Modelling and Simulation of Dry Sliding Wear V. Hegadekatte (1), *, N. Huber (2), (3) , O. Kraft (1), (4) of Karlsruhe,(2) Technical University of Hamburg - Harburg, (3) GKSS-Forschungszentrum Geesthacht, (4) Forschungszentrum Karlsruhe *
[email protected] (1) University
Motivation: How to simulate wear and assess the lifespan of micro-machines ?
Strategy: Two fold strategy Courtesy of Sandia National Laboratory
Background:
1. Identify local wear model / parameter from experiment 2. In the future, use this local wear model to predict wear in micro-machines
Sliding / Rolling Contact
Micro-machines fail predominantly because of wear Dominant surface forces High operating cycles Dynamic contact Poor fabrication tolerances
Spinning Contact
Micro Planetary Gear Train (SFB 499)
GIWM:
Methods:
FN
Global → Average contact pressure Incremental → Contact pressure update
1. Global Incremental Wear Model (GIWM) 2. Finite Element (FE) based wear simulation tool (Wear-Processor) Wear Models:
Pin
Wear-Processor:
Fit
FN
50
tD
100
200
s
300
0.1 1/2
P
0 0
500
100
[m]
200
s
300
400
500
iang and Arnell (1998)
B:
s
= (F
N
1
FN
N
iang and Arnell (1998)
J
-0.6
GIWM
Fit
-0.8
300
350
Wea
400
ack
r Tr
[m]
Width
450
500
FN
GIWM =
40
N
-0.4
-0.6
-0.8 300
Prediction 350
Wea
400
ack
r Tr
[µm]
Width
450
500
[µm]
N=0 3 mm 5 mm
x
0.02
0
-0.01 -0.02
0.01
0.02
0.03 10 mm
5 mm
s = 0 mm
Simulation Geometry
z [mm]
σYY
0.3
Sliding Direction
kD
0.15
3
=13.5E-9 mm
/Nmm
-300 0 0
x [mm] 20
30
s
40
50
60
70
0.02
80
-0.01 -0.02
0
0.01
0.02
0 0
s = 71.36 mm
z [mm]
1/2
2
/ (E R k s)) P
C
D
s
N
=
=
400
mN
200
mN
h
Wear-Processor
10
C
P
D
s = 600
300
0.03
|z| [mm]
D
References
0.04
3
=13.5E-9 mm
0
100
200
s
300
400
/Nmm
-0.03
500
Fast tool for quick identification Wear model (Global Scale) P s is smaller Simple geometries, single wearing surface
-0.02
-0.01
z
[m]
• • • • •
0.00
0.01
mm mm
0.02
0.03
[mm]
Wear model (Local Scale) Ideal for studies of local effects General geometry P sis bigger Computationally expensive
We thank the German Research Foundation (DFG) for financial support through the collaborative research centre, SFB 499.
[1] V. Hegadekatte, N. Huber, O. Kraft. (2005). Finite element based simulation of dry sliding wear. Modelling Simul. Mater. Sci. Eng., 13, 57-75 [2] V. Hegadekatte, N. Huber, O. Kraft. (2005). A finite element based method for wear simulation in microsystems. In D. Löhe, J Haußelt (Eds.), Microengineering in metals and ceramics Part II. Wiley-VCH Verlag GmbH, Weinheim, Germany, 605-623 [3] V. Hegadekatte. (2006). Modeling and simulation of dry sliding wear for micro-machine applications. PhD Thesis, University of Karlsruhe, Shaker Verlag, Aachen, Germany, in print
Gordon Research Conference, 2006
mm
36.35
60.64 mm 71.36 mm
s
-600
-1200
= 0.349
0.05
0.00 11.83
s = s =
0 -300
-900
kD
0
Acknowledgements 0.02
s = s =
900
2
= 800 mN
FN FN
[N/mm ]
FN
20
ZZ
GIWM
w
[µm]
1200
Expts.
