This relationship for the temperature heat flow is not accurate close to the welding arc. Since a point or a line source is assumed for thick and thin plates respectively, singularities will occur at the sources location where the temperature tends to infinity.

Figure 6 shows the influence of welding parameters on temperature. The welding speed and welding current have been varied and the temperature distribution has been solved using equation 2.12. In the upper two figures the welding speed has been varied and in the lower the welding current. It can be seen that both the welding speed and current have a strong influence on the heat distribution. Several welding defects, such as residual stresses and distortion (see section 2.3) are dependent on the heat input. If the heat input can be minimised, maintaining a full penetration, these defects will decrease.

The thermal condition in, and close to, the weld is very important since it controls the metallurgical events in the weld. Interesting parameters to control are; the distribution of peak temperature in the heat affected zone (HAZ), cooling rates in the weld metal and in the HAZ, and the solidification rate of the weld metal, figure 7.


−0.4 −0.2 0 0.1

−0.03

0

0.03400

400

800 12

0017

00

−0.4 −0.2 0 0.1

−0.03

0

0.03

400

400

800 12

0017

00

−0.4 −0.2 0 0.1

−0.03

0

0.03400

400

800 12

0017

00

−0.4 −0.2 0 0.1

−0.03

0

0.03400

400

800 12

0017

00

Figure 6: Temperature contour plots with different welding parameters. Upper left: welding speed 2.0 mm/s, Upper right: welding speed 3.0 mm/s. Lower left: welding current 100 A, Lower right 80 A.

Peak temperature

Cooling rate

Time (s)

Te

mp

era

ture

(K

)

Figure 7: Important temperature characteristics.


The peak temperature in the weld pool is given by [2]

00

121

TTH

tYCe

TT mnet

p

p −+=

−ρπ

(2.13)

where peak temperature, base metal melting temperature, initial base temperature are denoted by pT , mT , and 0T , respectively. The material properties, density,

and specific heat are denoted by ρ , and pC . Y denotes the distance from the

weld fusion boundary where the peak temperature is calculated, t states the plate thickness and e denotes the base of the natural logarithm. netH is net energy

input equal to υ

ηEI. This equation can be used to predict peak temperature in a

specific point in the HAZ, the width of the HAZ, as well as the effect of preheat on the width of the HAZ.

If the cooling rate in a specific point along the weld line is known, a prediction of the metallurgy in the welded area can be made. Cooling rates are important in the welding of heat-treatable steels. This is due to the formation of martensite in the welded area. In the case of carbon and low alloy steels, the temperature at which the cooling rate is calculated is not critical. Therefore, the major use of cooling rates is to calculate the preheating temperature [2]. The general cooling rate equation can be defined, using the moving coordinate ξ (2.9), which gives

νξ −=∂∂

t. Using the chain rule the cooling rate equation can be written as [3]

ξν

∂∂−=

∂∂ T

t

T (2.14)

An analytical solution of the cooling rate can be defined for both a thick (2.14) and a thin plate (2.15), respectively,

net

c

H

TTR

30)(2 −= πλ

(2.14)

( )30

2

2 TTH

tCR c

netp −

= πλρ (2.15)

where R is the cooling rate at a point at the weld centreline just at the moment when the point is cooled past the cT temperature. cT denotes the critical temperature for phase changes in the welded metal. The material properties,

2.3.2 Residual stresses and distortion 15

thermal conductivity, density, and specific heat are denoted by λ , ρ , and pC .

netH is net energy input equal to υ

ηEI.

The solidification rate can have an important impact on the metallurgical structure, properties, and material response to heat treatment. The solidification time tS of the weld metal, measured in seconds is given by

( )2

02 TTC

LHS

mp

nett

−=

πλρ (2.16)

where L is the heat of fusion.

2.3.2 Residual stresses and distortion

Residual stresses are self balanced internal stresses, which exist in the component without any external loads and can be classified as macro stresses and micro stresses [16]. The definition of macro stresses is that they are self-equilibrated in a cross section of the manufactured part. Micro stresses can be defined as stresses that are homogenous or inhomogeneous in a micro scale [16]. They are introduced in the component as results of manufacturing process such as welding and drilling, generated either on purpose or not, as the case may be.

Since the welding process heats the material locally, the temperature distribution is not uniform. In the melted weld pool stresses are released and can be assumed to zero. During the solidification of the melted weld pool the metal starts to shrink and to exert stresses on the surrounding weld metal and HAZ. These stresses remain in the material after welding and result in unwanted distortion. A typical example of distortion is given in figure 8. The stress level in the solidification area is proportionately low, but the stress level in the weld area increases and can be as high as the yield limit of the base material, which can cause unwanted fractures. Stresses in a welded plate are usually divided in two directions, transverse and longitudinal to the weld.

Longitudinal residual stresses can arise from different causes. The most common cause is the longitudinal contraction of the weld as it cools down. Another cause is superimposed by opposing transformation processes. Transverse residual stresses are generated by the transverse contraction of the weld during the cooling phase. It can also be generated indirectly due to the longitudinal contraction [10].

Three different types of residual stress induced distortion can be found in manufactured structures, figure 8. Longitudinal and transverse shrinkage can cause in plane distortion of the workpiece. Plane or axisymmetrical angular shrinkage can cause distortion perpendicular to the plane of the welded component. Another distortion is bending due to grids with longitudinal and transverse welds [2].


Figure 8: Example of distortion that can occur during welding [2]

3

Modelling techniques

This chapter describes the modelling techniques used in this thesis. A presentation of the off-line programming of robots is provided, followed by a presentation of temperature and residual stress prediction by the use of the Finite Element Method.

3.1 General principles of the off-line programming of robots

Computer Aided Robotics is a graphical tool for manufacturing simulation that can be used in several applications such as the off-line programming of robots, tele-operation of robots and simulation of general kinematic systems. The off-line programming of robots using a graphical simulation tool was first demonstrated in the beginning of the 1980. Developments in computer technology have significantly improved this technology. The simulation of the production of a component makes it possible to test, verify and optimise robot motions and to design fixtures and automation equipment before the real production process begins [25, 26]. Accessibility, collisions and timing can be verified before expensive machine tools and robots are purchased. Further, production accuracy can be improved by using off-line programming of robots, thus avoiding the complexity of manually programming a robot in three dimensions with a high degree of accuracy. The most common method to manually program a robot is still that an operator jogs the robot arm to each coordinate pose in space, a procedure which is error prone and depends heavily on the experience of the operator. Three-dimensional computer graphics tool for production planning makes it possible to achieve high accuracy for complex parts.


The technique of off-line programming is commonly used in industry. Several commercial software packages exist such as IGRIP, RobCad and Grasp. All of these systems have principally the same work order [8, 25]:

1. Modelling of the work cell 2. Work cell calibration 3. Programming of robot and other optional work cell equipment 4. Down loading of the program to the control system 5. Additional robot programming 6. Test running

In the first step the geometrical model of the work-piece, together with a geometric and a kinematic model of the workcell including the most important parts such as fixtures, robots and positioners are created. The geometrical model can either be created in a CAD/CAM software and imported to the robot simulation software or directly created in the robot simulation software. Most of the robots on the market have been predefined in the robot simulation softwares and can be retrieved from a library. If a new design of a robot or a new kinematic device is to be used, a kinematic model has to be created. The different parts of the kinematic devices, such as robot arms, have then to be modelled. These devices are usually successively created from the base part to the tool tip by connecting the parts in joints and describing movement patterns and boundary conditions for each joint.

The second step is the calibration of the workcell. It can include several sub-steps, such as TCP- (Tool Centre Point), workpiece- and signature calibration. A tool calibration is usually performed by rotating the real robot TCP around a sharp fixed point. Each new position is then stored and uploaded into the robot simulation program and a “best fit” is calculated using statistical regression. Similarly the position of the workpiece is calibrated by moving the robot to identified positions. To find errors in the geometrical model of the robot, an arm signature calibration can be used. This calibration finds the deviations between the physical and virtual model in the lengths between the robot joints and in the zero poses for the joints. Corrections to the virtual model are then made. A more detailed description of calibration is given in the OLP modelling validation section in chapter 4.

In the third step the programs for all the different devices are written. Three categories of commands are commonly used. The first category includes commands for the visualisation of the simulation. Examples of such commands are graphical commands such as viewpoint rotation and zooming towards or outwards from an object. These commands are not included in the robot code and consequently not downloaded to the robot.

The second category concerns commands that are used in the simulation but also included in the robot code. Movement commands for the robot are typical

3.2 Off-line programming in the present study 19

examples of this category. For a welding application, this category will also include commands such as ignition and termination of the arc.

The last category concerns commands that are not used in the simulation but which are directly downloaded to the robot. Examples of commands in this category are I/O´s, such as a gas protection set to on or off in welding.

Step four is to translate the simulation program to the specific robot code and to download it to the real robot control system. Step five concerns additional robot programming and adjustments that have to be performed on-line at the robot. In step 6 a test run of the program is performed. If the test result is satisfactory the production can be executed. Steps 1-6 might however have to be repeated iteratively until the required accuracy is achieved. Accuracy does not only depend on how well steps 1-6 have been performed but also on how well the simulation control system emulates the physical control system. A special module called RRS (realistic robot simulation) [19] can be used to increase the agreement between the physical and the virtual control system. Through RRS the original control system software for motion interpolation and transformation is integrated in the CAR system.

3.2 Off-line programming in the present study

The method to program robots off-line described above has been used in this study. An in house welding cell, see figure 9, was modelled in IGRIP (Interactive Graphics Robot Instruction Program, Deneb Robotics). Figure 9 shows a snapshot of the experimental setup.

Figure 9: IGRIP model of the experimental setup.


Both plane plates and a component with complex geometry were modelled and used as test cases. A tool calibration and a workpiece calibration were both performed. A high level graphical simulation language in IGRIP, GSL, was used for programming all devices in the cell. The part was a section of an aerospace part, a turbine component from the V2500 engine produced by Volvo Aero Corporation, figure 10, left. 1/13 was cut out of the real part, which originally consisted of an inner and an outer ring and 13 vanes, figure 10, right. All components were created in the UniGraphics CAD/CAM system and imported using both a direct translator and the neutral interface IGES. No RRS model was used.

Figure 10: Aerospace component, whole part left, 1 13 of the part right.

3.3 General principles of finite element modelling of welding

Numerical methods have been used since the beginning of the 1970’s to simulate welding processes. The focus has been to predict thermal histories, residual stresses and distortion before the real welding process begins. FEA simulations have been the most common numerical method and many papers have been presented [12, 14, 27, 28]. Large complex simulation models of three-dimensional components are, however, still rare, mainly due to the lack of computational power. One reason is that to be able to compute the temperature and residual stress fields in the affected zone a very fine discretization of the space variable is required.

Another complexity concerns the heat transfer between the electrode and the part to be welded [20]. This is a complex phenomenon with several interaction effects. The phenomenon can be divided into three different groups, the plasma arc, the weld pool and the solid material. Modelling the plasma arc is complicated since chemical reactions, ionizations, and vaporisations of both electrodes and the surface of the weld pool have to be considered. Developing a model of the weld pool is also complicated since, driven by various forces such as surface tension forces, electromagnetic and buoyancy forces the melted material undergoes vigorous circulation. The weld pool surface is also strongly influenced by a drag force that

Inner Ring

Vane

Outer Ring

Vane

Inner Ring

b)

a) Outer Ring

3.3.1 Boundary conditions 21

depresses the surface and induces a surface flow. The solidification process in the liquid–solid boundary region is also complicated to model.

Different assumptions and simplifications have therefore to be considered when building an FEA model of welding of a complex part. Examples of areas that have to be considered are part geometry simplifications, the type of material models to be used, load conditions, heat transfer and other boundary conditions and numerical strategy [14]. The most important considerations that have to be made are described in the following section.

3.3.1 Boundary conditions

All FEA problems are defined in terms of initial and boundary conditions. A typical type of initial condition for a welding application is the initial temperature that, in most cases, is set to room temperature. Examples of the most important boundary conditions are fixture forces, and heat transfer coefficients between the part and its surroundings.

Since the heat transfer between the electrode and the part to be welded is usually too complex to be integrated in the same model, an ad hoc heat source with parameters that are adjusted. Different types of heat sources have with this purpose in mind, been suggested [9, 10]. Three main methods are usually used. The first method is to apply an energy source within the part to be welded. The size and energy density are then adjusted to retrieve a fusion zone with good agreement with the real weld. The second method is to use a surface distribution to simulate the arc. Here the energy source heat flux depends on the distance from the centre of the arc. A very common distribution used is the Gaussian method. The last method is to use a double ellipsoid heat source, which was first recommended by Goldak [9]. This method combines the first two types since it includes a surface distribution as well as a heat source within the material. Although this method is the most realistic it needs more parameters to be fitted with experiments.

