no slide title · greece istram switzerland smr. comptest 2004: composites testing and model...
TRANSCRIPT
![Page 1: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/1.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
COMPTEST 2004
Experimental Measurement and Finite Element Analysis of Load Distribution and Strength in Multi-Bolt Composite
Joints with Variable Bolt Hole Clearances
M.A. McCarthy, V.P. Lawlor1, W.F. Stanley, G.S. Padhi, C.T. McCarthy2
University of LimerickIreland 1 Now National University of Ireland, Galway
2 Now University College Dublin
![Page 2: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/2.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Content
• Context for work
• Quasi-static loading
• Fatigue loading
• Finite Element Analysis
![Page 3: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/3.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Context – BOJCAS projectBOLTED JOINTS IN COMPOSITE
AIRCRAFT STRUCTURESEU FRAMEWORK V
COMPETITIVE AND SUSTAINABLE GROWTH(2000-2003)
Partners:
IRELAND University of Limerick (Project Co-ordinator)UNITED KINGDOM Airbus UK, QinetiQ (formerly DERA)SWEDEN SAAB, FOI, Royal Inst of Tech StockholmGERMANY Airbus GermanyITALY CIRATHE NETHERLANDS NLRGREECE ISTRAMSWITZERLAND SMR
![Page 4: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/4.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Why Study Clearance?
• Bolt-hole clearance results in 3D variations in stress/strain distributions
Good parameter to study for validation of 3D FE
• Clearance is inevitable in any practical manufacturing process – cannot be avoided, so effects should be understood
• Has not been studied experimentally in multi-bolt joints before
• Previous models of effects of clearance have been analytical or 2D FE
![Page 5: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/5.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Quasi-Static Loading
![Page 6: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/6.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Joint Geometries
• Single-lap joint
• HTA/6376 carbon/epoxy• Quasi-isotropic lay-ups• Titanium alloy bolts
• Double-lap joint
![Page 7: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/7.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Controlling Clearance
Four different size holes obtained with reamers specially manufactured to a tight (h6) tolerance
Drill Reamer
9µm tolerance on holeInside aerospace tolerances25µm tolerance on boltPossible 24µm variation for each nominal clearanceFour nominal clearances
Bolts (8 mm) obtained from SAAB (f7 tolerance)
![Page 8: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/8.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Clearance Cases
Nominal Clearance (µm)Case Code Hole 1 Hole 2 Hole 3C1_C1_C1 0 0 0 C1_C1_C2 0 0 80 C1_C1_C3 0 0 160 C1_C1_C4 0 0 240 C1_C3_C1 0 160 0 C1_C3_C3 0 160 160
![Page 9: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/9.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Centring/Aligning/Drilling Jigs
• Manufactured to very high precision
REF for jigs/fixtures: McCarthy, McCarthy, M.A., V.P. Lawlor, W.F. Stanley, 2004, An Experimental Study of Bolt-Hole Clearance Effects in Single-Lap, Multi-Bolt Composite Joints, Journal of Composite Materials, in-press.
![Page 10: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/10.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Measuring Load Distribution
Single-lap joints:
Instrumented bolts
Double-lap joints:
Strain gauges
![Page 11: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/11.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
SL Joints – Load Distr.Instrumented Bolts
0
1
2
3
4
5
6
0 0.1 0.2 0.3 0.4Corrected Stroke (mm)
Bol
t Loa
ds (k
N)
Bolt 1 (C1)Bolt 3 (C1)
Bolt 2 (C1)
C1_C1_C1
Instrumented Bolts
012345678
0 0.1 0.2 0.3 0.4 0.5Corrected Stroke (mm)
Bol
t Loa
ds (k
N)
Bolt 1 (C1)Bolt 2 (C1)
Bolt 3 (C4)C1_C1_C4
3D FEM
012345678
0 0.1 0.2 0.3 0.4 0.5Joint Displacement (mm)
Bol
t Loa
ds (k
N)
Bolt 1 (C1)
Bolt 2 (C1)
Bolt 3 (C4)C1_C1_C4
3D FEM
0
1
2
3
4
5
6
0 0.1 0.2 0.3 0.4Joint Displacement (mm)
Bol
t Loa
ds (k
N) Bolts 1 and 3 (C1)
Bolt 2 (C1)
C1_C1_C1
![Page 12: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/12.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
SL Joints - Failure
A
BA C
C BBolt Failure
No clear effect on ultimate strength
Net TensionFailure
![Page 13: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/13.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Most Interesting FailureC1_C1_C4 Specimen B
Two bolts failed simultaneously
Usual design rules (ignoring clearance) middle bolt NOT under any threat of failure
Instrumented Bolts
0
1
2
3
4
5
6
0 0.1 0.2 0.3 0.4Corrected Stroke (mm)
Bol
t Loa
ds (k
N)
Bolt 1 (C1)Bolt 3 (C1)
Bolt 2 (C1)
C1_C1_C1
Instrumented Bolts
012345678
0 0.1 0.2 0.3 0.4 0.5Corrected Stroke (mm)
Bol
t Loa
ds (k
N)
Bolt 1 (C1)Bolt 2 (C1)
Bolt 3 (C4)C1_C1_C4
But with clearance in one of the outer holes – failure of middle bolt becomes possible
![Page 14: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/14.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
SL Joints – Failure Initiation
0
5
10
15
20
25
30
0 10 20 30 40 50 60Load (kN)
Slop
e (k
N/m
m)
C1_C1_C1C1_C1_C4
C1_C1_C4 joint exhibits sharp losses in stiffness earlier than C1_C1_C1 joint
Slope of load-deflection curve, i.e. joint stiffness
![Page 15: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/15.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Measure of Failure Initiation
0
5
10
15
20
25
30
0 10 20 30 40 50 60Load (kN)
Slop
e (k
N/m
m)
C1_C1_C1C1_C1_C4
30% Loss
Load at 30% stiffness lossC1_C1_C1 case
REF: McCarthy, M.A., V.P. Lawlor, W.F. Stanley, C.T. McCarthy, 2002, Bolt-hole clearance effects and strength criteria in single-bolt, single-lap, composite bolted joints,Composites Science and Technology, Vol. 62, pp. 1415-1431.
