26/09/2017 1 Fretting wear of low current (DC) electrical contacts: quantification of electrical endurance S. Fouvry 1 , J. Laporte 1 , O. Perrinet 1 , P. Jedrzejczyk 1 , O. Graton 1 , R. Enquebecq 1,3 , O. Alquier 2 , J. Sautel 3 1 Laboratoire de Tribologie et Dynamique des Systèmes, Ecole Centrale de Lyon, France 2 PSA, Vélizy - France 3 Radiall,Voreppe, France 1 (contact : [email protected]) 2 Fretting Group @ LTDS Fretting Wear, Fretting Fatigue Fretting & Electrical Contacts Plateforme 2DF Durabilité, Fretting & Fatigue Plateforme numérique Bureaux des chercheur de la thématique fretting Lyon Ecole Centrale de Lyon
24
Embed
Fretting wear of low current (DC) electrical contacts ... · PDF fileelectrical contacts: quantification of electrical endurance ... 2 PSA, Vélizy - France 3 Radiall ... Cross Cylinder
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
26/09/2017
1
Fretting wear of low current (DC)
electrical contacts: quantification
of electrical endurance
S. Fouvry1, J. Laporte1, O. Perrinet1, P. Jedrzejczyk1,
O. Graton1, R. Enquebecq1,3, O. Alquier2, J. Sautel3
1 Laboratoire de Tribologie et Dynamique des Systèmes, Ecole Centrale de Lyon, France 2 PSA, Vélizy - France 3 Radiall,Voreppe, France
“Fretting mechanical criterion” If δ* < δt (PS/GS transition) => inner stick metal zone => R low & stable (infinite endurance) If δ* > δt (PS/GS transition) => full sliding => Wear => R rises (finite endurance)
0.0001
0.001
0.01
0.1
1
10
100
1000
0 1 2 3 4 5 6 7 8 9 10
R (Ω)
partial slip
gross slip
displacement amplitude, δ* ( µm)
4 t µm
-0.8
-0.4
0
0.4
0.8
-6 -4 -2 0 2 4 6
δ (µm)
Q*/P
-0.8
-0.4
0
0.4
0.8
-6 -4 -2 0 2 4 6
δ (µm)
Q*/P
N = 10000 cycles
S. Hannel et al. Wear 249(9), 2001
26/09/2017
6
11
Correlation between fretting sliding condition and ECR behaviour
displacement amplitude, *
Stick zone
sliding zone
Partial Slip
Q Q*,*
R()
time (fretting cycle)
low & stable ECR
Infinite endurance
full sliding
Q
2δ*
a
Gross Slip
t
Q*,*
R()
time (fretting cycle)
finite endurance
ECR increase
tangential force
amplitude Q*
1.0
10
100
1000
Ag/Ag Au/Au Sn/Sn
Fretting cycles
Ele
ctr
ical c
onta
ct
resis
tance (
10
-3
)
0. 1
101 102 103 104 105 106 107 100
Sn/Sn
Ag/Ag
Au/Au
Comparison between Noble & non Non noble coating
Non noble : very fast decay
Applied test conditions: Temperature: 25°C Relative humidity: 10% Frequency: 30Hz Normal Force : 3 N Displacement : 8 µm Thickness coat. : 1.3 µm
Noble & semi noble : Delay before EC failure !
GT GP
Rc Rc
*< t *> t t
GT GP
Rc
Rc
*< t *> t t
Rc When non noble substract reached: ECR failure !
Wear (delay)
Non noble (Sn alloy)
Noble (Au & Ag)