3.3.2 Material properties

The material properties that have to be included in the proposed simplified model when temperature simulations are to be performed are specific heat, heat conductivity, density, liquidus and solidus temperature. The conductivity is usually temperature dependent. Weld pool convection is a complex phenomenon that is difficult to simulate. This convection is therefore simulated by multiplying the thermal conductivity with a certain factor when the temperature exceeds the liquidus temperature. Different factors have been proposed in the literature. Common values are eight and ten. The specific heat for most welding materials are strongly temperature dependent. The value increases significantly during fusion due to the latent heat of transformation.


3.4 FEM-modelling in the present study

The welding paths, including initial weld velocities, were exported from the CAR model to the finite element software where predictions of temperature histories, residual stresses and fixture reaction forces were performed. The same CAD/CAM model as in the CAR model was imported to and meshed with the FEA software. The commercial FEA program MARC, from MSC Software, was used. In paper I a model with solid elements was created. Since this type of model is computationally very expensive, a small part of the component had to be selected for modelling. The model was divided by a non-uniform mesh with higher densities close to the weld path (where the highest temperature gradients were assumed to occur), figure 11. Eight-node brick elements were used. In paper III a shell model of the whole section of the aerospace part was created, see figure 12.

Figure 11: The non-uniform mesh used in paper one. Note the higher densities along the weld path

Flange

Weld path

Flange

Weld path

Figure 12: Shell model of aerospace component. Note the higher densities along the weld path

weld path

3.4.1 Boundary conditions 23

3.4.1 Boundary conditions

Different boundary types have been used. The boundary conditions used in paper I are given in table 1. Since the model had to be restricted to a subsection of the part, a “metal-metal” boundary condition was introduced which models the heat conduction through these interfaces, table 1. Different heat transfer coefficients were used for the surfaces with natural and forced convection i.e. forced convection on surfaces where root gas (see section 2.1) was applied. A typical set up for a plate is illustrated in figure 13 a. The applied boundary conditions for the plate are shown in figure 13 b. A “metal-meal” convection boundary condition which models the clamping of the plate to the fixture is used in group 1. Natural convection is used for group 2 and forced convection is used for group 3. The heat input from the arc, group 4 simulates the heat source.

12 4 2 1

22

1 3 1

Weld poolPlate

Fixture Weld gun

Arc

Weld pool

Shielding gas

Air

Plate

Air

Air Air

Fixture

a)

b)

Figure 13: a) Cross section of a plane plate mounted in a welding fixture. b) Applied boundary condition in a heat transfer simulation.

Natural convection was only used as heat transfer boundary condition between the part and the surrounding environment in paper III. The flanges on the inner- and outer- rings were assumed to be clamped since no fixture was used (the inner- and outer ring were welded on a steel plate). This was simulated by locking the FEM elements in all directions.


Table 1: Boundary conditions in the heat transfer analysis in paper I

Type of condition Interface

Value

( )2W

mm K

Corresponding number in figure 11b

Face film Metal – Metal 61000 10−⋅ 1

Face film Metal – Air 620 10−⋅ 2

Face film Metal – Gas (Argon as root gas)

6200 10−⋅ 3

As a heat source, a Gaussian surface distribution was used in all of the simulations. This distribution was selected since it requires fewer parameters to be calibrated and since the plates that were welded could be considered as to be thin. User subroutines had to be developed to simulate the moving heat source. The heat flux was expressed according to [10]

2

2

0

0

q

q

r

q

rq

q q e

EIq

EIq e

α

α

η απ

η απ

−

−

= ⋅

=

=

(3.1)

where q denotes the heat transferred to the workpiece, E the voltage, I the

current, η the efficiency factor, qα the concentration factor, and r the radial

distance from the centre of the heat source. Figure 14 shows a typical Gaussian heat flux distribution with a 5 % cut off limit. This means that the distribution is truncated when the heat flux reaches five percent of the maximum heat flux permitted i.e. maxmin 05.0 qq ⋅= . This truncation was proposed by Radaj [10] and was used in both paper I and paper III.

3.4.2 Material properties 25

−5 0 5

5

10

15

20

25

Radial Position (mm)

He

at

flux

(W/m

m2)

Figure 14: Heat flux Gaussian distribution with 5% cut off limit.

3.4.2 Material properties

Two stainless steels, namely Greek Ascaloy and SS316L, have been used. Temperature dependent properties were used for thermal conductivity and specific heat. The properties were taken from [11, 12, 13] and are given in table 2.

Table 2: Material properties for Stainless Steel 316L and Greek Ascaloy

Nomenclature Symbol SS 316L Greek Ascaloy

Density ρ -6 7.3 10 -3kg mm -6 7.3 10 -3kg mm

Latent heat of fusion H -52.47 10 J/kg -52.47 10 J/kg

Solidus temperature

solT 1673 K 1673 K

Liquidus temperature

liqT 1523 K 1523 K

Heat Capacity pC See chapter 3.4.2

Thermal conductivity

λ See chapter 3.4.2


The specific heat for Greek Ascaloy and 316L are strongly temperature dependent. The values used for 316L are presented in figure 12 [13].

500 1000 1500 2000 2500 3000400

450

500

550

600

650

700

750

Cp (

J/(k

g o C

))

Temperature ( oC)

Figure 15: Specific heat for Stainless Steel 316L

Simulations were performed with and without the consideration of weld pool convection. Figure 10 presents conductivity values used for 316L [13]

500 1000 1500 2000 2500 30000

0.01

0.02

0.03

Temperature ( oC)

λ (W

/(m

m o C

))

500 1000 1500 2000 2500 30000

0.1

0.2

0.3

Temperature ( oC)

λ (W

/(m

m o C

))

Figure 16: Conductivity for Stainless Steel 316L. Conductivity without considering convection (a) and conductivity when weld pool convection is considered by increasing the conductivity value above the melting point (b).

3.4.3 Properties for the thermal-mechanical modelling 27

The fusion interval selected in this study is in accordance with analyses conducted by Toselo et al. [12].

3.4.3 Properties for the thermal-mechanical modelling

The Greek Ascaloy’s initial microstructure consists of a mixture of ferrite and pearlite. In the numerical model in paper III the ferrite/pearlite to austenite transformation was assumed to occur only if the highest temperature experienced by the material was greater then a limit temperature, see paper III for further details. A thermal-elastoplastic model based on von Mises’s theory was used [14]. It was assumed that no creep strains occur during welding since the material is exposed to a high temperature for a very short period of time. The hardening behaviour of the material was assumed to be isotropic and piecewise linear. Transformation plasticity was not accounted for in the model. The principles that underpin the thermal-mechanical modelling are further described in paper III.

3.5 Principle of the integration between the off-line programming model and the finite element analysis model

Since two different softwares were used in the Off-line programming and in the FEA work, an interface between the softwares had to be developed. Figure 17 shows a block chart of the work principle. The same part geometry was used in both softwares. The principle is that the part to be manufactured is created in a CAD/CAM software and then, using either a direct translator or a neutral interface such as IGES or Step, imported to the robot simulation program. Here the engineer plans the production, following the steps described in section 3.1 above. A robot path is generated which includes the welding parameters to be used. The welding program is then exported to the FEA program where the thermal history, residual stresses and distortion are predicted. An optimisation can also be performed to reach a full penetration weld with minimised distortion (further described beneath). If such an optimisation is performed a new weld parameter is generated by adjusting the robot speed. This new speed, together with the simulation program for the robot motion, is then translated to a complete robot code. The robot code is finally downloaded to the robot control system and a test weld can be performed.


Robotsimulation

Geometrye.g. IGES

CAD/CAM

FEATranslator

Weldingpath

Thermal historyResidual stresses

Simulationprogram for therobot motion

IRBController

Completerobot code

Full penentration weldwith low distortion

Weldvelocity (wv)wv

Figure 17: Block chart showing the integration between OLP and FEM.

The information exchange between the different softwares is based on the method of adding attributes to a pose, the same principle as in the ABB operative system S4. Examples of arguments are robot speed and welding data, such as welding current and welding speed. Figure 18 shows three robot poses used in a welding application. When the robot passes a pose it will use the arguments belonging to the next pose, for example when the robot passes pose P2 towards P3 the welding speed is increased to 3.0 mm/s.

P2 Vw=2.5 mm/s I=100 A

P1 Vw=2.0 mm/s I=100 A

P3 Vw=3.0 mm/s I=100 A

Figure 18: Robot pose description for a path.

The information exchange between the softwares is based on text files using pose coordinates (right-handed Cartesian coordinate system [23]) together with the attribute’s welding speed and welding current. Figure 7 shows an example of an input file to the FEA simulation generated in the OLP software. The columns denote x, y and –z coordinates, welding speed and welding current respectively.

3.5 Principle of the integration between the off-line programming model and the finite element analysis model 29

Figure 19: Input file to the FEA simulation generated by the Robot simulation program

If an optimization is to be performed each node in the finite element mesh along the welding path is considered as a welding pose. During the calculation, a text file is generated with node numbers and corresponding node temperatures for each time step. Figure 8 shows a typical temperature cross section of a welded plate. The nodes along the weld path at the top and bottom surfaces are marked with dots.

Figure 20: Overview of a cross section from an FEA simulation showing the penetration. The colour represents a temperature interval close to the melting point.

A stand-alone Matlab program was constructed that reads this file and suggests a new weld speed. This weld speed is then used in a new input file for a new simulation. The optimisation algorithm used is presented in equation 3.2 where the input welding speed is denoted by 0S . Here λ is a relaxation parameter, meltT

the liquidus temperature and maxT the maximum temperature at each node.

1 imax melti i

melt

T Tns s

Tλ

− = +

(3.2)

The weld speed is iteratively adjusted by the program until the temperature on the root side is sufficiently close to a target temperature. This target temperature is usually set to the liquidus temperature. When an optimal velocity vector i.e. the velocity vector that maximises the speed while keeping full penetration is found, the velocity vector is exported to the weld program.

-75.0 0.0 0.0 2.5 100.0 -45.1 0.0 0.0 2.5 100.0 0.0 0.0 0.0 3.0 100.0 45.5 0.0 0.0 3.5 80.0 75.0 0.0 0.0 2.5 100.0

4

Model validation techniques

This chapter describes validation techniques for the off-line programming of robots, for temperature predictions and for residual stress predictions. Since a central part of off-line programming is calibration, a more extensive description of this is presented first.

4.1 OLP calibration

Several processes of calibration have to be performed to increase the accuracy in a CAR application. A CAR model consists of several different components that have to be calibrated, such as the robot, fixtures, manipulators and external positioning equipment, see chapter 3.1. All these components have to be modelled with the required accuracy and positioned according to their locations in the real work cell. The calibration of the cell can be divided into three different groups, namely signature calibration, tool calibration and work cell calibration [8]. In this thesis tool calibrations and work cell calibrations have been performed.

4.1.1 Signature calibration

The purpose of signature calibration is to increase the accuracy in the robot arm’s kinematic chain, both for the real robot and for the modelled version. The signature calibration can be divided into three different levels, namely joint level calibration, calibration of the robot kinematic and non-kinematic parameter calibration. The selected level of calibration depends on the type of robot and the process. In a welding application with a low weight robot and, comparatively to other processes, low robot speed, the joint level calibration is the most important signature calibration method. In the joint level calibration the joint values of the physical robot are compared with the corresponding values for the robot model.

32 Model Validation Techniques

A common way to perform this calibration is to rotate the robot arm around a measuring stick located at several positions in the work cell. The tool tip is rotated around the stick with different joint configurations and the joint values stored in a robot program. The model and the real robot values are then compared.

4.1.2 Tool calibration

Probably the most important part of the calibration is the tool calibration. When such a calibration has been performed the robot arm, together with the tool tip, can be used as a measuring tool with a high degree of accuracy. The tool calibration is usually performed using a measuring stick with a sharp point that is positioned in the work cell. The tool tip of the robot is moved towards this point in different directions. When the tool tip makes contact with the edge of the stick the position is stored in a robot program. It is important to move as many joints as possible during this positioning process. For each positioning at least five locations have to be recorded to achieve the required accuracy. The external axis in the same kinematic change such as servo track or position equipment must not be moved. They have to be calibrated separately. In a tool calibration for a welding application, the rotation of the tool is of critical importance since penetration and weld quality are strongly dependent on the angle between the electrode and the object to be welded. A second tool point calibration is therefore usually performed using a long tool-tip in the welding direction. This type of calibration can also be used to calculate the orientation of the tool. An alternative to using the robot to perform the tool calibration is to use external measuring equipment such as a theodolite or a laser interferometer.

