Measurement of friction during metal forming process

Citation for published version (APA):Wang, S. L., Ramaekers, J. A. H., & Smeets, M. J. H. (1993). Measurement of friction during metal formingprocess. In Y. S. Jin (Ed.), IST '93 : proceedings of the First International Symposium on Tribology : October 19-23, 1993, Beijing (pp. 637-645). International Academic Publishers.

Document status and date:Published: 01/01/1993

Document Version:Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can beimportant differences between the submitted version and the official published version of record. Peopleinterested in the research are advised to contact the author for the final version of the publication, or visit theDOI to the publisher's website.• The final author version and the galley proof are versions of the publication after peer review.• The final published version features the final layout of the paper including the volume, issue and pagenumbers.Link to publication

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, pleasefollow below link for the End User Agreement:www.tue.nl/taverne

Take down policyIf you believe that this document breaches copyright please contact us at:openaccess@tue.nlproviding details and we will investigate your claim.

Download date: 03. Apr. 2020

Laboratory for Forming Technology

Department of Production technology and Automation

Faculty of Mechanical Engineering

Eindhoven University of Technology

Measurement of friction

during metal forming process

S.L. Wang, J.A.H. Ramaekers and M.J.H. Smeets

Oct. 1993 WPA nr 1640

The paper presented on the first International Symposium on

Tribology, Beijing, China, Oct. 19-22, 1993.

Proceedmgs of the International Symposium on Tribology, Beijing. China, 19-22 October 1993

Measurement of f.·iction during metal forming process

S.L. Wang, J.A.H. Ramaekers and M.J.H. Smeets

Laboratory of Forming Technology Department of Production Technology and Automation

Eindhoven University of Technology The Netherlands

ABSTRACT

With the development of numerical simulation of the forming processes, there is . an increasing importance for developing a more suitable friction description to be used in the numerical simulation codes. In this paper, a plane strain compression tribometer is developed which can measure the vertical force, friction force and vertical displacement simultaneously and accurately. Three different friction models . are evaluated by the experimental results. A modified friction model is proposed and some directories are given for future research of friction in metal forming processes.

1. INTRODUCTION

With the advances in computer technology, numerical simulation of forming processes is of increasing importance. Numerical simulation can give valuable insight into the effects of various process parameters on the product quality and tool design[1]. The material behaviour and frictional characteristics are very'· important inputs in numerical simulation. The friction distribution is crucial when the evolution of microstructure and prediction of forming defects (like crack initiation etc.) are included in the numerical simulation. Therefore frictional· characteristics is a main concern in the research of tribology in metal forming.

Friction is most commonly introduced in the analysis of forming process by the Coulomb model and the von Mises model. The Coulomb friction model is only valid under condition of relatively low contact pressure which is rarely found in bulk metal forming processes. The von Mises model is valid only when the contact pressure is relatively high. The Coulomb model and the von Mises model are· usually used in a combined way in the analysis of forming processes.

Wanheim and Bay[3] have developed a generalized friction model by analysing the asperity deformation using the slipline field theory. The physical background of this model is that with the increase of the contact pressure, the plastic deformation fields around individual asperities will interfere. This model allows for a gentle transition from the Coulomb model towards the von Mises .

637

J;~d~i'~-Th'i~"mo'd-~'I-'~~'IY"~'~'Il-~er'n~-th-~-~s'peritydefo':mation"orthe'surfa~'e-iayer:-fn' ~etal forming the local asperity deformation interacts with the bulk deformation ~ield, so that the friction is not only determined by the interface boundary layer but also strongly influenced by the bulk deformation of the substrate material. I Ramaekers and Kals (R-KH 4] proposed a friction model based on the ~xperimental observations: Tfr = qpuA/Ao. Here p: normal pressure; u: relative I

cilisplacement between tool and workpiece; A/Ao: the surface strain. This model ¢omprises the substrate deformation by introducing relative displacement u and ~urface strain AI Ao. I This paper presents a newly developed tribometer based on plane strain Gompression and the preliminary experimental results obtained from it. The $xperimental results are compared with theoretical results derived from slab r!nethod. Different friction models are evaluated and it is found that the R--K model ~est fits the experiment. However,' the coefficient in the R--K model is found to be ~arying during the forming process. In order to improve the model, a modification 6t the R--K model is proposed. Finally discussion on the present work and ~roposition for future work are made.

!