12 S. Hannel et al. Wear 249(9), 2001
26/09/2017
7
2. Quantification of Nc (Electrical Contact Endurance) versus displacement & sliding amplitude ?
13
0
5
10
15
20
25
30
1 100 10000 1000000 1E+08
t_Au= 5µm
Finite endurance Domain (GS)
Infinite endurance Domain (PS)
Ap
plie
d d
isp
lace
men
t am
plit
ud
e
(µ
m)
fretting cycles, Nc (ΔR>4mΩ)
Au/Au interface
Electrical Endurances a function of the applied displacement amplitude
Asymptotic decreasing of NC (DRc=0.004 threshold): the larger the displacement the smaller the ECR endurance Nc
Applied test conditions: Temperature: 25°C Relative humidity: 10% Frequency: 30Hz Normal Force : 3 N Thickness coat. : 1.3 µm
14
26/09/2017
8
Quantification of the electrical endurance Curve : ”Fatigue like” approach :
perfect correlation between experiments and the exponential formulation (only 3 variables : δt , n, Ncδ
0
5
10
15
20
25
30
1 100 10000 1000000 1E+08 1E+10
fretting cycles, Nc (ΔR>4mΩ)
app
lied
dis
pla
cem
ent
amp
litu
de:
*
(µm
) n
t
NcNc
)(
*
y = -3.1898x + 16.748
R²= 0.9904
0
2
4
6
8
10
12
14
16
18
0 0.5 1 1.5 2 2.5 3
ln(N
c)
application to Au/Au interface
)ln( *
t
B
-n
BnNc t )ln()ln( *
With n = 3.18, B = 16.8 and µmt 5
Power law n
t
NcNc
)(
*
exp(B)
1µm) ( *
twhenNcNc
15 S. Fouvry et al. Wear 271 (9–10), (2011)
Comparison between coatings
- Small difference between δt transitions - Large difference between non noble Sn and noble Au&Ag ECR endurances - All the ECR endurance can be formalised using a simple power law function
Applied test conditions: Temperature: 25°C Relative humidity: 10% Frequency: 30Hz Normal Force : 3 N Thickness coat. : 1.3 µm
16 S. Fouvry et al. Wear 271 (9–10), (2011)
26/09/2017
9
Sliding amplitude formulation
The measured displacement depends on the test compliance => Results affected by the test signature !
measured δ (± µm)
real contact displ. δC (± µm)
test apparatus accommodation δA = Q x CA
ACAC CQ
δA : tangential test apparatus accommodation CA : test apparatus compliance
310 410 510 610 710 810 910 1010
Nc
0
5
10
15
20
25
30
1110
n
NcNc
)( 0
0 (±µm)
“power law function”
Ag/Ag Q(N)
(µm) displacement
Tangential force
*Q
δ*
δ0
δ0 : Sliding amplitude (=> δ when Q=0)
t *0
17 S. Fouvry et al. 58th IEEE Holm , 2012, 191-203
18
What about herogeneous interfaces ?
0
2
4
6
8
10
12
14
16
18
100 1000 10000 100000 1000000 10000000 100000000
fretting cycles, Nc
x10 /30
0 ( )µm
4
1.90
2.05 10Nc
5
20
2.03 10Nc
Ag/Ag
Sn/Sn
Ag/Sn7
2.80
4.74 10Nc
Ag/Sn still controlled by noble/noble fretting wear response (ECR failure is delayed by the coating wear) : But the formation of abrasive Sn oxides accelarate the Ag surface wear => mitigate benefit of Ag (only x 10 compared to Sn/Sn !)
O. Perrinet et al., ICEC 2014, p. 114-119.
26/09/2017
10
3. Correlation between fretting damage (surface wear & oxide debris)
& Electrical behavior
19
Study of the endurance degradation for homogeneous Ag/Ag interface
0
0.002
0.004
0.006
0.008
0.01
0 20000 40000 60000 80000 100000
Ré
sis
tan
ce
(Ω
)
Fretting cycles
23000 50000 83000
9500096000
107000
97300
ΔR<ΔRc
ΔR>ΔRc
Interrupted tests at different fretting cycles to follow the electrical degradation
Characterization of fretting scars (SEM, EDX 3D profil)
Prediction of ECR endurance: energy approach Application of friction energy approach
Nc prediction requires a local wear approach
z (µm)
)(xmax
maxh
Wear Profile
energy density
with wear ϕ(x) converge to a flat profile
Very nice prediction
(low dispersion)
0
200000
400000
600000
800000
1000000
1200000
1400000
1600000
1800000
2000000
0 1000 2000 3000
φf = Ed/Af (J/m²/cycle)
P=6NP=5NP=4NP=3NP=2NP=1N
𝑁𝑐 = 𝑁𝑐0.𝜑𝑓𝛽
Nc0=6.1012cycles
and β=-2.58
Nc (cycles)