Most of the common CAR softwares have a pre-defined function to calculate the tool tip position based on this calibration. In IGRIP, this operation is performed using statistical regression.

4.1.3 Work cell calibration

The position for each object in the work cell has to be determined in the virtual world. This can be achieved using the robot arm or by using external measuring equipment, such as measuring tape or a theodolite system. Objects that require less position accuracy, such as walls that are not critical for collisions with a kinematic device, are located most easily by using a measuring tape. Objects that need high position accuracy, i.e. those that are critical for collision or in a need of precise positioning, such as weld paths, are usually measured using the calibrated robot arm.

A work cell calibration is made by selecting critical points in the work cell. Points on the work-piece are typical examples. Using a calibrated tool, the robot is moved to these points. The co-ordinates of each point are stored in a program und uploaded in the CAR software. In the CAR model the same points have been

4.2 Temperature measurement techniques 33

identified. By comparing measured and modelled co-ordinates, the software calculates a best fit of the position of the parts using statistical regression.

4.2 Temperature measurement techniques

Temperature can be measured with a wide range of sensors. All of them measure the temperature by sensing a change in a physical characteristic. The most common methods are thermocouple, resistance temperature devices (RTD’s and thermistors), infrared radiators, bimetallic devices, liquid expansion devices and change of state devices [21]. In this work both thermal couple- and infrared imaging measurements have been used to measure thermal time histories on plates and on a complex shaped part. The main purpose of the thermocouple measurements was to obtain reference data by which the infrared imaging measurements could be calibrated [21].

4.2.1 Thermocouple instrumentation on plates

Thermocouple is the most common method used to measure temperature. This is due to the fact that they are cheap, interchangeable and can measure a wide range of temperatures. A thermocouple consists of two wires, of different metals, that are joined at one end. A change in the temperature at the connection of the two wires will induce a change in the electromotive force (emf) between the other ends. As the temperature changes, the emf will change. Often the thermocouple wires are located inside a metal or ceramic shield that protects it. The most commonly used thermocouple type is type K. It has one wire of nickel-chromium and one of nickel-aluminium. The contact point of the thermocouple is spot welded on the plate at the desired position where the temperature history is to be recorded [22].

Thermocouples of type K with a wire diameter of 0.11 mm have been used in all experiments in this work on the plates. Six thermocouples were positioned perpendicular to the weld seam. The first gauge was mounted as closely as possible to the melted zone at a distance of 4 mm from the centre of the weld. The remaining gauges were located at 4.5, 5, 6, 7 and 8 mm from the weld centre line, figure 9. A PC based data acquisition system was used to sample the signals from the thermocouples and to monitor the signals on a screen. The measurement data was simultaneously written to disk. The complete measurement system was calibrated in the temperature range 0 – 1350 °C


Weldingdirection

Linescan

Figure 21: Schematic diagram of a plate with thermocouples together with selected measurement line for the IR camera measurements.

4.2.2 Infrared imaging measurement techniques

The infrared (IR) temperature sensor technique is a non-contact measuring method. It measures the temperature by recording the IR energy emitted by the object. As the temperature increases in the object the amount of infrared radiation also increases. Different materials radiate different amounts of IR energy at the same temperature. This efficiency factor is called the emissivity, which is defined as the fraction of radiation emitted by an object as compared to the radiation emitted by a perfect radiator, called the blackbody, at the same temperature. The emissivity may vary from close to 0 (highly reflected mirror) to almost 1 (for a blackbody). The problem with the emissivity is that it can vary with wavelength, component curvature, component surface roughness, viewing angle, and due to surface film effects. An accurate temperature can’t be measured if the object’s emissivity is unknown.

The infrared camera used in this work is a VarioScan 3021 high resolution 16 bit Stirling cooled camera produced by Jenoptic GMbH. The camera has two scanning mirrors to image an object on a point detector of MCT type (HgCdTe), see figure 22. The camera operates in the wavelength range 8-12 µm and has an image resolution on 360(h)×240(v) pixels.

4.3 Residual tress measurements techniques 35

Horizontalscanner

Verticalscanner

Detector

Lens

Figure 22: Principle overview of the VarioScan 3021 high resolution camera.

Temperature measurements have been performed in full frame mode (1 Hz) and in line scan mode (270 Hz). The temperature measurements using the line scan mode were performed in combination with thermocouple measurements, see figure 21. The selected measuring line was then scanned continuously at a rate of 270 lines/s. The position of each thermocouple was registered after welding using a microscope. These positions were used to make comparisons with the IR recordings where the corresponding image pixels were selected. Different techniques of surface treatments were evaluated in order to overcome the problem with surface emissivity variations due to oxidation. Soot from an acetylene flame was found to give a high temperature resistant black surface with a constant emissivity value [24]. Using this technique, reliable measurements could be performed on the complete part, with the exception of the region where fusion had occurred.

4.3 Residual stress measurements techniques

After welding, residual stresses can be measured. Both destructive and non-destructive measurement techniques exist. The techniques can be divided into three main groups namely stress-relaxation, X-ray or neutron diffraction and cracking methods. Tentative measurements of residual stresses using neutron diffraction have been performed in this work. However no results have, as yet, been published.


The stress-relaxation method is a destructive method. It can be divided in two sub-groups. The first group is based on mechanical or electrical strain gauge measurements. The second group uses no electrical or mechanical strain gauges. The residual stresses are, instead, measured by estimation of the elastic strain release. By cutting the test object in several pieces or by drilling and removing a piece of the part, the residual stresses can be relaxed and measured. The stress relaxation is the most common method used since reliable, quantitative data can be obtained.

The x-ray and neutron diffraction methods are very similar. Both methods measure crystal lattice parameters in the welded material and produce interference phenomena that are related to the interplanar spacing of the lattice. The residual stresses are then determined from the change in the interplaner spacing via Bragg’s law and Hooke’s law and compared to the stress-free state. The difference between the methods is that neutron diffraction uses neutrons scattered by a nuclear power source whilst X-rays are used to scatter the electrons in the latter method. The neutron diffraction method provides deeper penetration depths, approximately 30 mm in steel as compared to X-ray penetration depths of about 10 µm.

The cracking residual stress measurement method determines residual stresses by studying the cracks in the melted zone. The cracks may have been introduced by hydrogen or stress corrosion. The method is particularly useful for analyses of components with complex stress distributions.

4.4 Distortion measurements

Distortion of a welded component can be measured using common simple length and angular measuring techniques. Depending on the accuracy desired measuring tapes, Co-ordinate Measuring Machines (CMM) or other more advanced techniques, such as laser- and vision- based systems can be used. Tentative measurements of distortion using CMM have been performed in this work. However, no results have, as yet, been published. Figure 15 shows an example of a typical distorted plate after welding. The surface has been measured using a CMM machine and post processed in UniGraphics.

Figure 23: Distorted welded plate measured using a CMM machine. Results post processed in UniGraphics.

5

Results

In paper I, a CAR software package was used to simulate welding operations and to program robot motions off-line for plates as well as for an aerospace component. The results showed a high degree of accuracy and very little fine-tuning after calibrations had to be made. The method seems to be a powerful tool, specifically in small batch production such as within the aerospace industry. FEA was used in the same paper to predict temperature time histories. The peak temperature was shown to be strongly dependent on the distance from the weld centre line. Good agreements between predicted and measured temperatures, both in peak temperatures and in the cooling histories, were found. The overall conclusion from the simulations was that the model predicted the thermal cycle very well.

In paper III, residual stress distributions were predicted and evaluated along three different sampling lines. The stresses were recorded after 200 s from the start time of welding. The stresses were analysed longitudinally and perpendicularly to the weld seam. The longitudinal stress level and distribution was discovered to be very similar along all three sampling lines. The stress components perpendicular to the weld direction were, however, discovered to have significant differences. The reason for these differences was most probably the varying component stiffness along the weld seam.

An integration between the CAR software and FEA software was also constructed. Using this interface, the optimization of the welding parameters and sequences could be performed. It was shown that penetration control could be achieved by maximizing the weld velocity while maintaining full penetration.

Validations of the FEA simulations were performed by using both thermocouples and IR camera measurements on both plates of Greek Ascoloy and Stainless Steel 316L. An Acetylene/Oxygen soot deposition method was evaluated in paper II, making the IR measurements emissivity independent. Good agreement between IR and T/C temperature measurements was found in the same paper. It was concluded that IR imaging is a useful non-contact method to measure temperatures on complex shaped parts.

38 Results

Several developments of the FEA model are possible. One simplification in the present model is that the tack-welding (performed before the main weld) was not considered. This tack welding will most probably affect the stress level. Including transformation plasticity in the material model will also change the stress state. The present model can however, provide a powerful tool to qualitatively evaluate different weld parameters and fixture designs off-line.

6

Summary and conclusion

In this thesis a method and a simulation tool by which robot trajectories can be defined and thermal, residual stress distributions can be predicted on parts with complex shapes have been developed. The method was evaluated on a piece of an aerospace component where robot weld paths were defined off-line and automatically downloaded to a Finite Element Model, where temperatures and residual stress distributions were predicted. The temperature predictions were compared with experimental measurements using both thermocouple and infrared emission measurements and good agreements were found. No residual stress distributions have as yet been validated but measurements using neutron spectroscopy have been performed which are to be compared with corresponding predictions.

The method described provides a powerful means to construct and optimise torch trajectories and process parameters off-line. Using this system, thermal histories and most probably residual stresses can be predicted on complex shaped parts and in this way resulting changes in the microstructure and mechanical properties can be estimated. Using the developed interface between the Off-line programming and the Finite Element software in combination with the developed optimization, algorithm penetration control can be achieved. Using this method productivity can be maximised whilst still maintaining full penetration.

40 Summary and conclusion

7

Proposals for future work

Several extensions of the modelling work described in this thesis are possible. To extend the work to include filler wire, pulsed current, as well as considering tack-welding would be valuable. To include transformation plasticity in the material model would also be of interest.

The predicted residual stress distributions have not been validated. Measurements on the Aerospace component have recently been performed at Studsvik Neutron Research Laboratory, Uppsala University. A comparison between measured and predicted distributions is to be performed.

To develop a good validation method for distortion simulations would be of interest. Using such a method the optimisation of deformation patterns could be performed. Of specific interest is to evaluate the possibility to use forced cooling during welding and thereby affecting changes in the residual stress-state. Using the developed simulation tool specific cooling profiles which minimise distortion could then be designed.

42 Proposals for future work

References

1 K. Easterling, “Introduction to the Physical Metallurgy of Welding”, second edition, Oxford: Butterworth Heinenmann Ltd., 1992, ISBN 0-7506-0394-1

2 L.P. Connon, “Welding handbook, Welding technology”, Eight edition, Miami: American Welding Society, pp 67-87, 218-264, 1991, ISBN 0-87171-281-4

3 D. LeRoy Olsen, T. A. Siewert, S. Liu, G. R. Edwards, “ASM handbook, Volume 6, Welding, Brazing and Soldering”, pp 7-24, Library of Congress Cataloging-in-Publication Data, 1997, ISBN 0-87170-382-3

4 M. Olsson, “Simulation and Execution of Autonomous Robot Systems”, PhD thesis, Department of Mechanical Engineering, Lund University, Sweden, pp 7-86, 2002,

5 P. Cederberg, M. Olsson, G. Bolmsjö, "Virtual Triangulation Sensor Development, Behaviour Simulation and CAR Integration Applied to Robotic Arc-Welding", Journal of Intelligent and Robotic System, 34(4), pp. 365-379, 2002

6 Y. Ting, W. I. Lei, H.C. Jar, "A path planning algorithm for industrial robots", Computers & Industrial Engineering , 42, pp. 299-308, 2002

7 R. Meredith, U.S. Patent 2,274,631

8 G. Bolmsjö, M. Olsson, K. Brink, "Off-line programming of GMAW robotic systems - a case study", Int J. for joining of Materials, 9, pp. 86-93, 1997

9 J. Goldak, A. Chakravarti, M. Bibby, "A new finite element model for welding heat sources", Metallurgical Transactions, 15B, pp. 299-305, 1984

44 References

10 D. Radaj, "Heat effects off welding", Berlin: Springer Verlag , pp. 18-67, 1965

11 R.T.C Choo, J. Szekely, R. C. Westhoff, "On the calculation of the free surface temperature of gas-tungsten-arc weld pools from the first principles - Part II: Modelling the weld pool and comparison with experiments", Metallurgical transaction B, 23B, pp. 376 - 384, 1992