40 30
Verify each other 1/2
2
/ (E R k s))
200
300
400
500
600
N *103 [-]
σYY
Summary
Design Tool
50
GIWM
= 0.009148
= (F
100
[mm]
Experiment
•
0.1
0.03
Verification
•
Lower Flat Ring (Steel)
0.2 0.15
0.05
0.01
10
N = 6E5
Upper Hemispherical Ring (Brass)
0.25
s = 71.36 mm
-600
•
A:
0.01
20
=
r Tr
-0.4
0.0
-0.2
Spinning contact
0.04
0.00 0.00
ack Depth [µm]
J
0.0
-0.2
Wea
ack Depth [µm] r Tr
Wea
[µm]
[µm]
/Nmm
400
Guidelines
0.02
= 0.349
/ (E R k s)) C
mN
mN
h [mm]
[µm] 1/2
3
=13.5E-9 mm
0
-900
0.20
w [µm]
w N
400
200
10
kD
0
-1200
Disc GIWM
Front side of Pin
h (F
=
=
mN
10
Back side of Pin
P
R )
w
h
:
s
s
0.01 2
FN FN
s zz [N/mm 2 ]
A
e
1E-3
20
s yy [N/mm 2 ]
Pin
0.25
0.08
3
= 13.5E-9 mm /Nmm
= 0.009148
200
=
30
2·RD 0.30
3
1E-3
s
mN
0 0
0.06
1E-4
/Nmm)
= 800 mN
y
2·RWT
3
= 12E-8 mm /Nmm
B
1E-5 1E-5
400
h FN
0.01
= 21E-11 mm /Nmm
/ (
e
=
:
h
FN
w 20
FN
Sliding Direction
x [mm]
= 92E-8 mm /Nmm
0.01
1E-4
=10.2 E-9 mm
/Nmm)
s = 0 mm
Wear Track
Πs
3
D
3
-300
h
Πe D
kD
3
=13.5 E-9 mm
F 20 mm
Effect of elastic deformation
4
2 ai
-600
x
Linear wear from Archard (1980) Normal elastic displacement from Oliver & Pharr (1992) Dimensionless elastic displacement Dimensionless system parameter he 1 FN = π EC2 RP k D s h w RP 2
3
= 800 mN
-900
0
4
GIWM - Sarkar (
kD
40
FN
w
GIWM
30
-1200
z
0.00
3
Disc
2 a0
s zz [N/mm 2 ]
0.05
Si N /Si N , k
GIWM - Archard (
s yy [N/mm 2 ]
FEM Model
0.10
0.1
Expts.
Pin-on-disc
Local wear General geometry Elastic material model 2D and 3D FE model Wear on both surfaces Efficient re-meshing technique Handling of different wear rates Coulomb friction from experiment
D
Disc
Expts.
40
2·RP y
Steel/Steel, k
2 ai
Prediction
50
h
hw = k D p 1 + 3µ 2 s µ : Coefficient of Friction
DLC/WC, k
Disc
2 a0
Linear disc wear
Pin
Direction of sliding
Modified Archard‘s wear model (Sarkar, 1980)
1
Disc
FN
Pin
Linear pin wear
Pin
2aH i
Normal elastic displacement (Oliver & Pharr, 1992) Global linear wear integrated over sliding distance Fit and predict experimental data
hw = kD p s
D
FN
FN
Salient Features:
Archard‘s wear model (Archard, 1953)
PTFE/Steel, k
GIWM for disc wear
GIWM for pin wear
Identification of the included parameters in a wear model Two tools were developed to simulate pin-ondisc experiments - GIWM and Wear-Processor GIWM is an easy tool for fast identification and validation of wear models (existing or new) Wear-Processor is a computationally expensive, but a general tool, which has the scope for handling any arbitrary geometry A criterion based on the evaluation of Π e and Π s for choosing the more suitable of the above tools was given This work forms the basis for the prediction of wear in micro-machines
