~ Pra o egmn

DEFORMATIONS

Int. J. Mach. Tools Manufacl. Vol. 37, No. 3. I~'-309-318. 1997 1997 Published by Eltevier Scicace D d Printed in ~ Britain. All ~ rcsc,.rv~ 0~)O--6955/97$17.~ + .GO

PII: S0890--69a3~96)00067-3

AND PRESSURE DISTRIBUTION TOOL SLIDEWAYS

ON MACHINE

|. HOSEYIN

FiLiZ,t,

ABDULLAH AKPOLATt and |BRAHIM H. GISIZELBEYt

(Received 12 March 1996)

A l ~ t r a e t - - A finite element analysis of machine tool slideways has been carried out by using two-dimensional slideway models. Models are considered to consist of contact elements at the interface of sliding surfaces. Flat and dovetail types of slideways are modelled, and pressure distributions and deformations are determined with the use of ANSYS Rev. 5.0A non-linear contact elements. Close correlation of the results obtained from ANSYS with the experimental results given in the literature demonstrates the validity of the models used. Ease of construction of the models and faster solution times seemed to be the advantages of this method. The existing designs can easily be modified if the requirements on the allowable pressure distribution and deformations on the contact surfaces are specified. 1997 Published by Elsevier Science Ltd. All rights reserved

NOMENCLATURE
C

F k. I P A

constant relating normal contact deformation and pressure depth of the model coefficient of friction applied load contact stiffness length of a contact element constant relating normal contact deformation and pressure mean interface pressure normal contact deformation I. INTRODUCTION

Machine tool structures generally consist of fixed or bolted joints and sliding joints. Forces are transmitted across the joint interfaces and the overall static and dynamic charaeteristics of the machine tool are influenced by the compliance at these individual connections. The main requirements of these joints are the stiffness and the stability but, for sliding joints, the friction and wear properties must also be considered by the designer, since these can influence both the short- and long-term accuracy of the machine. Welded and bolted joints have received considerable attention from researchers. In addition to analytical and experimental methods, there are many works using finite element methods (FEMs). Slideways, however, have not received similar attention. Studies on contact deformations could be seen in the early 1960s. There were some studies on slideways in the 1970s. With the advent of high speed and high storage capacity computers, the analysis and design studies with FEMs were made easier after the 1980s. But, there are only a few studies which used the finite element technique for the determination of pressure distribution and deformation of slideways. Surface compliance for a range of mean interface pressures and surface finish used in the machine tool slideways has been analysed by Levina [1, 2], Tenner [3] and Ostrovskii [4]. They suggested an equation that is based on experimental data to describe the relationship between the normal pressure and the approach of the surfaces as h = c/'~ (1)

tUniversity of G-aziantep, Mechanical Engineering Department, Gaziantep, Turkey

:~Author to whom correspondence should be addressed. 309

310

i. Hiiseyin Filiz et al.