12 Toselo, F. X. Tissot, M. Barras, "Modelling of the weld behaviour for the control of GTA process by computer aided welding ", Matehematical Modelling of Weld Phenomena 4, pp 80-103, 1997

13 S. K. Choong, "Thermophysical properties of stainless steel", Argonne national laboratory, Illinois, 1975

14 D. Berglund, "Simulation of welding and stress relief heat treating in development of aerospace components", Licentiate thesis, Department of Mechanical Engineering, Luleå University of Technology, Sweden, pp 1-19, 2001

15 P. Sicard, M. Levine, "An approach to an expert robot welding system", IEEE Transactions on System, Man and Cybernetics, 18, pp. 204-222, April, 1988

16 R. Lin, "On residual Stresses and Fatigue of Laser Hardened Steels", PhD thesis, Department of Mechanical Engineering, Lidköpings University, Sweden pp. 62-83, 1992

17 P. Nylen, X. Guterbaum, P. Jonsson, B. Nordin, L. Pejeryd, "Relationship between arc welding parameters asn weld bead geometry in pulsed and non-pules TIG welded IN718", Trends in weding research, pp 408-413, 2002

18 F. P. Incropera, D. P. DeWitt, "Fundamentals of heat and mass transfer, Fourth edition", Fourth edition, New York: John Wiley & Sons, pp. 52-55, 1996, ISBN 0-471-30460-3

19 R. Bernhardt, G. Schreck, C. Willnow, "Realistic robot simulation", Computing and control engineering journal, pp. 174-176, 1995

20 T. Zacharia, J. M. Vitek, J. A. Goldakt, T. A. DebRoy, M. Rappaz, H. K. D. H. Bhadeshia, "Modeling of fundamental phenomena in welds", Modelling Simulation Material, Sci. Eng., 3, pp. 265-288, 1995

21 A. S. Morris, “Principles of Measurement and Instrumentation”, New Jersey: Prentice-Hall, pp 235-267, 1993

Reference 45

22 W. Bolton, “Measurement and Instrumentation systems”, Butterworth Heinemann, pp 346-348, 1996

23 American National Standard, "For Industrial Robots and Robot System, Point-to-Point and Static Performance Characteristics - Evaluation" ANSI/RIA R15.05-2-1992 (R1999)

24 "Table of Various Surfaces", www.mikroninst.com, Micron Instrument Company, USA

25 Y. F. Yong, J. A. Gleave, J. L. Green, and M. C. Bonne, “Off-line programming of robots, Handbook of Industrial Robotics”, New York, John Wiley & Sons, 1985

26 G. C. Carvalho, M. L. Siqueira, S. C. Absi-Alfaro, “ Off-line programming of flexible welding manufacturing cells” Journals of materials processing technology 78, pp 24-28, 1998

27 Y. Ueda, T. Yamakawa, “Analysis of thermal elastic-plastic stress and strain during welding by finite element method”, Trans JWRI, Vol 2, pp 90-100 1971

28 H. D. Hibbit, P. Marcal, “A numerical thermo-mechanical model for the welding and subsequent loading of a fabricated structure”, Comp. & Struct, Vol 3, pp 1145-1174, 1973

Included Papers

Paper I

M. Ericsson, P. Nylen, and G. Bolmsjo,"Three-Dimensional Simulation of Robot Path and Heat Transfer of a TIG-Welded Part with Complex Geometry." In Proceedings of the 11th International Conference on Computer Technology in Welding, pp 309-316, December 6-7, 2001, Colombus, Ohio, USA.

Also published as SME Technical Paper AD02-292 (Dearborn, Mich. Society of Manufacturing Engineers, 2002).

Paper II

P. Henrikson and M. Ericsson “Non-contact Temperature Measurements using an Infrared Camera in Aerospace Welding Applications” Trends in Welding Research: Proceedings of the 6th International Conference, pp 930-936, 15-19 April, 2002, Pine Mountain, Georgia, USA.

Paper III

M. Ericsson, D. Berglund and P. Nylén “Three Dimensional Simulation of Robot path, Heat Transfer and Residual Stresses of a TIG-welded Part with Complex Geometry” Trends in Welding Research: Proceedings of the 6th International Conference, pp 973 – 979, 15-19 April, 2002, Pine Mountain, Georgia, USA.

48 Included papers

Paper I

Three-dimensional simulation of robot path and heat transfer of a TIG-welded part with complex geometry

M. Ericsson*, P. Nylén** G. Bolmsjö***

* University Trollhättan/Uddevalla Box 957 S-461 29 Trollhättan Sweden

** Volvo Aero Corporation, S-461 81 Trollhättan, Sweden.

***Department of Production Engineering, Lund Institute of Technology, Box 118, S-221 00 Lund, Sweden

In Proceedings of the 11th International Conference on Computer Technology in Welding, pp 309-316, December 6-7, 2001,

Colombus, Ohio, USA.

Also published as SME Technical Paper AD02-292 (Dearborn, Mich. Society of Manufacturing Engineers, 2002).

Three-dimensional simulation of robot path and heat transfer of a TIG-welded part with complex geometry

M. Ericsson*, P. Nylén** G. Bolmsjö***

* University Trollhättan/Uddevalla Box 957 S-461 29 Trollhättan Sweden

** Volvo Aero Corporation, S-461 81 Trollhättan, Sweden.

***Department of Production Engineering, Lund Institute of Technology, Box 118, S-221 00 Lund, Sweden

Abstract

The applications of commercial software (OLP) packages for robot simulation, and programming, us interactive computer graphics, provide powerful tools for creating welding paths off-line. By the use of such software, problems of robot reach, accessibility, collision and timing can be eliminated during the planning stage. This paper describes how such software can be integrated with a numerical model that predicts temperature-time histories in the solid material. The objective of this integration is to develop a tool for the engineer where robot trajectories and process parameters can be optimised on parts with complex geometry. Such a tool would decrease the number of weld trials, increase productivity and reduce costs. Assumptions and principles behind the modeling techniques are presented together with experimental evaluation of the correlation between modeled and measured temperatures.

1 Introduction

The metallurgical structure of a metal, which determines its mechanical properties, is a function of its chemical composition, its initial structure and the thermal effects of the welding process. Theoretically, if both the thermal events and the response of the material to the thermal process is known, the resulting changes in microstructure and properties can be predicted. Several papers have been published concerning numerical modeling of thermal histories, residual stresses, and distortion (Ref. 1-7). Mainly two-dimensional studies have been performed. Three-dimensional studies are still restricted to simpler shapes such as plates and pipes.

The use of robots for arc welding started in the early 70´s and is now extensively used in the MIG/MAG processes. Using robots for TIG (GTAW) welding is however still rare. One of the reasons is the increased demand for precise

52

programming and control. Programming of welding robots is usually done manually by the jog teach method. Using this method the robot is off-line, the part stationary, and the robot arm jogged through the program under reduced power and at reduced speed, via a joystick. Generating a path by hand in this way can be time consuming. On a complex geometry, it is virtually impossible for a programmer to maintain constant gun velocity, distance from, and orientation to, the part. However, by using computer simulation this problem can be overcome. Using this method, the programming is moved away from the robot to a graphical computer system, often referred to as a “off-line programming” system (OLP). The technology in this area is well established and has been a research area (Ref. 8-11) for some ten years. Despite these extensive investigations, the two different simulation techniques (numerical process modeling and OLP) seems only to have been studied separately.

The need for a better simulation tool for arc welding was the starting point for a research program at the University Trollhättan/Uddevalla in collaboration with Volvo Aero Corporation. The objective of this program is to provide temperature -time histories and metallurgical- and mechanical-properties predictions on robot welded parts. The program is divided into four parts:

1. to off-line program parts with complex shapes, 2. to numerically predict the shape of the molten pool by the use of

Computational Fluid Dynamics (CFD) techniques, 3. to numerically solve the energy equations in the solid material with

sufficient accuracy that metallurgical predictions can be made, as well as to link the off-line programming model with this numerical model, and

4. to empirically establish relationships between temperature-time history and metallurgical and mechanical properties

This paper is concerned with parts 1 and 3 above; namely methods of programming robots off-line and of predicting temperature-time histories on parts with complex shapes.

2 Principle of Off-line Programming (OLP)

Several commercial software packages for off-line programming of robots exist (CATIA, IGRIP, Robcad GRASP etc.). A brief description of the methodology using such systems is given below. A more detailed description is given in (Ref. 8). The methodology of OLP includes the following steps (Ref. 8):

1. modeling of the work cell, 2. modeling of the work-piece, 3. calibrating the work-cell, 4. adjusting and fine-tuning up and down loading of programs, 5. programming, and

53

6. test runs and macro programming enhancements

The first step to model the work cell concerns the construction of a geometric and a kinematic model of the robot, positioner etc. This demands access to design drawing of the cell together with measurements of critical dimensions in the cell. The workcell model is usually constructed directly in the OLP system. The IGRIP (Interactive Graphics Robot Instruction Program, Deneb Robotics) system was used in this study. In the second step a geometrical description (CAD data) of the part to be welded is generated either in a CAD/CAM software or in the off-line programming software (OLP). If this model is created in a CAD system the data is imported to the OLP software either using a neutral interface (for instance IGES) or a direct reader.

The accuracy of the modeled workcell is usually not high and the third step is therefor to make a calibration by measuring different points in the physical welding cell. This procedure might include several sub-steps depending on the complexity of the workcell. In this work a tool calibration and a calibration of the workpiece were performed. Tool calibration is performed to determine the tool center point and to determine the orientation of the weld torch. The procedure used in this study was to have a measuring arrow in a fixed position in the work cell and to move the robot to this position in different directions. The positions from the real robot cell were then uploaded to the OLP software and a “best fit” was performed by the system. The calibration of the workpiece was performed similarly by moving the robot to clearly identified positions on the workpiece. These positions were recorded and uploaded to the OLP software where the difference between model and measurements was calculated and an adjustment of the model using least squares fitting was done.

The motion of the robot is then programmed in steps four and five, either in a high level programming language (for instance GSL, which is the graphical simulation program in IGRIP) or in a specific robot language (such as RAPID, which is the program language for ABB robots). A robot trajectory is then defined by a set of coordinate frames specifying locations and gun orientation. After that, the motion may be simulated to check the results on the computer. High level languages are then translated and the program finally downloaded to the robot controller where in the final step, test runs are performed. Figure 1 shows a screen-capture during the simulation in the OLP system.

54

Figure1: OLP model.

3 The Computational Heat Transfer Model

The primary aim of the numerical finite element model is to predict the temperature evolution outside the molten zone on a part with complex geometry. Here, the software ICEM CFD, HEXA, which is a commercial pre-processor for CFD and structural applications, was used to mesh the part. The commercial FEM program Marc from MSC Software was used in the heat transfer predictions. User subroutines had to be developed to simulate the moving heat source. A Gaussian surface distribution was used to model the heat source from the weld. This distribution was preferred to a volumetric one (Ref. 4) since it reduces the number of parameters (unknown variables) to be fit and because the plates to be welded were considered thin (<1.5mm) The heat flux was expressed as (Ref. 4):

2

2

0

0

rq

q

r

q

q

eEI

q

EIq

eqq

α

α

παηπαη

−

−

=

=

⋅=

where q denotes the heat transferred to the workpiece, E the voltage, I the

current, η the efficiency factor, qa the concentration factor, and r the radial

distance from the center of the heat source. The distribution was truncated in the radial direction, at a cut off limit of 5 % of the maximal heat input, as proposed by D. Radaj (Ref. 4). Temperature dependent properties as well as phase change was included in the analysis. The thermal conductivity was increased by a factor 10 when the temperature reached the liquidus temperature to account for connective heat transfer in the melted zone (Ref. 4). This simplified model was used instead of more advanced CFD models to simulate the physics in the molten zone, since the

55

latter methods are too computationally demanding. The part was divided into a non-uniform mesh with higher mesh densities close to the weld path (where the highest temperature gradients were assumed to occur), Figure 2. Eight-node brick elements were used. Six elements were defined in the thickness direction. Grid sensitivity trials were made by successively refining the mesh. The final number of nodes in the model was 181500.

Figure 2: The non-uniform mesh. Note the higher densities along the weld path.

The boundary conditions for the analysis are given in Table 1. The model had to be restricted to a subsection of the part to be welded since otherwise to long computational times would have been required. To compensate for this simplification a “Metal-Metal” boundary condition was introduced which symbolizes pure heat conduction through these interfaces, Table 1.