Ag/Ag HR=10% e=2 µm f=30Hz T=25°C
)( f
NcNc
Ncϕ= 6 1012 cycles β=2.58
Af=A final
fretting loop
contact area
ff
Ed
A
29 C. Mary et al. Wear 263 J. Laporte et al., Wear 330-331 (2015), 170–181.
Prediction of ECR endurance: energy approach Simplified approximation
Possibility to approximate Af using a single power law function of normal force: with A0=72,600 µm² and m=0.256
Previous analysis requires the measurement of Af (long & fastidious)
0
20000
40000
60000
80000
100000
120000
140000
160000
0 2 4 6 8
P(N)
A (µm²)
𝐴𝑓 = 𝐴0 .𝑃𝑚
The analysis requires the computation of Ed! (integrale of the frettting loop) Considering a quadratic shape:
04Ed P
µ.P
02
30 Ed
26/09/2017
16
Prediction of ECR endurance: energy approach Simplified approximation
1000
10000
100000
1000000
10000000
1000 10000 100000 1000000 10000000
Nc
exp
(cycle
s)
Ncth (cycles)
e=2µm, 2µm<δg< 16µm
P=6NP=5NP=4NP=3NP=2NP=1N
Very good correlation between experimental lifetime and
theoretical prediction
𝐴𝑓 = 𝐴0 .𝑃𝑚
Ed
ANc
NcNc
f
f )( )1(
0
0 1
4 mPµ
ANcNc
31 J. Laporte et al., Wear 330-331 (2015), 170–181
PµEd ...4 0
Prediction of ECR endurance: energy approach Influence of the coating thickness
32
A thicker coating induce a significant lateral contact extension!
VAg α eγ with γ>1
e
Parabolic evolution:
𝑁𝑐
𝑁𝑐𝑟𝑒𝑓=
𝑒
𝑒𝑟𝑒𝑓
𝑝
p=2.85 eref=2µm Ncref=Nc(2µm)
0
200000
400000
600000
800000
1000000
1200000
1400000
0 1 2 3 4 5 6
Coating thickness (e)
Nc (cyclesAg/Ag P=3N δ0= 9µm HR=10% f=30Hz T=25°C
J. Laporte et al., Wear 330-331 (2015), 170–181
26/09/2017
17
Prediction of ECR endurance: energy approach Global formulation
Very good correlation confirming the proposal !
Ncth=f(P, δ0, µ)
𝑁𝑐
𝑁𝑐𝑟𝑒𝑓=
𝑒
𝑒𝑟𝑒𝑓
𝑝
Ncth=g(P, δ0, µ, e)
Normal force
Sliding amplitude
Friction coefficient
Thickness
1000
10000
100000
1000000
10000000
1000 10000 100000 1000000 10000000
Nc
(experi
me
nta
l) (c
ycle
s)
Nc(predicted) (cycles)
e=2µme=3µme=4µme=4.8µm
Eq. (21)P=3N,
2µm <δ0< 16µm
Eq. (15), e=2µm, 2µm <δ0< 16µm
P=6NP=5NP=4N
P=3NP=2NP=1N
p
ref
m e
e
Pµ
ANcNc
)1(
0
0 1
4
33
J. Laporte et al., Wear 330-331 (2015), 170–181
5. Complex Fretting-Reciprocating slidings
34
26/09/2017
18
before insertion after insertion
repetitive insertions
Surface degradations
Influence of repetitive clipping & uncliping slidings
pin
clip flexible
?