where ~ is the deformation or normal approach of the surfaces in ~m, P is the interface pressure in kgf/cm 2, and c and m are constants determined by the modulus of elasticity and the surface finish of the approaching surfaces. The factors affecting the relationship between the normal or interface pressure and the approaches of the surfaces are: material of the slideway elements, machining process, asperities height, orientation of machined lays, hardness, flatness deviation and size of the contact area. Connolly et al. [5] investigated the effect of material hardness on the initial approach and static stiffness of joints formed by surfaces machined with single point tools, and Thomley et al. [6] analysed the effect of surface topography upon the static stiffness of machine tool joints. The effects of these factors on c and m were evaluated in detail by Back [7], and he recommended that it is adequate to have an approximate knowledge of the surface compliance parameters obtained from a test, and concentration must be given more to the compliance of the components of machine tool joints, because the static stiffness of joints is more dependent upon the elasticity of the components surrounding the surfaces in contact rather than the compliance of the contacting surfaces. This does not mean that all the work on the surface contact compliance is unnecessary. Instead, such surface characteristics are necessary to formulate the solution of the complete joints. Masuko and Ito [8] analysed contact pressure distribution on the slideways under mixed lubricated condition with the use of the ultrasonic wave method. The slideway design methods which had been used commonly among machine tool designers were re-examined through the experimental results of contact pressure distribution, and it was confirmed that the design method is correct for most cases except for a heavy cutting machine tool. Hinduja [9] and Back [7] have investigated the deformation and pressure distribution in the joints by FEM. They have developed hydrostatic, plate and spring methods for modelling the contact between fixed and sliding joints. On comparison with the experimental results, it was shown that the finite element technique is a powerful engineering tool that can be applied to joints with fixed and sliding components. Furukawa and Moronuki [10] analysed contact deformation of a machine tool slideway and its effect on machining accuracy. They proposed a design policy to minimize the machining error by obtaining three-dimensional displacements of the slider caused by contact deformation. Hashimoto et al. [ 11 ] developed a method for contact stiffness estimation. They estimated contact stiffness by a beam model vibrating on an elastic foundation, and constructed an experimental equation as a function of the clamping load and surface topography. In this study, by using ANSYS Rev. 5.0A's non-linear contact elements, contact elements are defined between components of the machine tool slideways to determine deformations and pressure distribution under static loading conditions. 2. FINITE ELEMENTANALYSIS In modelling the slideways, the first thing to consider is the behaviour of the surfaces in contact after a force is applied to the elements. The model must be constructed such that deformations at each point in the joint and the pressure distribution at the surfaces in contact can be easily evaluated. The model must be simple, but adequate to give accurate results. In the models, it is always necessary to consider that the contact between the surfaces is elastic. If it were assumed to be rigid, then the calculated deflection of the slideways would be smaller than the actual values. This error cannot be neglected. For precise calculations of the deformations and pressure distributions, it is necessary to consider the compliance of the structural components surrounding the contacting surfaces. In general, qualitative influences of all factors on the behaviour of the joints are known from experiments and from simple theoretical models. A method of solution taking into account quantitatively all the factors that permit the calculation of the pressure distribution and deflections is still not available. For this purpose, the ANSYS Finite Element package, which can handle models consisting of non-linear contact elements, is used.

Deformations and Pressure Distribution on Machine Tool Slideways

311

In the ANSYS non-linear contact elements, analysis of surface-to-surface contact with large deformations, contact and separation, and sliding with friction is possible. General contact is represented by following the positions of points on one surface (the contact surface) relative to lines or areas of another surface (the target surface). The contact elements are used to track the relative positions of the two surfaces. The point-to-point contact element in two dimensions is Contactl2, planar and axisymmetric. These contact elements are finite elements that apply appropriate forces to nodes on two surfaces to account for contact and friction [ 12]. The solution accuracy depends on a number of factors such as mesh density, used element boundary conditions, analysis type and contact stiffness. Among them the contact stiffness of a surface is a procedure variable which should be improved during the loading process. The contact stiffness of a surface is the function of surface finish, pair of contact material and interface pressure distribution. It is obvious that the interface pressure is the only important factor for the contact stiffness. The contact stiffness is large for large mean interface pressures [11]. A new contact stiffness value must be calculated according to the new interface pressure for each element along the contacting surface during the loading steps. For an isotropic material the following relationship is valid [7]:
hk,, P - 106ld

(2)

where P is mean interface pressure, A is the contact deformation, l is the length of a contact element, d is the depth of the model, and k, is the contact stiffness for a contact element. From Equation (1) the following identity can be established:
I

c
or

106ld

(3)

]--m

ld(X) ,,k, = (10 l) cl/,"

(4)

At the initial stage of the process, a constant value is assigned to the contact stiffness for every contact element. By assuming that the interface pressure is uniform along the contacting surface, the surface compression can be calculated from Equation (1). By using Equation (4), the new contact stiffness values for each element along the surface in contact can be calculated. For the next iteration, contact stiffness values are calculated from previous deformation values. This procedure is followed until the difference in the successive deformation values is less than an allowed error.
3. EXAMPLES

Examples for fiat and dovetail types of slideway models are prepared to illustrate the determination of the deformation and the pressure distribution at the contacting surfaces of machine tool slideways. Dimensions and finite element division of a first flat joint slideway model is shown in Fig. 1. Contact elements are defined between contacting surfaces along lines AB, BC, DE and EF. The lengths of AB and DE are 30 mm and of BC and EF 20 mm, and the depth of the joint is 60 mm. The load was applied at the centre and plane stress was assumed in the analysis. The surface compliance parameters are c=0.69 and m=0.5, as defined in [7]. The coefficient of friction, f, for contact elements is taken as 0.1. The materials for both fixed and sliding elements are cast iron with the modulus of elasticity of 95 GPa

312

i. H/lseyin Filiz F/4

e t al.