Table 1: Boundary conditions in the heat transfer analysis

Type of condition

Interface Value

Face film Metal – Metal 61000 10−⋅

Face film Metal – Air 620 10−⋅

Face film Metal – Gas (Argon as root gas) 6200 10−⋅

Flux Metal – Heat source user subroutine (q above)

weld path

56

The simulations were to time consuming to be run on a single workstation why parallel computing on a Linux cluster was used. By the use of the parallel computing option in Marc, the part is subdivided into several domains. For each domain, subsolutions are then calculated in parallel on different processor, and an iterative procedure assembles the subsolutions to the global solution. Ten 800Mz processors were used in the present study.

4 Integration of the Heat Transfer and Off-line Programming Models

By the integration of the heat transfer model above with the off-line programming system IGRIP, a powerful, yet efficient tool for temperature prediction and optimization may be obtained. To that end, an interface translating the data; robot coordinates, welding speeds and currents between the two softwares was developed. This interface calculates a linear motion between each robot point, which controls the moving heat source in the finite element calculation. The two softwares (IGRIP and Marc) have to be installed on the same workstation since communication between different operating systems has not been considered. The overall architecture of the simulation system is given in Figure 3.

Robotsimulation

Geometrye.g. IGES

CAD/CAM

FEATranslator

Weldingpath

I, v for full penetration,microstructure, HAZ etc.

Optimizationof I and v


IRBController

Completerobot code

Good andoptimum weld

Figure 3: The overall architecture of the simulation system.

57

5 Experimental

TIG (GTAW) welding was performed using an in-house robotised welding cell. The torch used is from Binzel AB and is linked to a six-axis robot from ABB, IRB1400. The power source is a TIG Commander 400 AC/DC from Migatronic AB. Throughout all experiments, thoriated tungsten electrodes were used. Welding experiments were performed on both plane plates and on a section of an aerospace part (in the following referred to as the production part), namely a turbine component from the V2500 engine provided from Volvo Aero Corporation. 1/13 was cut out of the real part, which originally consisted of an inner, an outer ring and 13 vanes, see Figure 4. The reasons for the experiments on the plane plates were twofold: to calibrate the heat source parameters used in the simulations, and to determine the emissivity in the infrared temperature measurements. Both the plane plates and the production part were made of Greek Ascoloy with a thickness of 1.25 mm. To avoid oxidation, argon was used as root gas in all weld trials. The types of welds performed were bead on plates, and no filler material was used. All plates were tac-welded together before performing experiments so that no gap or misalignments could be introduced. Two special fixtures were designed, one for the plane plates and one for the production part, Figure 4.

Outer ring

Vane

Inner ring

Figure 4: Fixture for the plan plates (left) and fixture with component (right).

To obtain temperature measurements, both thermocouples and high-resolution infrared (IR) emission measurements were used. Six thermocouples were positioned perpendicular to the welding direction. The first gauge was positioned as close to the melted zone as possible at a distance of 4mm from the center of the weld. The second and third were positioned 0.5 mm radially from the previous one. The remaining gauges were positioned 1.0 mm radially from the preview one. The sampling frequency for all thermocouples was 270 Hz. The IR-camera used is a VARIOSCAN 3021 High Resolution, from JENOPTIK, Laser, Optik, Systeme Gmbh, which works in the IR radiation spectrum of 8 – 12 µm. The camera was used both in a line scan mode with a scanning frequency of 270 Hz as well as in a full-frame mode with a frequency of 1 Hz. The analysis of the IR measurements were made using the IRBIS Plus software provided by JENOPTIK. A comparison between the IR results with the thermocouple was made. Different techniques to

58

soot the plane plates were evaluated to reduce emissivity -dependency in the IR-measurements. Finally a method using an acetylene flame was selected. This technique was also used on the production part.

6 Results and Discussion

The results of the robot programs made off-line showed a high accuracy and very little fine-tuning after calibrations had to be made. The method appears as a powerful tool, particularly in small batch production such as within the aerospace industry.

The computational time from the parallel calculation was 34 hrs for the production part. The predicted fusion zone was 5.0 mm on the top-side and 4.8 mm at the root-side which agreed well with measured widths.

The predicted and IR-measured temperature histories in the point located 4mm and 7 mm from the center of the weld are given in Figure 5. The IR-measurements were performed twice, corresponding to the captions IR1 and IR2 in the Figure 5. There is an excellent agreement between predictions and measurements for the 4.0 mm case both in peak temperature and in the cooling history. The agreement for the 7.0 mm case is not as good as for the 4mm case but still good. An example of the comparison between the thermocouple and the IR-measurements are given in Figure 6. There is a very good agreement, which implies that that the technique to soot worked well. The reason why data is lacking in the IR-curve is that the camera-system collects data during a maximum time interval of 20 seconds. This data then have to be written to disk before a new sampling sequence can be gathered. A more extensive evaluation between thermocouple and IR measurements is planned.

Figure 5: Predicted and measured temperature-time histories, 4mm (left) and 7mm (right) from the weld centerline.

0

200

400

600

800

1000

1200

1400

1600

0 5 10 15 20 25 30 35 40

Time (s)

Tem

per

atu

re (

C)

0

100

200

300

400

500

600

0 5 10 15 20 25 30 35

Time (s)

Tem

per

atu

re (

C)

Simulation IR 1 IR 2

SimulationIR 1 IR 2

59

Figure 6: Comparison between thermocouple and IR measured temperatures.

The overall conclusion from the simulations is that the model predicts the thermal cycle very well. However, further research is needed until welded structures can be optimised without experiments. The temperature predictions are naturally dependent on the heat transfer coefficients (boundary conditions in Table 1). To determine these values, experiments are required since the heat flow can be convection dependent, specifically if the workpiece and fixturing are small. Also, the heat source parameters (in the expression for q above) have to be calibrated by experiments. The on-going work at the laboratory at University Trollhättan/Uddevalla to numerically model the magnetohydrodynamics of the arc and to predict the shape of the molten pool by the use of Computational Fluid Dynamics (CFD) techniques seems as a promising tool to compute the heat source parameters without experiments. Such a model will establish a direct relationship between the welding current, speed, voltage and the shape of the molten zone, which can be used as a boundary condition in the temperature predictions in the solid region. Such model will also increase the knowledge of the stirring of the weld pool, the weld pool surface shape and the physics of the arc.

Several extensions of the modeling work described in this article are possible. The simulations can be extended to compute residual stresses, distortion and in the longer term to predict fracture strength and fatigue life of a structure. To extend the modeling work to include filler wire and pulsed current would also be valuable.

7 Summary and Conclusions

An engineering method and a simulation tool to define robot trajectories and to predict thermal histories on parts with complex geometries have been developed. The method was evaluated on a part with a complex shape where robot weld paths were defined off-line, and automatically downloaded to a FEM-model where transient temperatures were predicted. These predictions were compared with experimental measurements using both thermocouple and infrared emission measurements and good agreement was found. The described method provides a promising means to construct and optimise torch trajectories and process

0

100

200

300

400

500

600

700

800

900

1000

0 10 20 30 40 50 60 70

Time (s)

Tem

per

atu

re (

C)

Thermocouple IR

60

parameters off-line. Using this system, thermal histories can be predicted on complex shaped parts and thereby resulting changes in microstructure and mechanical properties be estimated. The models used may after futher development enable the optimization of welding processes, thus increasing productivity and reducing the need of weld trials.

8 Acknowledgement

The authors wish to acknowledge the guidance in the temperature measurements of Per Henrikson (Volvo Aero Corporation), and the assistance in the laboratory by Xavier Guterbaum (University of Trollhättan/Uddevalla) and Börje Nordin (Volvo Aero Corporation). Appreciation is expressed to Peter Jonsson of Volvo Aero Coropration for providing samples for this research and to Anita Hansbo of University of Trollhättan/Uddevalla for linguistic revision. The work was funded by the Foundation for Knowledge and Competence Development and EC Structural Founds.

9 References

1 Eagar T. W., Tsai, N. S. 1983. Temperature Fields Produced by Traveling Distributed Heat Sources. American Welding Society Journal 62(12) 346-s to 355s

2 Gu, M.; Goldak ,J.; Hughes, E. 1993. Steady state thermal analysis of welds with filler metal addition. Canadian Metallurgical Quarterlv. 32 (1): 49-s to 55-s.

3 Radej, D. 1992 Heat Effects of Welding: Berlin: Springer Verlag.

4 Goldak, J.; McDill, M.; Oddy, A.; House, R.; Chi, M.; Bibby, M. 1987. Computational Heat Transfer for Weld Mechanics. Proc. of Int. Conf. on Trends in Welding Research, Advances in Welding Science and Technology. Eds S. A. David: 15-20. Metals Park ASM Int.

5 Jonsson, M.; Karlsson, L; Lindgren, L.E. 1985. Deformation and Stresses in Butt Welding of Large Plates with Special References to the Material Properties, J. of Eng. Mat. And Tech. 107: 265-s to 270-s.

6 Lindgren, L.E.; Karlsson, L. 1988. Deformation and Stresses in welding of Shell Structures. Int. J. for Numerical Methods in Eng. 25: 635-s to 655-s.

7 Bolmsjö, G.; Olsson, M.; Brink, K. 1997. Off- line programming of GMAW robotic systems – a case study. Int. J. for the Joining of Materials, 9 (3): 86-s to 93-s.

8 Buchal, R.O.; Cheras, D.B.; Sassani, F.; Duncan J.P. 1989. Simulated Off-Line Programming of Welding Robots. Int. J. of Robotics Research 8 (3): 31-s to 43-s.

61

9 Bolmsjö, G. 1999. Programming robot welding system using advanced simulation tools. Proc. of the International Conf. on the Joining of Materials JOM-9, 284-291. May 16-19, 1999, Helsingør, Denmark.

10 Walter S. 1994. Simulation and Calibration for Off-line Programming of Industrial Robots. Proc. of Computer Technology in Welding: Paper 54. Paris 15-16 June.

Paper II

Non-contact temperature measurements using an infrared camera in aerospace welding applications

P. Henrikson*, M. Ericsson**

* Volvo Aero Corporation, S-461 81 Trollhättan, Sweden.

** University Trollhättan/Uddevalla Box 957 S-461 29 Trollhättan Sweden

In Proceedings of the Trends in Welding Research: Proceedings of the 6th International Conference, pp 930-936, 15-19 April, 2002,

Pine Mountain, Georgia, USA.

Non-contact Temperature Measurements using an Infrared Camera in Aerospace Welding Applications.

P. Henrikson*, M. Ericsson**

* Volvo Aero Corporation, S-461 81 Trollhättan, Sweden.

** University Trollhättan/Uddevalla Box 957 S-461 29 Trollhättan Sweden

Abstract

This paper describes the application of infrared (IR) thermal imaging and temperature measurements in welding applications, both on single plane plates and on an aero engine turbine component with complex geometry. Temperature profiles were measured on the plates using thermocouples (T/C) in combination with an IR camera system, and the results were compared. The IR camera was used both in line scan mode (270 Hz scan frequency) and in full frame mode (1 Hz frame rate). Different methods of surface treatments have been tested to handle the problem of the surface emissivity variations due to oxidation during welding. Results from measurements using thermocouples and IR camera is presented in the paper as well as temperature measurements using the IR camera on an turbine exhaust case (TEC) engine component.

1 Introduction

Experimental temperature data is required in welding research and development, and for validation of numerical simulations of the welding process (Ref. 1). Temperature measurements near or in the weld pool are of special interest, and this is a challenge in selection and application of measurement methods and for assessment of measurement quality.

1.2 Thermocouple measurement in welding.

Temperature measurements using T/C type K (NiCr-NiAl) is commonly used in welding research. They are relatively inexpensive and can be used at the high temperatures near the weld pool. There are however some issues that need to be addressed concerning measurement quality when using T/C type K in welding applications (Ref. 2). Measurement response time is critical, especially when the T/C is installed close to the weld. Generally, it is not possible to state the response time for a single T/C in this situation, since it is the whole measurement system response time that is measured. The T/C wire diameter is an important parameter

66

to consider when discussing response time, but also the attachment method is an important factor, which effects the response time. Another issue is measurement system calibration (including T/C wire), which must be performed over the whole measurement range. Measurement inaccuracy needs to be addressed when using T/C type K at high temperatures in the range 1250 °C – 1372 °C, which is above the recommended range of use. There are however other T/C types that can be considered to be used in welding experiments, for example T/C type B (Pt-30%Rh-Pt-6%Rh) which can be used at higher temperatures, but can be more difficult to install.