2. DESIGN OF THE TRIBOMETER ,

to get the friction coefficients in various friction models, it is necessary to measure the friction force, contact pressure, relative slip velocity, interface temperature etc. ~imultaneously, simply and accurately. There is no universal test procedure in metal I

fiorming tribology. Most bulk forming processes are compressive processes. This i!s a reason for choosing plane strain compression as the prototype of bulk forming precesses. ,

--

--force transduser

Fig. 1 Principle of the tribometer

.. In Fig. 1, the principle of the tribometer is given. The workpiece is plane $train compressed, in such a way that the friction force is exposed to the force transducer and detected by a Kistler load cell. The amount of side surface friction Ffr2 in total friction Ffr is reduced by using Teflon foil as lubricant in the side surface and by choosing the workpiece dimension Lo/Ho = 2.5, leading to Ffr :::: 2Ffr1 •

Fig. 2 is the schematic drawing of the tribometer. A high stiffness die base is used to assure the plane strain condition. Also changeable upper and lower platens are used for evaluating different tool surface finish or tool coatings. The

.. '''-''''-'~'''-.-.-- .... --.---.--.----... ---... -.. -.---.. --. ___ .. ~I 638

........ --.~ .. -----... ---- -""-"-"'-"-----'----'---1 3 2 11 !

.--.-.--~~~rr--~~--~ I

1

(a). front view

(b). bottom view

Fig. 2 Schematic drawing of the tribometer 1 . die base 2, 6. steel blocks 3. upper tool 4. workpiece 5. lower tool 7. Kistler load cell 8. adjustable side platens 9. side platen 10. pre-stress screw 11. position fix screw

Amplifier

o Numeral display ~III;~ L-____________ ~ ~ F.~~

L-______________ ~~

Kistler cell

Computer

Hydraulic system

Vertieal force sensor

Fig. 3 Experimental arrangement

639

I I i

~'i~t~~~"~-r~' '~~-;d--f~'~-'~~~'I~-~ti~-g'dif'f;'~;'~t-t~~r;~rf~~;-f-i~ish -'~;-t~-~i~~~ti'~g~~' Th~ 4entre line of the Kistler load cell is set to the average height of the workpiece with II

~he consideration that relative reduction is 40% (see Fig. 2). i In Fig. 3, the experimental arrangement is schematically drawn. Besides I friction force, the vertical force and vertical displacement signals are also recorded I ~uring the forming process. The three analog output signals are first converted by Ii

ell pel718 data acquisition labcard (AID conversion), the digitalized signals are I

~ollected and stored in a computer by using a special made software triggered I trogram. I 3. EXPERIMENTAL RESULTS ;

~sing the newly developed apparatus, the mean values of friction coefficient p, friction factor m and friction constant q during the forming process can be derived ~see Appendix B). Using the different friction models in the slab method analysis 6f plane strain compression[7] (see Appendix A), the theoretical load-displacement rlelation are derived which are used as a comparison with the experimental data. !

*.1 Experimental material and procedure

: The material used in experiment is soft aluminum. The workpiece dimensions ~re Wo *Bo *Ho = 50.02*25.00*19.84mm3. Forthe uniformity ofthe surface finish J

<;tuality, all the workpieces are processed by means of abrasive powder. The ~astagaev test at average strain rate t = 8 * 10-4 S-1 showed the flow stress is: trf = 120* 8°·24.

I I I i

II

The surfaces are polished to have I i The tool material is high speed steel. r:oughness values Ra = 1.25 - 2.5 pm. i :;3.2 Experimental results I : Two sets of experiments are carried out at room temperature with average strain rate of e = 8 * 1 0-4 s-1. One set (a) uses Tallow as lubricant, another set (b) I

~ses no lubricant. Fig. 4, 5 and 6 show the results.

;

!~ ~~~ ''0 !.3 160

120

80

~300 '0 250

lOS .s 200

150

100

50

I

i

I

I I

I g.OO 0.05 0.10 0.15 0.20 0.25 II

(a) Tallow lubricant (b) no lubricant

_, '._. _., __ .. __ .,'" ~i9 ~_~_ R:pr~~uc i b i I i~_~::a s~~~.me_~~~_____ ___ '_". __ , ___________ .J

40 ~FlfHo O~--.---.---,--,,--.---,

0.00 0.05 0.10 0.15 0.20 0.25 0.30

640

I FromF'ig~4;'it"~'~'~-b'~~~-~~-th~t-'th;-r~'prod~cibility i~ ~-eryg~~d :-This shows' rat the new apparatus works well and the experiments are controlled well.

I ~ 0.04 14)

~ 0.03 I~ ~ 0.02 I

m ~ 0.6 'U 0.5 IE gO.4 C,) 0.3

0.2

0.1 I

I 0.01

aH/Ho i 0.00 -F:::lIO<...r----.--,--,-----r--, 0.0 aH/Ho , 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.00 0.05 0.10 0.15 0.20 0.25 ,

(a) Tallow lubricant (b) no lubricant Fig. 5 Coefficients jJ, m and q

From Fig. 5, it can be seen that the coefficients in all the three friction ~odels are not constant during the forming process. jJ and m increase with the ~unch travel, q first increases rapidly then decreases dramatically with the punch ~ravel. So all three friction models find difficulty to be used in forming processes, the search for a true friction constants has to be continued. !