Fig. 1. Finite element division of the joint and load application points. F is the total external load. and Poissons's ratio of 0.21 I. Mean interface pressure is obtained by dividing the total applied external load to total contact area of the joint. The deformed shape of the model for 1.6 MPa mean interface pressure is shown in Fig. 2. Deformations of the elements are obtained and pressures are calculated from Equation ( l ) at each pair of nodes along the contacting surfaces. Fig. 3(a) and (b) show the calculated pressure distribution for several mean interface pressures. The trends of pressure distribution are similar, but for high pressures, the distribution become less uniform, because of the influence of body detbrmation. At low mean interface pressures, maximum interface pressure is about the same as the mean interface pressure. The ratio of maximum interface

Fig. 2. Deformed shape of the joint for 1.6 MPa interface pressure.

Deformations and Pressure Distribution on Machine Tool Slideways

313

,-, el

---C~---0.$

IS.

~" .=,
IS

s
4

3.2 I . . . , R,,r~ I ~, ~.oR.,t'n I

\ \ y~

h
tJ

__o
, y

I\
+ +

\
\
:"

.~
o

. . . .

: ' , ' ' :

. . . .

. . . .

. . . .

. . . .

. . . .

:"~'~. I~1- -,-+ ,---I- -HCl

40 46 50

10

15

20

2S

30

36

Contact length (ram)

(a)

I :t
l

:o

--~--lzs

k
25 30 35 40 45 SO

,o
0 5 10 15 20

Contact Length (mm)

(b)
Fig. 3. Pressure distribution at the contacting surfaces for several interface pressures.

pressm'c to me~.n interface pressure increases with the mean interface pressure. A n d also the mean inted'~e pressure is increased, the contact length on B C and ~ becomes smaller. W h e n the deflections at the centre of the joint are plotted a g e n t the interface pressure, the diagram in Fig. 4 is obtained. It allows the designers to determine the required load for the allowable deflection. Non-linearity between deflections and mean interface pressure can be observed. The structural elements which are used in the second sfideway model are shown in Fig. 5. This is a typical example on closed-type sfideways. Depth of the joint, surface compliance parameters, other assumptions and material properties are the same as in the first example. In this problem, contact elements are defined along the fines AB, CD, DE and FG, HI, IJ. The lengths of AB and FG are 40 nun, CD and HI are 90 mm, and DE and IJ are 35 nun. Contact elements are not defined along the fines BC and GH, because when the joint is loaded due to the deformation, a separation at this surface will be developed. The deformed shape of the model is shown for 1.6 MPa mean interface pressure in Fig. 6. The calculated pressure distribution for five mean interface pressures is shown in

314

I. Htlseyin Filiz et aL

~'
~
0

.~+~+
20 40 60 Deflection ()un) 80 leO 120

Fig. 4. Relationship between the applied interface pressure and the deflection at the centre of the first example.

F~

F/4

F/4 tO n~n

I1111111 11III111 .11111111 I1111111 IIIIIIII I1111111

Illlllll IIIIIIII I1111111 IIIIIII1 Illlllll llllllll III11111 llllllll

11111111 11111111
I

~r X D

' iiilJJJJlJttiii,,II ~1]11111

mmmlmmmli

I1111111 IIil1111, , , , ,,,, II111lll I1111111 I1111111 I1111111

'IIIlllll" " "'

,.4JJ.kkkkk
Fig. 5. Finite element division of the joint and load application points.

I1111111

4-4"t"?"
,~L,lli~ I!~!! FI.' :x :

--i

,iiljii[

1.

Illl~
[llll
~llll

::::

I i I I I

~ r

....

I',III

~
IIIIIII

,
11 i.
11]

iiil
: : :t
....

IIII111

iiiiii
...... L

[I

Fig. 6. Deformed shape of the joint for 1.6 MPa interface pressure.

Deformations and Pressure Distribution on Machine Tool Slideways

315

Fig. 7(a) and (b). In this example, at low mean interface pressures, the maximum interface pressure is about 1.5 times the mean interface pressure, but at high mean interface pressures, the maximum interface pressure is about three times the mean interface pressure. As the third example, finite element division and position of the loads for a dovetail slideway is represented in Fig. 8. The material is cast iron and surface compliance parameters are c--0.25 and m--0.70, and the other parameters are the same as in the previous examples. The length of contacting lines AB and CD is 42.42 mm. The deformed shape of the model is shown for 1.6 MPa mean interface pressure in Fig. 9. Fig. 10(a) and (b) show the calculated pressure distribution at contacting surfaces of the dovetail slideway model for five different mean interface pressures. As the pressure increases, the length of contact decreases. In this example, for low mean interface pressures, the maximum pressure is about twice the mean interface pressure, but for high mean interface pressures, it is about four times the mean interface pressure.