1.3 Infrared radiation measurement in welding.

Infrared radiation measurement of temperature has several advantages compared to T/C measurement. Using a pyrometer, non-contact temperature measurement can be done on a single point on the object (Ref. 3). A scanning mirror IR-camera with a photon detector is an attractive device in welding research and applications. It allows both full field temperature measurement, as well as high speed measurement using line scanning. The use of IR cameras for quantitative temperature measurements in TIG (GTAW) welding is in many cases limited by the difficulties to handle the surface emissivity variations and the reflected radiation from the electrode during welding. When performing spectral radiance temperature measurements, the spectral radiance temperature measured is not only depending on the object emissivity, but also on the wavelength spectral band used for the measurement (Ref. 4). The emissivity used in the measurement must then be spectral emissivity used for the detection device. A number of issues must be addressed when performing IR radiation temperature measurements in the weld pool. One important factor is the uncertainty in the surface emissivity in the melt at the measurement wavelength, and the contribution from electrode background reflection (Ref. 5). Another factor to consider is the occurrence of slag in the melt (Ref. 6). The melted steel and the slag have different radiance due to their emissivity, which are different. Slag has in general higher emissivity than steel and it appears hotter compared to the melted steel at the same temperature. There is also an uncertainty in the IR measurement due to the change in surface emissivity during material solidification and the effect of the surface oxidation in the cooling phase. The surface emissivity is assumed to increase during cooling and oxidation, but it is difficult to measure actual emissivity during the cooling phase (Ref. 7). A well known fact in radiation measurements is that when there are uncertainties in the emittance of a surface it is general best to do measurement at as short wavelength as possible. This is due to that the spectral radiance as a function of temperature increases very rapidly towards shorter wavelengths. A given uncertainty in emissivity then leads to smaller uncertainties in temperatures at shorter wavelength.

67

2. Instrumentation

2.1 Thermocouple instrumentation on plates.

Type K thermocouples were used in all experiments on plates. Six T/C were positioned perpendicular to the welding direction. The first gauge was positioned as close to melted zone as possible at a distance of 4 mm from the center of the weld. The rest of the T/C were positioned at 4.5, 5, 6, 7 and 8 mm from the center of the weld. The T/C were coupled to a signal conditioning unit and the T/C signals were amplified to give a calibrated and linearised output from 0 to 5 volt. A PC based data acquisition system was used for sampling of the T/C signals, and the measurements were presented on-line on the PC monitor and measurement data was written to disk. The complete measurement system (including T/C wire) was calibrated over the whole measurement range. The T/C used in the test had a wire diameter of 0.11 mm and surrounded by ceramic cement and Inconel protection for lead out. The T/C wires were attached to the plate using spot welding.

2.2 Infrared camera.

The IR camera used is a Varioscan 3021-ST high resolution 16 bit Stirling cooled camera from Jenoptik GMbH, Germany. The camera uses scanning mirrors to image the measurement object on a point detector. The camera resolution is 360(h)×240(v) pixels, and the operating wavelength range is 8 µm – 12 µm. The camera detector is of MCT type (HgCdTe). The camera has four filters for different calibrated temperature ranges, and at the highest measurement range, the system is calibrated in the temperature interval from 200 °C -1700 °C. For higher temperatures linear extrapolation of the Planck black body radiator is used. The camera was used both in line scan mode with a horizontal line scanning frequency of 270 Hz, as well as in full frame mode with a frequency of 1 Hz. The measurements and analysis of the IR tests were made using the IRBIS Plus software from Jenoptik.

2.3 Surface preparation.

Different techniques of surface treatment have been tested to handle the problem with surface emissivity variation on the metallic surface due to oxidation outside the weld joint. The surface treatment should ideally have a low emissivity variation over a wide temperature range and should be insensitive to emissivity variations due to oxidation during welding. Diffuse black high temperature paint for engine exhaust pipes was tested. Initial tests showed that the paint could be used up to about 650 °C. In order to find a surface treatment that could be used at higher temperatures, different kind of soot deposition techniques were tested. From the experiments it was found that by using an acetylene/oxygen flame, a thin high temperature resistant soot layer could be produced. The optimal gas mixing is reached by starting from a pure acetylene flame, and then gradually increase the

68

oxygen gas until no soot is visible in the flame. An emissivity value of 0.96 has is reported in Ref. 8 for soot applied to a solid in the range 50-1000 °C.

3 Experimental setup

TIG (GTAW) welding was performed using an in-house robotised welding cell. The torch used is from Binzel AB and is linked to a six-axis robot from ABB, IRB1400. The power source is a TIG Commander 400 AC/DC from Migatronic AB. Throughout all experiments thoriated tungsten electrodes were used. A special fixture designed to avoid distortion has been used during the welding of the plates. The aero-engine component, a part of a V2500 engine turbine exhaust case (TEC) from Volvo Aero Corporation, was TIG welded using the robotised welding cell. A segment of 1/13 was cut out of the TEC, which originally consists of an inner and outer ring and 13 vanes, see figure 1. The TEC is made of Greek Ascoloy with a vane thickness of 1.25 mm. The vane was spot welded between the outer and inner ring.

Figure 1: Overview of the aero engine component and the IR-camera setup.

4 Measurements

Welding experiments were performed on both plane plates and on the turbine component. The purpose of the T/C measurements was to get reference data against which the IR measurements could be calibrated. Initial test measurements were performed on T/C instrumented stainless steel plates. For comparison of the IR measurements against T/C measurements, the acetylene/oxygen sooting technique was used on the plates.

69

Figure 2: Weld test on the Turbine exhaust case (TEC).

Using this technique, all but the weld joint remained sooted during and after the weld experiment. This allowed quantitative temperature measurement on the outside the weld joint using the IR camera. In fig. 2 the TEC component is shown after a weld test. In the experiments, the Greek Ascoloy plates and the TEC vane had a thickness of 1.25 mm. The stainless steel plate thickness was 2.0 mm. To avoid oxidation on the backside during welding, Argon gas was used as root gas in all weld trials. The types of welds performed were bead on plates, and no filler material was used.

4.1 Measurement on plates.

Figure 3 are showing temperature measurements on a Greek Ascoloy plate with 6 T/C installed.

10 20 30 40 500

250

500

750

1000

1250

Time (s)

Tem

pera

ture

(

o C)

4mm4.5mm5mm6mm7mm8mm

Figure 3: Temperature profiles (T/C) on Greek-Ascoloy, plate thickness 1.25 mm, I=75 A, v=2.5 mm/s.

70

The T/C were spot welded to the plate and positioned in radial direction to the weld. The first T/C was mounted as close as possible to the weld (fig 4). The position of the T/C was measured in a microscope after welding. These T/C positions were later used in the analysis of the IR line scan temperature images, for which the corresponding pixels were selected and used for comparison of temperatures.

4.0 mm

150 mm

50 m

m

ABWeld line

Wel

ding

dirc

etio

n

Line

scan

Figure 4: The drawing shows the measurement positions on a plate for T/C and IR camera line scan.

When using the IR-camera in line scan mode, two different acquisition modes can be used. In one mode, 5600 lines are scanned continuos at a rate of 270 lines/s and then the data is saved do disk. The scanning time and the readout time is both 21 sec, which means that only a part of the whole temperature cycle will be measured, and this can be seen in fig. 16. In the other line scan acquisition mode, images like those in fig. 5 are captured, with an image size of 360(h)×x240(v) pixels. After a camera readout time of approx. 0.1 s, which is indicated in the fig. 5 by the white field, another line scan image is taken, and so on until the end of the weld test. During the IR camera measurements a macro lens was used with a working distance of 100 mm to the plate, see figure 15. Due to the viewing angle, only a small section of the plate will be in focus. In the camera software, a camera line perpendicular to the welding direction (se fig. 4) is selected at the focused position on the plate and scanned at 270 Hz. The optical magnification in the IR camera system gives a spatial resolution of 7 pixels/mm on the object during line scan. Care has been taken to select the correct pixel in IR scan lines that correspond to the T/C position on the plate. In fig. 6 radial temperature profiles representing different line scans have been plotted.

71

Figure 5: Two 360 (h) × x240(v) infrared line scan temperature images taken during welding. One image represents 0.9 s, and the welding direction is from right to left in the pictures.

As can be seen from fig. 6, the radial temperature profile for the sooted stainless steel plate has three peaks. For the curve going through the maximum temperature in the weld pool, the soot is probably attached to the surface outside the weld pool for temperatures until the curve drops on each side of the weld. For comparison, temperatures measured with T/C positioned at 4, 5 and 6 mm from the center of the weld are also plotted in figure 6. The agreement between IR- and T/C measurements is good in this region, as can be seen in figure 6. Over the melting temperature, the soot layer has disappeared and the surface emissivity changes instantly. This can be seen as a sudden temperature drop at the edge of the weld pool. At the center of the weld, the high temperature peak is due to reflection of radiation from the electrode, and this is explained from the results in figure 14 and 15. The line plotted 6.7 s after the maximum temperature profile shows that the temperature wave has propagated far out on the plate, and that temperature in the center of the weld is in the same range as at the edge of the weld (where the soot is still attached). The lower temperature seen at the weld joint is due to a different surface emissivity compared to outside the weld. The surface emissivity of the weld is changing (increasing) over time due to material solidification and surface oxidation. During the IR camera measurements, a constant emissivity value of

0.99ε = was used at every pixel in the image.

72

50 100 150 200 250

500

1000

1500

2000

Pixel

Tem

pera

ture

(

o C)

Max+3.0s+4.8s+6.7sT/C

Figure 6: IR line temperature profiles and T/C measurements in radial direction from the weld for a stainless steel plate.

4.2 Measurements on the aero-engine component.

Several welding experiments were performed on the component and temperatures were measured using the IR camera at 1 Hz full frame rate and the 270 Hz line scan mode. No thermocouples were used in the experiments on the TEC component. In fig. 7 is shown a full frame (1 Hz) IR temperature measurement during welding on the sooted TEC (compare the weld path and the component set-up in fig. 2 and fig. 1). In this experiment the filter range was up to 800 °C, therefore temperatures higher than 800 °C are shown in black in the image. On the outer ring of the component, there are two rigid supports (which can be seen in fig. 1), and the cooling effect of these supports can be clearly seen in fig 7.

Figure 7: Full frame thermal image of turbine component showing the heat propagation in the range 100-800 °C.

73

Line scan temperature measurements were also performed during the weld test, and a contour plot of the temperature distribution of the middle section is seen in fig. 8. The cooling effect of the support is clearly also in this graph.

0 100 200 3000

1000

2000

3000

4000

5000

250

500

750 10

00

Weld line

Pixel

Line

num

ber

Figure 8: Temperature contour plot of the middle section of the turbine exhaust case (TEC) measured using line scanning.

Figure 9 shows line scanned temperature profiles on the front part of the vane. All profiles are measured to the right side of the weld as seen in fig. 7 and in the contour plot in fig 8. The temperature profiles in figure 9 correspond to the radial positions for which the T/C were instrumented on the weld tests on the Greek-Ascoloy plane plates.

Figure 10 shows four temperature profiles from the middle part of the TEC. The highest temperature in this part of the component is much lower compared to the front and back part due to the big heat sink on the outer ring support. The measured IR line scan temperature profiles for the back part of the TEC is shown in figure 11. Here the wall thickness in the vane and the outer ring can be assumed to be homogenous, like the front part.

74

5 10 15 200

250

500

750

1000

1250

Time (s)

Tem

pera

ture

(o C)

4 mm5mm6mm7mm

Figure 9: Infrared line scan temperature profiles measured on the front part of the TEC.

5 10 15 20200

400

600

800

Time (s)

Tem

pera

ture

(o C)

4mm5mm6mm7mm

Figure 10: Infrared line scan temperature profiles measured on the middle part of the TEC.

5 10 15 20200

450

700

950

1200

Time (s)

Tem

pera

ture

(

o C)

4mm5mm6mm7mm

Figure 11: Infrared line scan temperature profiles measured on the back part of the TEC.

75

5 Discussion

5.1 Thermocouple measurements.

Close to the weld, the temperature gradient is very high and the spot welded T/C installation will effect the transient response of the T/C, and this may cause a significant error in the measurement. Comparison of T/C and IR measurements indicates that there may be a significant difference in peak temperature between the T/C temperature measurement and the IR measurement near the weld joint. The spot-welded T/C had a diameter about 0.7 mm on the plate. Measurements were done to study the effect in transient response and measured peak temperature, using a smaller spot weld diameter, approx. 0.56 mm. In order to study the potential lag effects of different spot sizes, four T/C were installed in pairs on two plates, see fig 4. One plate had T/C separated at the radial distance 3.8 mm (C1 and C2) and 4.3 mm (D1, D2) from the center of the weld. The T/C pairs at the same radial distance were separated 3 mm in the axial direction. The other plate had the T/C separated at the radial distances of 4.3 mm (A1, A2) and 4.8 mm (B1, B2). The two plates were welded with different welding currents. The reason for this is that at the first plate, the first T/C pair was at the very edge of the weld seam. On the other plate the welding current was lowered so the T/C should be about 1-1.5 mm away from the edge of the weld pool. The two measurements have been plotted in the same graph, se fig. 12. Experiments were also done with smaller spot weld sizes, but these did not survive the weld test.