~250 Experiment ~ 300

Experiment

R--Kmodel 250 R--K model t'i 200 '0

is 150 .3 200

100 150

100

50 l1H/Ho

0 0.00 0.05 0.10 0.15 0.20 0.25

(a) Tallow lubricant (b) no lubricant Fig. 6 Load--displacement curve

In Fig. 6, the comparison of the load--displacement curves is given. Using the experimental results of jJ, m and q and the theoretical derivation by means of the slab method, three load curves corresponding different friction models are ~btained. Firstly, it can be seen that the slab method analysis is strongly influenced by the friction distribution at the interface introduced by different friction models. $econdly, the load predicted by the slab method is lower than the experimental tesults. Both the load curves by the Coulomb model and the von Mises model ~nderestimate the load. The load curve by the Coulomb model has the largest discrepancy from the experimental results. The load curve by the R--K model agrees very well with the experimental results. This implies that the R--K model predicts a more reasonable friction distribution .

i

I . .. ....... . --.---.-.. __ ._. _____ . ______ .. ___ ........ ___ ... ______ J

641

,- .... ' .. - ...... --, ....... --_ .. _ ....... ,._" ... _ .... , ... ,,, ......... --_ ........ , ........ , ............ ,, ...... - --'-'''' ... _ ..... _- --.-.--.. - ..... ---- .. -- ... - ...... ---.-----.--- .. ""--". "'-- """---'-'j

:B.3 Modification of the R--K model i I The proposition of the R--K model is based on assuming the friction rjnechanism is mixed film lubrication (Fig. 7). When the shear stress of the lubricant fl can be neglected compared with that of the boundary film Tb, then the friction ~tress is expressed as: I

) Tfr = Art ATb

1 In the R--K model, it is assumed that real contact area increases with normal pressure, relative displacement between tool and workpiece and nominal surface

I

~train, the expression is: Tfr = qpuA/ Ao. From the experimental results, the R--K-rlnodel can predict load very well but the coefficient q in the R--K model is not a 90nstant during the forming process. Here a modification on the R--K model is ~ade: assume the real contact area A/A increases with normal pressure and ~elative slip velocity Us between tool and workpiece: Tfr = q'pus • The physical I

rineaning is that real contact area is influenced both by normal pressure and the flow of the substrate material.

The coefficient q' in the modified R--K model is calculated from the experimental results (Fig. 8). It can be seen that coefficient q' first increases very quickly and when reduction is .ilH/Ho > 0.05, q' is oscillating around a constant value.

.c;"' 1.2 I

~ 1.0 *00 0.8

---1T 0.6 0.4

0.2

0.0

jp Normal stress

Friction sterss 'tfr

Relative slip velocity Us

Workpiece (af) A

Nominal contact ar

Fig. 7 Mixed film lubrication

0.00 0.05 0.10 0.15 0.20 0.25 0.30

(a). Tallow lubricant I . __________ 1

642

. ... .. _ .. --............... _ ... -··"'·"··-1

4

AH/Ho o~----~--~----~----~--~

0.00 0.05 0.10 0.15 0.20 0.25

(b) no lubricant Fig. 8 Coefficient q'

Here is an explanation of the oscillation of q'. In this experimental set-up, the ~ertical displacement is detected by an inductive sensor which gives integer pitches of the inductive coil (0.01 mm). The average strain rate is e = 8 * 10-4

S-1.

When calculating q', the differentiation of the vertical displacement in every 5s (about 0.07mm) is used. So the discontinuous vertical displacement signal is an i'mportant cause of the oscillation of the q' curve.

I

! When the relative slip velocity is constant (no bulk deformation of the ~ubstrate material), the modified R--K model turns back to the Coulomb model. i The nearly constant coefficient in the modified R--K model allows it to be ~asily implemented in forming analysis. When the relative reduction rate ~H/Ho < = 5%, assume that the q' increases linearly with the reduction rate, when teduction rate ~H/Ho> 5%, assumes that the q' is taken as a constant. ! More experiments will be necessary to see the influence of the punch travel velocity on the coefficient q'. ! I r- DISCUSSION AND CONCLUSIONS

The friction in metal forming processes has a unique feature: it is influenced ~y both the deformation of the surface layer asperity and the deformation of ~ubstrate material. Therefore, it is necessary to combine the analysis of the ~eformation of surface asperity and the deformation of substrate material. Due to the complexity of the process models, even with many simplifications, the pure theoretical and numerical analysis is difficult. The combination of physical ~imulation with theoretical and numerical simulation is a promising way in the analysis of friction in metal forming. The Moire method[8] is an effective way for ~nalysis of plastic deformation, so that the future work will be conducted using Moire method to verify the modified R--K model and to explore new models by I

taking into account both the deformation of surface layer asperity and the deformation of substrate material.