0.0001
0.001
0.01
0.1
1
0 100000 200000
R [Ω
]
Nombre de cycles N
electrical failure
Nc Nc
pin
clip flexible
35
tangentiel forcesensor
flexiblestrips
samples
upper sampleholderflexible
strips weight
electromagneticlinear motor
(reciprocating)
electromagneticshaker
(fretting)
laser sensor
9 µm
ΔR >ΔRc
δ0= 9 µm (30 Hz)
Test stopped when
D
D= 250µm to 1 500µm vGC,ref= 8,3µm.s-1 to 124,5µm.s-1
fretting Nf 5 000 à 60 000cycles
Ag/Ag HR=10% e=2 µm f=30Hz T=25°C
Experimental strategy
Real clip assembly
fretting track
reciprocating track
J. Laporte et al., Wear 376-377 (2017) 656–669.
26/09/2017
19
N=60 000 cycles
0 = 9µm
Large sliding
D=1mm
R = 2.14mΩ A
δ*g
[Ag]≈17at%
δ*g
Fretting zone
Reciprocating zone [Ag]≈31at%
[Ag]≈87at%
refilling process : Ag is transferred from the external reciprocating track scar to the fretting scar ! => The application of reciprocating increase the ECR endurance !
Ag/Ag RH=10% e=2 µm f=30Hz T=25°C
Effect of reciprocating sliding regarding ECR fretting response
37
R = 2.14mΩ
B
plain frettingfretting-reciprocating
δ*g
δ0
0
2
4
6
8
10
12
0 50 000 100 000 150 000 200 000
fretting cycles (N)
ΔR=4mΩ
plain fretting(D=0µm)
Nc= 100,000 cycles
reciprocating sliding
ele
ctr
ical
co
nta
ct
res
ista
nc
e,
R (
mΩ
) Nf = 10000 cycles
Fretting
N cycles
D
9 µm
reciprocating stroke (D= 1 mm)
fretting
sequence
fretting & reciprocating
Nc= 201,700 cycles
300 µm
[Ag]=2.2at%
[O]=12.8at%
[Ag]=3.4at%
[O]=43.6at%
[Ag]=5.17at%
[O]=48.13at%
All Ag present in reciprocating track is transferred
max (0)
(0)
,1 /f f tr
Nc NcNc Nc
N N
0
5
10
15
20
25
0 2 4 6 8 10
x 1
00
00
Nf,tr
plain fretting
(NR=0), Nc(0)
Ncmax
x104
x104
NR=0
NR=1
NR=2
NR=4
NR=10
NR=20
NR=40
fre
ttin
ge
nd
ura
nce
, N
c(c
ycle
s)
(Eq.23)
fretting cycles between eachreciprocating sliding, Nf (cycles)
Nf too long => no transfer
Influence of fretting block (periodicity of large sliding) ?
38 J. Laporte et al., Wear 376-377, (2017) 656–669.
Nf
Nc : total number of fretting cycles before ERC failure
26/09/2017
20
39
Influence of reciprocating stroke ?
0
5
10
15
20
25
30
35
40
0 250 500 750 1,000 1,250 1,500 1,750
x 10
000
reciprocating stroke, D (µm)
Dc
Dth
plain fretting,
NcPF
x104
fre
ttin
g e
nd
ura
nce
, Nc
(cyc
les)
Non monotonic evolution
Nc prediction (global formulation)
40
Vc : total Ag volume involved in fretting wear process
0
1
2
3
4
5
6
7
8
9
10
0 500 1,000 1,500 2,000
reciprocating stroke, D (µm)Dth
Steady-state
PFfV ,
ccf
NVV
we
ar
rate
,
(µ
m3/
cycl
e)
Nf=Nf,ref =10,000cycles250µm D 1,500µmplain fretting
PFfreff VhV ,,
determined for Dref=1mm
with h=1.66
v
tr
PFf
D
D
hhVV
1
1,
VcV
Nc
(µm3/cy.)
[ ]2
f
Ag fVc k V k e D
wear rate
1
,
*
,1
1
124
v
tr
w
trf
f
f
gPFf
fcc
D
D
hh
N
N
D
PK
ek
V
VN
0
5
10
15
20
25
30
35
40
0 10 20 30 40
Nc e
xp
(cyc
les)
Ncpred (cycles)
x104
x104
D=Dref=1mm5,000cycles Nf 60,000cyclesNf=Nf,ref=10,000cycles250µm D 1,500µmplain fretting
V
VcNc
total Ag volume involved in fretting wear process @ Nc
V : wear rate per fretting cycle
J. Laporte et al., Wear 376-377, (2017) 656–669.
k = 0.94 (proportion of volumetric Ag volume involved)
26/09/2017
21
41
6. Simplified strategy to compare coatings : ECR versus Coatings properties ?
Ncref : Reference ECR endurance defined for a reference sliding amplitude δ0=± 9µm
ECR endurance, Nc
δ*0 =±9µm
310 410 510 610 710 810 910 10100
5
10
15
20
25
30
1110
0 (±µm)
“power law function”
δ0=9µm
Ncref
the larger Ncref and _ref , the better the electrical performance !