25 30 35 40 45850

950

1050

1150

1250

1350

Time (s)

Tem

pera

ture

(o C)

A1A2B1B2C1C2D1D2

Figure 12: Comparison of T/C peak temperature difference at the thermal gradient near the weld for different T/C spot weld sizes and radial positions.

As can be seen, the T/C pair (C1, C2) closest to the weld, and at a distance of 3.8 mm from the center of the weld, is showing a relatively large difference in peak temperature. The T/C pair (D1, D2) at a distance of 4.3 mm shows a smaller difference in peak temperature. In the second test, the weld width was smaller, and the T/C pairs (A1, A2) and (B1,B2) was positioned 4.3 and 4.8 mm from the center of the weld. As can be seen in fig. 12, there is a small peak temperature

76

difference for T/C pair (A1, A2), but for T/C pair (B1, B2) the two different spot weld sizes shows identical temperatures. During the passing of the TIG weld temperature transient, no difference in transient response time due to the different spot weld sizes could be measured during the test, only a difference in peak response temperature. The peak T/C temperature difference due to surface attachment size is significant only very close to the weld. Using a non-contact fast response IR detector in this region, it can be expected to give even higher peak temperatures (if the surface emissivity and the background reflection are known).

5.2 Infrared image data processing.

In order to generate temperature profiles at different radial distances from the weld, the center of the weld has to be defined in the IR image. Two different methods based on pixel averaging and peak temperature detection have been used to accomplish this, and both have been found to give the same result, within ± one pixel.

0 100 200 3000

1000

2000

3000

4000

5000

250

500

500

500

500

750

750

750

1000

1000

Pixel

Line

num

ber

Figure 13: IR line scan temperature contour plot from a weld test on a sooted stainless steel plate.

In fig. 13 a temperature contour plot from a weld test on a sooted stainless steel plate is shown. In fig. 14 the high temperature region of fig. 13 is contoured.

77

145 150 155 160 165

185019001950200020502100215022002250 75

0

1000

1000

1250 12

50

1500 15

00

1750

1750

2000

2000

PixelLi

ne n

umbe

r

Figure 14: Temperature contour plot showing the peaks where the electrode position is right above the measurement position (line 1844) and the peak due to reflection of electrode radiation (line 2192).

By counting the number of lines between the two high temperature peaks in figure 14, and using the line sampling frequency, 270f = Hz, together with the welding speed ( 2.5v = mm/s), gives that the weld torch has traveled the distance of 3.22 mm between the two peaks. Using the electrode to plate distance 1.5 mm gives the angle 25α = ° in figure 15, which is close to the viewing angle during the experiments

TorchTorch

αα

IR

camera

Welding direction

Figure 15: Optical and geometrical calculation gives torch positions of maximum reflected radiation in the IR-image.

In fig. 16, temperature measurements using T/C and IR are shown from a welding experiment on a sooted Greek-Ascoloy plate. The measurement position is just at the edge of the weld, 4 mm in radial direction from the center of the weld. The explanation for the high IR temperatures up to 500 °C in the beginning of the heat transient is due to the weld torch that comes into the field of view of the camera. The next part of the curves shows the temperature peak response. It is seen that the IR measurement reach a higher peak value ( 1418IRT = °C) compared to the T/C

measurement ( 1243T CT = °C), and this can also be seen in the measurements in

fig. 6 for stainless steel. From the T/C response tests, it was stated that an IR measurement is expected to reach a higher peak temperature very close to the weld,

78

compared to a T/C measurement. In all experiments, the T/C position and the IR scan line is not taken at the same position on the plate, see fig. 4, but are a separated by a few millimeters. Due to process variations, the IR measurement position may be inside the melt, explaining the sudden drop in temperature as a result of a sudden emissivity variation. As can be seen in figure 16, during the cooling phase, the IR and T/C temperature curves show good agreement. This means that surface emissivity is close to the value set in camera, 0.99ε = , indicating that the surface oxidation during the cooling phase results in a high emissivity. It should be pointed out that the agreement between the T/C and IR measurements increases with the radial distance from the weld, and this can be seen in fig. 17.

10 20 30 40 50 60 700

250

500

750

1000

1250

Time (s)

Tem

pera

ture

(

o C)

T1IR 1

Figure 16: Comparison of T/C and IR line scan temperatures on a sooted Greek-Ascoloy plate at 4 mm from the weld line.

An example of this is shown in figure 17 for a stainless steel plate, showing good agreement between T/C and IR measurements at 6 mm and 7 mm from the weld line.

10 15 20 25 30 35200

400

600

800

Time (s)

Tem

pera

ture

(o C)

TC 6mmIR 6mmTC 7mmIR 7mm

Figure 17: T/C and IR temperatures on a sooted stainless steel plate 6 mm and 7 mm from the weld line.

79

6 Summary and conclusion

Temperature measurements have been successfully performed on an aero engine turbine component using an infrared camera system. Both full field temperature images and time resolved line scan profiles have been measured and analyzed. By deposition of a soot layer on the metal surfaces to be welded, a surface with high emissivity was produced that made it possible to handle the emissivity variation due to surface oxidation outside the weld joint, and to suppress reflected radiation. Thermocouple- and infrared measurements have been performed on plane plates made of stainless steel and Greek Ascoloy, and comparative analysis has been made of the results. Infrared radiation temperature measurements has also been made in the weld pool and during solidification and the cooling phase, and the results have been analyzed and problem areas have been identified that promote further work in this field

7 Acknowledgements

The authors would like to thank Xavier Guterbaum (University of Trollhättan/Uddevalla), for the assistance and valuable discussions in the laboratory. The authors also would like to thank Dr. Per Nylén for comments and suggestions in the preparation of the paper. This work was made in collaboration between Volvo Aero Corporation and University of Trollhättan/Uddevalla. The work done by the University of Trollhättan/Uddevalla was funded by the Foundation for Knowledge and Competence Development and EC Structural Founds.

8 References

[1] T.Zacharia, S.A. David and J.M. Vitek, Effect of Evaporation and Temperature-Dependent Material Properties on Weld Pool Development, Metallurgical Transactions B, Volume 22B, 233-240 (1991)

[2] Manual on the use of Thermocouples in Temperature Measurement, ASTM Special Technical Publication 470B (1981)

[3] D. Farson, R. Richardson and X. Li, Infrared Measurement of Base Metal Temperature in Gas Tungsten Arc Welding, Welding J., Vol 77(9), 396-401 (1998)

[4] D.P. DeWitt and G.D. Nutter, Theory and Practice of Radiation Thermometry, J. Wiley & Sons, Inc. (1988)

[5] H.G. Kraus, Experimental Measurement of Stationary SS 304, SS 316L and 8630 GTA Weld Pool Surface Temperatures, Weld. J., vol. 68(7), 269-279 (1989)

80

[6] G.R. Peacock, Thermal Imaging of Liquid Steel and Slag in a Pouring Stream, Proc. Of SPIE Vol 4020, Thermosense XX22, 50-60 (2000)

[7] P.D. Jones and E. Nisipeanu, Spectral-Directional Emittance of Thermally Oxidized 316 Stainless Steel, Int. Journal of Thermophysics, Vol. 17, No4, 967 - 978 (1996)

[8] Table of Emissivity of Various Surfaces, www.mikroninst.com, Mikron Instrument Company, USA.

81

Paper III

Three Dimensional Simulation of Robot path, Heat Transfer and Residual Stresses of a TIG-welded Part

with Complex Geometry

M. Ericsson1, D. Berglund2, P. Nylén13

1 University Trollhättan/Uddevalla Box 957 S-461 29 Trollhättan Sweden

2 Luleå University of Technology, Division of Computer Aided Design, SE-97187 Luleå, Sweden

3 Volvo Aero Corporation, S-461 81 Trollhättan, Sweden.

In Proceedings of the Trends in Welding Research: Proceedings of the 6th International Conference, pp 973 – 979, 15-19 April, 2002,

Pine Mountain, Georgia, USA.

Three Dimensional Simulation of Robot path, Heat Transfer and Residual Stresses of a TIG-welded Part

with Complex Geometry

M. Ericsson1, D. Berglund23, P. Nylén13

1 University Trollhättan/Uddevalla Box 957 S-461 29 Trollhättan Sweden

2 Luleå University of Technology, Division of Computer Aided Design, SE-97187 Luleå, Sweden

3 Volvo Aero Corporation, S-461 81 Trollhättan, Sweden.

Abstract

In this paper a system is presented that combines a robot off-line programming software with a finite element model that predicts temperature-time histories and residual stress distributions. The objective is to develop a tool for the engineer where robot trajectories and welding process parameters can be optimised on parts with complex geometry.

The system was evaluated on a stainless steel gas turbine component. Robot weld paths were defined off-line and automatically downloaded to the finite element program, where transient temperatures and residual stresses were predicted. Temperature dependent properties and phase change, were included in the analysis. Assumptions and principles behind the modeling techniques are presented together with predicted temperature histories, residual stresses, and fixture forces.

1 Introduction

A large number of aerospace components have complex shapes and are manufactured in several steps, often including joining operations. Joining by welding, however, induces changes in the base metals microstructure and can generate unwanted stresses and deformation. To avoid deformations, expensive and complex fixtures often have to be used. Furthermore the planning of optimal welding sequences in aerospace component welding is difficult and requires highly experienced operators. Therefore a simulation tool to be used for evaluation of features such as structural behavior, welding sequences and fixture solutions during the design stage would be desirable. By the use of such a tool it would be possible to evaluate manufacturing processes in the early stages of product development in

84

order to reduce both the number of welding experiments and the need of welding operator experience. The tool should preferably be capable of simulating the welding torch path, of detecting collisions between torch and workpiece, and of optimizing welding parameters considering penetration and component deformation behavior. Thus a combination of the simulation techniques finite element analysis (FEA) and off-line programming (OLP) is necessary.

FEA simulation of thermal history, residual stresses and distortion has been performed since the early 70ies and several papers have been presented [1-6]. Large complex simulation models of three-dimensional components are however still rare, mainly due to lack of computational power. The reason is that to be able to compute temperature and residual stress fields in the affected zone a very fine discretization of the space variable is required to accommodate sharp gradients properly.

The OLP technology is also well known and has been a research area in many years [7-9]. By the use of this technique, the programming of the robot is transferred from the workshop to a computer system where the programming can be performed without disturbance of the production. The technique can be used to simulate the welding torch path, to detect torch-workpiece collisions and to control torch orientation as well as electrode distance.

In this paper an integrated approach, using two commercial OLP and FEA codes, is described making it possible to optimise torch trajectories and weld parameters off-line. The best welding parameters, i.e., the parameters that generate the lowest component deformation while keeping full penetration, can thus be found. The objective of this paper is to describe the system principle and to demonstrate some of its capabilities. The principle of the OLP-FEA integration which previously has been presented in [10] is summarised here for completeness.

2 Integration of OLP and FEA

The overall principle of the integrated system is given in Figure 1. The following steps describes the procedure:

1. The component to be manufactured is created in a CAD/CAM system. 2. The model in step one is imported to the OLP software and the FEA

software, either using a direct translator or by using a neutral file format, such as IGES or STEP. In the OLP software a model of the work cell is created. A welding program (including robot motion, weld speed, and arc current, etc.) is developed. Check for collisions are also made.

3. The welding parameters, i.e., robot co-ordinates, weld speed and arc current, are exported from the OLP software to the FEA software using an interface developed as part of this project.

85

4. Thermal histories and residual stresses are predicted in the FEA software. An optimization of weld velocity is performed to generate the lowest component deformation while keeping full penetration.

5. The optimised welding parameters are exported to the robot motion program. A translation of the program to a specific robot manufacturer language is made.

6. The final program is downloaded to the manufacturing equipment (Irb and weld controller systems).

Robotsimulation

Geometrye.g. IGES

CAD/CAM

FEATranslator

Weldingpath

Thermal historyResidual stresses


IRBController

Completerobot code

Full penentration weldwith low distortion

Weldvelocity (wv)wv

Figure 1: Simulation system architecture.

A more detailed description of the principles of the robot simulation (OLP), step two above, and the FEA simulation, step four, is given in the following sections.

3 OLP

Several commercial software packages for OLP exists (GRASP, IGRIP and Robcad etc.). The procedure using these systems can be summarised in the following steps. A more detailed description can be found in [7].