I_+_1. I

~ , - ~ - -

o.8

10

2o

3o

4o

so

so

7o

.0

so

lOO 110

120

1~0

140

1so

lm

170

Contact Length (mm)

(a)

lO

10

20

30

40

SO

IO

"0

IO

I10

100

110

120

130

140

180

IM

170

Contact Length (ram)

(b)
Fig. 7. Pressure distribution at the contacting surfaces for several interface pressures.

316

i. Htlseyin Filiz et al.

Fig. 8. Finite element division of the joint and load application txfints.

Fig. 9. Deformed shape of the dovetail slideway model for 1.6 MPa interface pressure. 4. DISCUSSION AND CONCLUSIONS

Accuracy of shape and dimension of the workpiece, and high technical performance with economic efficiency are the basic factors considered in the design of machine tools. These are mostly dependent upon static and dynamic stiffness of machine tool structures. Machine tools have complicated structures and their analysis is extremely difficult by analytical methods. The models presented in this study cover some common types of joints used in machine tools. As seen in Fig. 3(a), there is almost 100% correlation with the results given by Back [7]. Stiffness and pressure distribution in machine tool joints can be calculated easily and precisely. Depending upon the deflection requirements, modifications such as changing the length of main sliding surfaces or auxiliary sliding surfaces, changing the length and locations of narrow guideways, and changing the geometry of the slideway systems could

Deformations and Pressure Distribution on Machine Tool Slideways

317

_o__o,i
_ 3.2 I

7.07

14.14

21.21

28.28

38.38

42.42

Contact Length (mm)

(a)

7.07

14.14

21.21

28. 28

35.38

42.42

Contact Length (mm)

(t))

Fig. 10. Pressure distribution at the contacting surfaces of the dovetail model for several interface pressures.

be done in a very short time. Ease of construction of the models and faster solution times are the advantages of this method. Careful consideration must be given to increasing the mean interface pressure, because maximum pressure may reach a value greater than the permissible value for the selected slideway material.
REFERENCES [I] Z. M. Levina, Calculation of contact deformations in slideways, Machines and Tooling 36, 8-17 (1965). [2] Z. M. Levina, Research on the static stiffness of joints in machine tools. Proc. 8th Machine Tool Design and Research Conference, Manchester. Macmillan, pp. 737-758 (1967). [3] D. G. Tenner, Contact stiffness of friction slideways, Machines and Tooling 39(3), 3-6 (1968). [4] V. I. Ostrovskii, The influence of machining methods on slideway contact stiffness, Machines and Tooling 36, 17-19 (1965). [5] R. Connolly, R. E. Schofield and R. H. Thomley, The approach of machined surfaces with particular reference to their hardness. Proc. 8th Int. Machine Tool Design and Research Conference, Manchester. Macmillan, pp. 759-775 (1967). [6] R. H. Thornley, R. Connolly, M. M. Barash and F. Koenigsberger, The effect of surface topography upon the static stiffness of machine tools joints, International Journal of Machine Tool Design and Research 5, 57-74 (1965). [7] N. Back, Deformations in machine tool joints. Ph.D. thesis, U.M.I.S.T., 1972. [8] M. Masuko and Y. Ito, Distribution of contact pressure on machine tool slidcways. Proc. lOth Int. Machine Tool Design and Research Conference, Manchester. Macmillan, pp. 641-650 (1969).

318

1. Htlseyin Filiz et al.

[9] S. Hinduja, Analysis of machine tool structures by finite element method. Ph.D. thesis, U.M.I.S.T., 1971. [I 0] Y. Furukawa and N. Moronuki, Contact deformation of a machine tool slideway and its effect on machining accuracy, International Journal of Japanese Socie~. of Mechanical Engineers, 30(263), 868-874 (1987). [l I] M. Hashimoto, E. Marui and S. Kato, Estimation of contact stiffness at interfaces in machine structures by a beam model on an elastic foundation, Tribology International 27(6), 423-431 (1994). [12] ANSYS User's Manual for Revision 5.0,4, Vol. IV, Theor)'. Swanson Analysis Systems Inc., U.S.A., 1992.