![Page 16: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/16.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Failure “Initiation” Loads
Specimen Load at 30% stiffness drop (kN)C1_C1_C1 (A) 51.06 C1_C1_C1 (B) 47.26 C1_C1_C1 (C) 51.00
Average 49.77
C1_C1_C4 (A) 48.25 C1_C1_C4 (B) 47.64 C1_C1_C4 (C) 46.20
Average 47.36
Failure initiated earlier in C1_C1_C4 joints
![Page 17: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/17.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Double-Lap JointsAgain, no clear effect on ultimate strength
![Page 18: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/18.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
DL Joints – Failure Initiation
05
10152025303540
0 10 20 30 40 50 60 70 80Applied Load (kN)
Bol
t Loa
ds (k
N)
Bolt 2 (C1)
Bolt 3 (C1)
Bolt 1 (C1)
Significant failure event
C1_C1_C1
Load at 30% loss of stiffness matches load at interruption of strain gauge pattern very well
• Strain gauge method of load distribution measurement much cheaper can test to failure
• strain gauge readings interrupted at a “significant” failure event
![Page 19: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/19.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
DL Joint - C1_C3_C1
05
10152025303540
0 10 20 30 40 50 60 70 80Applied Load (kN)
Bol
t Loa
ds (k
N)
Significant failure event
Bolt 3 (C1)
Bolt 1 (C1)
Bolt 2 (C3)
C1_C3_C1
• Again load at 30% loss of stiffness matches load at interruption of straingauge pattern very well (true for all six clearance cases)• Load is significant lower in C1_C3_C1 case than C1_C1_C1• From consideration of bearing yield allowable, the “first significant failure” was found to be bearing failure at one of the holes
![Page 20: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/20.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Effect of Clearance on first bearing failure
Code Load at first bearing failure (kΝ)
Percentage Difference from
C1_C1_C1
C1_C1_C1 50 0% C1_C3_C1 44 12% C1_C1_C4 44.3 11.4% C2_C1_C1 43.2 13.6%
C4_C1_C1 40 20%
C3_C3_C1 37.2 25.6%
![Page 21: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/21.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Conclusions – QS Loading• Clearance:
No significant effect on ultimate tensile load
DID affect ultimate tensile mode
Small effect on failure initiation load in SL joints
LARGE effect on failure initiation load in DL joints (load at first bearing failure affected by 25%)
• Strain gauge load distribution method cheaper than instrumented bolts – can be used up to failure (cannot easily be used for SL joints though)
• Load at 30% loss in stiffness appears to be a good measure of first “substantial” failure
![Page 22: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/22.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Fatigue Loading
![Page 23: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/23.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Fatigue Cases
Nominal Clearance (µm)Case Code Hole 1 Hole 2 Hole 3C1_C1_C1 0 0 0 C1_C1_C4 0 0 240
• Both Single-Lap and Double-lap joint
![Page 24: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/24.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Test Set-up
• Constant amplitude fatigue loading, R = -1 (σmin/σmax = -1)
• Anti-buckling guides
Cooling system (compressed air)
• To avoid temperature rise, frequencies between 0.66 and 5 Hz
• Temperature of each bolt monitored – maintained < 25oC
![Page 25: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/25.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Failure Criterion
• Hole elongation criterion for failure (Starikov and Schon, 2002)
• Increase in peak-to-peak displacement ∆δ of 0.8 mm
![Page 26: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/26.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Ultimate Failure Modes• However, in single-lap joints, other failure modes occurred on continuation of tests beyond the hole elongation failure point
Bolt failure Net tension failure
Extreme hole elongation
• Double-lap joints exhibited only extreme hole elongation
![Page 27: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/27.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Ultimate Failure Cycles
• Cycles to “ultimate failure” were also recorded
• “Ultimate failure” – displacement to + or – 10 mm
• Reached suddenly in catastrophic failure modes and gradually in extreme hole elongation cases
![Page 28: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/28.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
SL Joints – Hole Elong
Joints with loose-fit bolt have shorter fatigue life in general
![Page 29: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/29.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
SL Joints – Ult Failure
Joints with loose-fit bolt have shorter fatigue life in general
![Page 30: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/30.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Temperature and Displ. History
Hole elongation initiates here
0
5
10
15
20
25
30
35
40
1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06No. of cycles
Tem
pera
ture
(deg
C)
012345678910
Del
ta_P
k_Pk
(mm
)
Temperature - Bolt 1Temperature - Bolt 2Temperature - Bolt 3Delta_Pk_Pk
C1_C1_C1 +/-27 kN
Single-lap joint
![Page 31: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/31.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Temperature and Displ. History