1. Modeling of the work cell 2. Work cell calibration 3. Programming of robot, positioner and other optional work cell equipment 4. Down loading of the program to the robot controller 5. Additional robot programming 6. Test running

86

The first step considers the construction of a geometrical model of the workpiece, and a geometrical as well as a kinematic model of the work cell. These models are constructed using drawings, CAD/CAM models, or if such not are available, by doing measurements of all critical equipment positions in the cell. The work cell model might alternatively be constructed directly in the OLP system.

In the second step a calibration of the model with the real cell is performed. This step can include several sub steps such as tool point, work piece and signature calibration. [7]. Tool calibration is performed to determine the tool center point and to determine the weld torch orientation. A procedure using a measuring arrow in a fixed position in the work cell and moving the robot to this position in different directions is usually used. The positions from the real robot cell are then uploaded to the OLP software and a “best fit” is found using regression. The calibration of the workpiece is performed similarly by moving the robot to identified positions on the workpiece. These positions are recorded and uploaded to the OLP software where a similar adjustment of the model is done. To find errors in the geometrical model of the robot, an arm signature calibration can be used. This calibration finds any deviation in the length of the robot joints and in the zero points for the joints.

In step three the robot motion is programmed using either a high level language or a specific robot language. If a high level language is used the program is translated to the specific robot language and downloaded (step four) to the robot controller system.

Usually equipment specific additional programming is needed (step five), which is performed manually at the robot. Validation of the program by test runs is finally performed in step six.

The IGRIP (Interactive Graphics Robot Instruction Program, Deneb Robotics) system was used in this study. A tool calibration and a workpiece calibration were performed. The high level language GSL, which is the graphical simulation language in IGRIP, was used for programming all devices in the cell. The programmed part was a section of an aerospace part, a turbine component from the V2500 engine provided from Volvo Aero Cooperation. 1/13 was cut out of the real part, which originally consisted of an inner, an outer ring and 13 vanes, Figure 2. The welding paths, including initial weld velocities, were then exported to the finite element software where predictions of temperature histories, residual stresses and fixture reaction forces were performed. The principle of this model is described in the next section.

87

Vane

Outer Ring

Inner Ring

Vane

Outer Ring

Inner Ring

a)

b)

Figure 2: a) Turbine component consisting of an inner, an outer ring and 13 vanes. b) experimental component (1/13 of the turbine component).

4 FEA

The commercial FEA program MARC from MSC Software was used. The model contained 3056 shell elements and 3182 nodes, see Figure 3. A staggered approach was used for the coupled thermal-mechanical simulation. This means that the updating of the geometry used in the thermal calculation is lagging one time step behind.

Weldpath

Flange

Figure 3: Shell model of experimental component.

88

4.1 Boundary Conditions

Different methods to simulate the thermal load in a welding analysis exist. One method is to prescribe the temperature in a certain volume of material and to adjust the temperature level in order to obtain an acceptable dimension of the fusion zone. A more sophisticated method is to use a moving heat source. User subroutines were therefore developed to simulate a moving Gaussian surface distribution [4]. This distribution was preferred to a volumetric one since it reduces the number of parameters (unknown variables) to be fit and because the plates to be welded were considered thin (<1.5mm). Natural convection was only used as energy dissipation to the surrounding. The flanges on the Inner- and Outer ring were assumed to be clamped in the model since no fixture was used. The Inner- and Outer ring were instead welded on a steel plate, see Figure 2b.

-0,004

-0,002

0

0,002

0,004

0,006

0,008

0,01

0,012

0 200 400 600 800 1000 1200Temperature [ºC]

εth

Martensite curve

1

2

a)

-0,004

-0,002

0

0,002

0,004

0,006

0,008

0,01

0,012

0 200 400 600 800 1000 1200Temperature [ºC]

b)

Figure 4: Thermal dilatation, thε vs. temperature for, a) a cooling rate of 10 °C/s, b) a cooling rate of 0.3 °C/s.

Both cases gave the same amount of martensitic transformation. It should be noticed that it is assumed that pure martensite is formed. It should also be noticed that the martensite start temperature, Point 2, decreases when the cooling rate is increased. This has not been accounted for in the numerical model. The thermal dilatation for reheated martensite follows the martensite curve, Figure 4a.

89

A thermo-elastoplastic model based on von Mises theory was used. It was assumed that no creep strains occur during welding since the material is exposed to a high temperature for a very short period of time. The hardening behaviour of the material was assumed to be isotropic and piecewise linear. Transformation plasticity is not accounted for in the model. The temperature dependent Young’s modulus and Poisson’s ratio are shown in Figure 5a, whilst Figure 5b shows the virgin yield limit for the initial ferrite/pearlite mixture, yfσ , and the yield limit of

pure martensite, ymσ

σ y[M

Pa

]

0

200

400

600

800

1000

1200

0 500 1000 1500

σym

σyf

Temperature [ºC]b)

Heating,

Cooling, If Tpeak>850°C

If T peak<850°C

Heating,If Tpeak>850°C

Temperature [ºC]

0

50

100

150

200

0 500 1000 1500

00,050,10,150,20,250,30,350,40,45

E [

GP

a]

ν

νE

a)

Figure 5: Temperature dependent mechanical properties, a) Young’s modulus, E, and Poisson’s ratio, ν, b) Temperature dependent yield limit for the ferrite/pearlite phase,

yfσ , and yield limit for the martensitic phase, ymσ .

The yield limit of the material changes due to the phase transformation from ferrite/pearlite to martensite. If the peak temperature during welding has been higher than 850°C ( 3eA ) the yield limit follows the curve for martensite when the

material is cooled. The curve denoted yfσ in Figure 5b is followed during cooling

if the peak temperature has been lower than the 3eA -temperature. To avoid

90

convergence problems in the numerical calculations the minimum yield limit was set to 20 MPa and the maximum Poisson’s ratio to 0.45. The temperature dependent thermal properties are shown in Figure 6a. The latent heat of melting was set to 338kJ/kg, solidusT to 1480 °C and liquidusT to 1600 °C. The

emissivity factor used for the radiation boundary condition is shown in Figure 6b.

0

0,05

0,1

0,15

0,2

0,25

0,3

0,35

0,4

0 500 1000 1500

λ [ W

/m·º

C]

Temperature [ºC]b)

Temperature [ºC]

C [

J/kg

·ºC

]

0100200300400500600700800900

0 500 1000 1500

0

20

40

60

80

100

120

140

160

c

λ

a)

Figure 6: a) Temperature dependent conductivity λ , and heat capacity C. b) Emissivity.

5 Experimental

The component was TIG welded using an in-house robotised welding cell. The torch used is from Binzel AB (thoriated tungsten electrodes) and lined to a six-axis robot, ABB IRB 1400. The power source is a TIG Commander 400 AC/DC from Migatronic AB. The vane was spot welded between the outer and inner ring prior to welding the seam. Argon gas was used, both on the topside and on the root side to avoid oxidation of the component. No filler material was used during the welding.

91

6 Results and Discussion

The result of the robot program made off-line showed a high accuracy and very little adjustments after the calibration had to be made. Previously predicted temperature histories on a similar weld on the part showed an excellent agreement with IR measurements [10]. Example of temperature-time histories, located 4.2, 6.3 and 8.4 mm from the weld centerline, are given in Figure 8. Residual stresses were evaluated along three lines, located in the front, in the middle and at the back of the weld. The predictions are summarised in Figure 9 and 10 along each line. Figure 9 shows the calculated longitudinal stresses (i.e. in the weld direction) and Figure 10 the stresses perpendicular to the weld direction. Because of the properties of the shell model the stresses in the thickness direction are zero. The length axis in Figures 9 and 10, represents the distance from the weld center and the stresses shown were recorded after 200 s from the start time of the welding. The longitudinal stress level and distribution are similar along all sampling lines and the maximum longitudinal stress is about 800 MPa. The stress components perpendicular to the weld direction at different reference line locations show large differences. Compressive stress is generated both along the middle- and back reference line but this is not the case in the front. The reason for these differences is the different stiffness of the component along the seam. Reaction forces in a rectangular coordinate system located in a point along the weld between the outer ring and steel plate, see Figure 2b, are given in Figure 11. The forces are an example of a result that can be used to evaluate fixture solutions on a real part. Another example is given in Figure 12 where an optimization of the weld velocity to generate the lowest component deformation while keeping full penetration has been performed.

Several developments of the FEA models are possible. One simplification in the present model is that the tac welding (performed before the main weld) was not considered. This tac welding will most probably affect the stress level. The development of a new solid model, instead of a shell model, including these tac welds is planned. A validation of the predicted stresses by measurement is also planned. Including transformation plasticity in the material model will also change the stress state [6].

The present model can however be a powerful tool to qualitatively evaluate different weld parameters and fixture designs off-line.

92

Figure 7: Lines where evaluations were performed.

50 100 150 200

250

500

750

1000

1250

Time (s)

Tem

pera

ture

(

o C)

4.2mm6.3mm8.4mm

Figure 8: Temperature histories in three points along a perpendicular line to the weld seam.

Front Middle

Local co-ordinate system

Back

93

0 20 40 60 80 100 120

−200

0

200

400

600

800

Length (mm)

Res

idua

l Str

ess

(MP

a)

FrontMiddleBack

Figure 9: Longitude stress component along the three lines in Figure 7.

0 20 40 60 80 100 120

−150

0

150

300

Length (mm)

Res

idua

l Str

ess

(MP

a)

FrontMiddleBack

Figure 10: Stress component (perpendicular to the weld) along the three lines in Figure 7.

94

0 50 100 150 200

−8000−6000−4000−2000

02000400060008000

10000

Distance (mm)

For

ce (

N)

XYZ

Figure 11: Reaction forces along the weld between the outer ring and the steel plate (see figure 2).

0 50 100 1500

0.51

1.52

2.53

3.54

4.5

Distance (mm)

Spe

ed (

mm

/s)

Speed 1speed2

Figure 12: Weld velocity before (solid) and after optimization (dashed).

7 Summary and Conclusions

A simulation tool to define robot trajectories and to predict thermal histories and residual stress distributions on parts with complex geometries has been developed. The tool was evaluated on a part with a complex shape where robot weld paths were defined off-line, automatically downloaded to a FEA-model, where transient temperatures, residual stresses and fixture reaction forces were predicted. The described method seems as a powerful tool to construct and optimise torch trajectories and process parameters off-line. Further work in validating the residual stress distributions is planned.

95

8 Acknowledgement

The authors wish to acknowledge the assistance in the development of the material models by Andreas Lundbäck of Luleå University of Technology. The work was funded by the Foundation for Knowledge and Competence Development and EC Structural Founds.

9 References

[1] Y. Ueda and T. Yamakawa, Analysis of thermal elastic-plastic stress and stress during welding by finite element method, Trans. JWRI, Vol. 2 (90-100, (1971).

[2] Y. Ueda and T. Yamakawa, Thermal stress analysis of metals with temperature dependent mechanical properties, Proc. of Int. Conf. on Mechanical Behavior of Materials, Vol. III,10-20, (1971).

[3] H.D. Hibbit and P.V Marcal., A numerical thermo-mechanical model for the welding and subsequent loading of a fabricated structure, Comp. & Struct., Vol. 3 1145-1174 (1973).

[4] Radej, D. Heat Effects of Welding, p33, Springer Verlag, Berlin (1992)

[5] T.W Eagar, N.S. Tsai, American Welding Society Journal 62(12) 346-s to 355-s. (1983)

[6] F.G. Rammerstorfer, D.F Fischer, W. Mitter, K.J Bathe and M. D. Snyder, On thermo-elasto-plastic analysis of heat-treatment processes including creep and phase changes, Comput. Struct., Vol 13, 771-779. (1981).

[7] G. Bolmsjö, M. Olsson, K. Brink, Off-line programming of GMAW robotic systems – a case study. Int. J. for the Joining of Materials, Vol. 9 (3), 86-93, (1997).

[8] S Walter Simulation and Calibration for Off-line Programming of Industrial Robots, paper 54, Proc. of Computer Technology in Welding, Paris, June 1997.

[9] R.O. Buchal, D.B. Cheras, F. Sassani, J.P. Duncan, Int. J. of Robotics Research 8 (3): 31-43 (1989).

96

[10] M. Ericsson, P. Nylén, G. Bolmsjö, Three-Dimensional Simulation of Robot Path and Heat Transfer of a TIG-welded Part with Complex Geometry, 11th International Conference on Computer Technology in Welding, Colombus, Ohio, Dec. 2001, to be published.

simulation of robotic tig-welding - product development ... · simulation of robotic tig-welding...

Documents