cold welding index _ref=1/µmax (δ0=9µm)
0.0
0.5
1.0
1.5
2.0
2.5
0 5000 10000 15000
µmax
µ=Q*/P
fretting cycle
Definition of two driving parameters to describe GS ECR endurance & Cold welding
26/09/2017
22
Ni
AgSn
CuZn37 (substrat)
e=2µm
2 µm
Ni
AgCrN
CuZn37 (substrat)
e=2µm
2 µm
Ni
AgSnIn
CuZn37 (substrat)
e=2µm
2 µm
Ni
AgaC
CuZn37 (substrate
e=2µm
2 µm
New coating via Ag PVD
Possibility to explore new hardnesses and new conductivities !
Comparison of coatings
0
5
10
15
20
25
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1 x104
AuNi AgaC AgCrN AgSn AgSnIn Ag
ref=1/µmax Ncref (cycles)
considering ref & Ncref
the best compromise is obtained with AgSnIn (conductive ITO oxydes)
Conditions de test: P=3N δ0 = 9µm RH=10% f=30Hz T=25°C
e=2µm et eAuNi=1,3µm
44
Comparison of coatings
cold welding index
GS fretting endurance
index
Laporte et al. , IEEE 61st Holm Conference, 2015, 287-297
26/09/2017
23
Correlation of Nc_ref versus Hardness & conductivity
0
5
10
15
20
25
0 100 200 300 400 500
AgSnIn
conductive oxydes (ITO)
Hardness, H (hv)
Nc (x 10000)
Ag
AgaC
AuNi AgCrN
AgSn
High scattering : Hardness is not a relevent parameter
0
5
10
15
20
25
0.00 0.20 0.40 0.60 0.80
electrical conductivity , σ (µS.cm-1)
Nc (x 10000)
AgCrN
AgSn AuNi
AgSnIn
AgaC
conductive oxydes
Ag
Removing (AgSnIn) => Linear increase
)(NcNc VVCNc
CVNc c ≈ Cst (base Argent)
CVNc
NcV
45 Ncref is proportional to the coating conductivity
46
Tribological interpretation :
0
5
10
15
20
25
30
35
40
0 0.2 0.4 0.6 0.8
σ (µS.cm-1)
x104
AgSnIn
AgSn
AgCrN
Ag
AgaC
AuNi
0,NcV
NcVC
VNc (µm3)
the wear volume at the ECR failure is proportional to the coating conductivity
VNc ( x 104 µm3)
The higher the coating electrical conductivity , the larger the wear volume required to reach ECR failure → Nc is proportional to the coating conductivity
0
5
10
15
20
25
30
35
40
45
50
0 50000 100000 150000 200000 250000
KNc=1,82µm3/cycle
AgSnIn
AgSn AgCrN
Ag
AgaC
AuNi
Nc (cycles)
wear volume measured at Nc is proportional to the Nc
VNc ( x 104 µm3)
26/09/2017
24
47
Conclusions
- The partial slip/ gross slip displacement transition controls the transition from infinite to finite ECR endurance (Nc : N_fretting when Δ R>4 mΩ) - Nc is controlled by surface wear processes: ECR failure is reached when [O] > 45 at% (noble metal eliminated and replace by a oxide layer) - Nc can be expressed as a power law function of friction energy density ϕ (Wear depth is controlled by ϕ) - Nc can be expressed as a power low function of sliding amplitude (deduced from the general friction energy density formulation)
- Application of large sliding induced a noble metal “refilling” process of fretting scar (Increase of Nc) - The global response of a coating can described by two variables - Ncref (GS endurance index) & χ ref (cold welding index)
- Nc_ref is controlled by the electrical conductivity of the coating