Hole elongation initiates earlier
0
5
10
15
20
25
30
35
40
1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05No. of cycles
Tem
pera
ture
(deg
C)
012345678910
Del
ta_P
k_Pk
(mm
)
Temperature - Bolt 1Temperature - Bolt 2Temperature - Bolt 3Delta_Pk_Pk
C1_C1_C4 +/-30 kN
Single-lap joint
![Page 32: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/32.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
SL - Cycles to small hole elong
• Clearer distinction between clearance cases
![Page 33: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/33.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
DL Joints – Hole Elong
Run-out
Joints with loose-fit bolt have shorter fatigue life in general
![Page 34: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/34.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
DL - Cycles to small hole elong
• Again - clearer distinction between clearance cases
![Page 35: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/35.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Load distribution during Fatigue
C4_C1_C1 200th Cycle C4_C1_C1 20000th Cycle
• Due to hole wear, clearance has less effect as wear progresses (load distribution evens out) clearance most affects initiation of failure, less effect on final failure
![Page 36: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/36.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Conclusions – Fatigue Loading
• Joints with a loose-fit bolt had shorter fatigue lives than joints with all neat-fit bolts (SL and DL)
• Clearance had a particularly strong effect on failure initiation, i.e. cycles to a small hole elongation
• Effect of clearance less pronounced as failure progresses since failure causes elongation of the neat-fit holes in the C1_C1_C4 joint causing the clearance to even out over time
![Page 37: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/37.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Finite element Analysis
![Page 38: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/38.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Model Creation Tool: BOLJAT
REF: Padhi, G.S., M.A. McCarthy, C.T. McCarthy, 2002, BOLJAT – A tool for designing composite bolted joints using three-dimensional finite element analysis, Composites, Part A, 33/11, pp. 1573-1584
![Page 39: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/39.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Contact AnalysisContact analysis performed between all parts
![Page 40: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/40.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Contact Area DevelopmentSingle-Bolt, Single-Lap Joint
C1 Clearance (Contact Area)
C4 Clearance (Contact Area)
![Page 41: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/41.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Model of Double-Lap Joint
Quarter Symmetry assumed
![Page 42: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/42.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Stress Distribution
Net tension stresses in central laminate highest at this hole (i.e. bypass stresses correctly accounted for)
Stresses vary through thickness of each laminate even in DL joint (due to bolt bending)
![Page 43: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/43.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Progressive Damage Analysis
![Page 44: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/44.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Bolt Loads (C1_C1_C1)
SimulationExperiment
![Page 45: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/45.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Bolt Loads (C4_C1_C1)
SimulationExperiment
![Page 46: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/46.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Compressive Matrix Damage (C1_C1_C1)
Hole 1(0 µm)
Hole 2(0 µm)
Hole 3(0 µm)Applied Load
10 kN
30 kN
50 kN
![Page 47: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/47.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Compressive Fibre Damage (C1_C1_C1)
Hole 1(0 µm)
Hole 2(0 µm)
Hole 3(0 µm)Applied Load
10 kN
30 kN
50 kN
![Page 48: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/48.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Compressive Matrix Damage (C3_C3_C1)
Hole 1(160 µm)
Hole 2(160 µm)
Hole 3(0 µm)Applied Load
10 kN
30 kN
50 kN
![Page 49: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/49.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Compressive Fibre Damage (C3_C3_C1)
Hole 1(160 µm)
Hole 2(160 µm)
Hole 3(0 µm)Applied Load
10 kN
30 kN
50 kN
![Page 50: No Slide Title · GREECE ISTRAM SWITZERLAND SMR. COMPTEST 2004: Composites Testing and Model Identification University of Bristol, UK, Sept 21st –23rd, 2004](https://reader031.vdocuments.site/reader031/viewer/2022020414/5bdde80709d3f2995d8bc87d/html5/thumbnails/50.jpg)
COMPTEST 2004: Composites Testing and Model IdentificationUniversity of Bristol, UK, Sept 21st – 23rd, 2004
Composites Research CentreUniversity of Limerick
Conclusions – FEA
• 3D FEA with contact can account for: variable contact in each hole and through thickness; bypass stresses; non-uniform bearing stresses through thickness
• PDA gives insight into failure processes at each hole and gives prediction of initial failure (bearing failure in one hole)