Some special points can be concluded from the present work: The newly i developed tribometer works well and can get the coefficients in various friction I

.1

643

I

I I

rh~'d;I~; --~~li~bIY-.-Th-~-;i;b-~;ti~~d"~~--~~I-t;;~~--~t~~-~~iIY---i'~--fi ~~~~~-d---by- th~"'f~'i~tjo n---I __ ~istribution due to the introduction of a certain friction models. The R--K friction I rjnodel predicts forming load best. A modified R--K friction model is proposed and j 9an be easily used in forming analysis with improved accuracy. A more general :

-- ftriction model should be developed by taking into account both the deformation of I surface layer asperity and the deformation of substrate material. i i I I r 5. REFERENCE I I ! ~1]. S. Kobayashi, S-1. Oh and T. Altan, Metal Forming and the Finite-Element I Method, Oxford University Press, 1989 !

~2]. J.A. Schey, Tribology in Metal Working: Friction, Lubrication and Wear, American Society of Metals, Metals Park, Ohio 44073, 1983 [3]. T. Wanheim and N. Bay, A Model for Friction in Metal Forming Processes, ~nnals of the CIRP, VoI.27/1/1978 [4]. J.A.H. Ramaekers and J.A.G. Kals, Mathematical Representation of Friction in I

Metal Forming Analysis, Annals of the CIRP, VoI.35/1/1986 I

[5]. W.R.D. Wilson, Friction Models for Metal Forming in the Boundary Lubrication ~egime, ASME Journal of Engineering Materials and Technology, 113(1 )(1991 )60 [6]. A. Nagamatsu, T. Murota and T. Jimma, On the Non-Uniform Deformation of I

Block in Plane-Strain Compression Caused by Friction, Bulletin of the JSME, Vol. ~ 3, No. 66, 1970 r7]. S.L. Wang, unpublished work, 1993 I

(8]. O.X. Cao, S.Y. Ye, B. Xie and X.T. Ma, Principle and Applications of Moire ~ethod, Tsinghua University Press, 1983 I i

APPENDIX A [

i Slab method analysis I (

dx do =2*"t'6-X 'f H

8

CJ z= ~f"t' ,,dx+c o

(A 1)

(A2)

(A3)

Introducing different friction models into (A3), the theoretical load equations ~an be obtained:

The Coulomb model: Tfr =PP

I .. --- .--------- ... -.- --.-------. -- ----.- .. -.... -.--~- .. ---,--- -. _ .. -...... ---- ...... -.. -------. --- --.. ---... -------_ . ___ ._ ._--.. ---- .... ______________ . ____ .1

644

,,_,_, _" ____ , __ ,_"_ "' __ ,. _".", ,'" _"_"",,_, ,'' __ ' __ cc,', ',c,_ ,_ .-',_,_._. __ ,_, ____ ... _, ___ .. ' ___ ,_ ,_ , __ .' __ ,_" _____ , ____ ' __ ' __ ----,

The von Mises model: Tfr = mK = mUfN3

p = WBa'(2+ mB) VITI /3 H

(A4) r

I (A5) I

! The R-K model: Tfr = qpuA/Ao, the friction stress Tfr and contact stress Uz are I

obtained, then the vertical force Prk and total friction force Ffr are obtained by the [' humerical integration of Uz and Tfr •

a =-~a,exp[ qIlH(B2_x~] (A7) I z.f3 H2 I

2a,qxllH [qIlH(B2 2\] -.:....---exp -- -x J /3H H2

APPENDIX B i

i

perivation of Ji, m and q

/3F,r m=----2*B*W*a,

(AS)

(81) "

(82)

I I

Integration of (A 7) and (AS) give the total vertical force and friction force, ~nd q is obtained as:

I' I

i F.H q=1.334 'r (83)

NOMENCLATURE ! i

P K u Us

Wo, 80 , Ho W, 8, H LlH Pel' Pvrnt Prk

Ffr

Fver

IlHBFv8r

: normal pressure : shear strength of the specimen : relative displacement between tool and specimen : relative slip velocity between tool and specimen : initial length, width and height of the specimen " : current length, width and height of the specimen : punch travel : load by slab method correponding to the Coulomb, von Mises ,'I'

and R-K model respectively : measured friction force

. _: meas~r:d=e~iear fore: ___ .... _ .__~_ __ _______ J 64S