Comparison of different joining techniques in a ... · PDF file1 Comparison of different joining techniques in a crashworthiness perspective Massimiliano Avalle –Politecnico di Torino

1

Comparison of different joining

techniques in a crashworthiness

perspective

Massimiliano Avalle – Politecnico di Torino

Pierfranco Mauri – Henkel Loctite Italia

2

Contents

Comparison of joining techniques

Joining technique for car body construction

Mechanical testing

Results and discussion

Design of crash structures with different

joining techniques

Experimental results from crash boxes

Redesign of crash box with adhesive bonding

Comparison of the results

3

Why using alternative joining

solutions?

Adhesive bonding helps increasing stiffness

During polymerisation other fixing system is

needed

Adhesive bad compatibility with spot-welds

Adhesive and other mechanical fasteners can

join different materials

Repairing possible

4

Joints testing

Spot-weld strength

Static

Fatigue

Dynamic/Impact

Loading conditions

5

KS2 specimen

60°

30°

0°

90°

• Formerly used by Hahn

et al. for fatigue

• Material ZST340

• Samples from

for

6

Testing system

Double sliding

frame

Connections with

spherical joints

Inclined grips

(0°-90°)

piezoelectric loadpiezoelectric load--

cells cells (during

dynamic tests)

Details of the 30°

loading system

7

Testing apparatus

•• DropDrop--towertower

•• Height 12 m, Height 12 m,

•• Mass 60Mass 60--120 kg 120 kg

•• 300 kN max.300 kN max.

•• 13 m/s v13 m/s vmaxmax

•• Equipped for Equipped for

impact testing (in impact testing (in

compression)compression)

•• Equipped with Equipped with

tensile test griptensile test grip

•• Load measured Load measured

with piezoelectric with piezoelectric

loadload--cellscells

•• Stroke measured Stroke measured

with an optical with an optical

encoderencoder

•• Universal Universal

hydraulic hydraulic

material material

testing testing

equipment equipment

DARTEC DARTEC

HA100HA100

•• 100 kN max100 kN max

•• 100 mm/s v100 mm/s vmaxmax

•• Load Load

measured with measured with

a straina strain--gage gage

loadload--cellcell

•• Stroke Stroke

measured with measured with

LVDTLVDT

8

Spot-welds

Experimental results

Load & stroke measurement

Number of loading speed: 3

Low-speed: 0.01 mm/s

Medium-speed: 80 mm/s

High-speed (impact): 5.5×10³ mm/s

Load-curve characteristic

Failure surface is derived from maximum load

as a function of the loading angle

9

Spot-weld results

Low speed

0°

30°

60°

90°

• Apparatus: DARTEC HA100

• Loading speed: 0.01 mm/s

Stroke (mm)

Load (kN)

10

Spot-weld results

Medium speed

0°

30°

60°

90°


• Loading speed: 80 mm/s (=0.08 m/s)

Stroke (mm)

Load (kN)

11

Spot-weld results

High speed

0°

30°

60°

90°


• Loading speed: 5.5 m/s

Stroke (mm)

Load (kN)

12

Spot-welds

Low speed samples

13

Spot-welds

High speed samples

14

Analysis of the results

Joint strength is analysed as a function of the

loading angle:

An elliptic limit curve is assumed:

A different limit curve is obtained for each

loading speed

1

2

max

2

max

=

+

N

N

T

T

15

Spot-weld strength vs. loading

Limit curve as a function of the loading angle

( )2maxmax

2

max

tan1

1

cos NT

TF

αα +=

Loading angle (deg)

Maximum load (kN)

16

Spot-weld strength vs.

loading components

1

2

max

2

max

=

+

N

N

T

T

Maximum tensile load (kN)

Maximum shearing load (kN)

17

Failure surfaces

Static (0.01 mm/s) Dynamic (5 m/s)

Peel (90°) Shear (0°) Peel (90°) Shear (0°)

18

Alternative joining systems

for automotive constructions

Riveting

Self-riveting, punch riveting, Henrob joint

Clinching

Adhesive bonding

punch riveting Clinching

19

Clinching compared to

spot-welds (1/2)

Material: Mild Steel ( approx. 300 N/mm2)

1. Round die, ∅ 5mm 2. Round die, ∅ 8mm, 3a. Rectangular die, w. 4mm, shear 90 ° 3b.

Rectangular die, w. 4mm, shear 0 ° 4. Spot Weld, Standard spec. minimum, ∅3 & 4mm

Copyright© 2001 ATTEXOR Clinch Systems SA

20

Clinching compared to

spot-welds (2/2)

Mondino, I., Properzi, M., Giunti, T., Calderale, P.M., “La fatica di

giunzioni meccaniche per strutture veicolistiche innovative”

(Fatigue of mechanics joints for innovative car body structures)

Proceedins XXVIII AIAS Conf., 1999

21

KS2 specimen

60°30°0° 90°• Formerly used by Hahn

et al. for fatigue

• Material DC04

• Samples from

22

Clinching and Bonding

Experimental plan

KS2 specimen loaded at different angles

Clinched 21 samples

Bonded LOCTITE Hysol® 9466 21 samples

Clinched+bonded 22 samples

Loading speed: quasi-static 0.01 mm/s

Bonding procedure

Sanding (paper sand P80)

Degreasing (LOCTITE® 7063) and bonding

Polymerisation: 90 minutes @ 100°C

23

Clinched and bonded

KS2 specimen

Clinched+bonded

Clinched

24

Clinched & clinched+bonded

samples

Clinched+bonded,

90°Clinched+bonded,

0°

Clinched+bonded,

60°

Clinched+bonded,

30°

Clinched, 30° Clinched, 90°Clinched, 60°Clinched, 0°

25

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 0.5 1 1.5 2 2.5 3

Stroke (mm)

Load (kN)

Clinching

Test results

0°

30°

60°

90°



26

0

2

4

6

8

10

12

0 2 4 6 8 10

Stroke (mm)

Load (kN)

Bonding

Test results



27

0

2

4

6

8

10

12

0 1 2 3 4 5 6

Stroke (mm)

Load (kN)

Clinching+bonding

Test results



0°

30°

60°

90°

28

0

2

4

6

8

10

12

0 15 30 45 60 75 90

Loading angle (deg)

Maximum load (kN)

Clinched

Bonded

Clinched+bonded

Clinched and bonded joint

strength vs. loading

Limit curve vs. Loading angle

( )2maxmax

2

max

tan1

1

cos NT

TF

αα +=

29


strength vs. load components

0

2

4

6

8

10

12

0 2 4 6 8 10 12

Normal load (kN)

Shear load (kN)

Clinched

Bonded

Clinched+bonded

1

2

max

2

max

=

+

N

N

T

T

30


energy vs. loading angle

Failure energy vs. Loading angle

0

10

20

30

40

50

60

0 15 30 45 60 75 90

Loading angle (deg)

Failure energy (J)

Clinched

Bonded

Clinched+bonded

31

Different joining solutions:

conclusions

Joining by clinching is effective and a good

alternative to spot-weld

The use of adhesive strongly increase

strength and energy absorption capability

Clinching can be use to make bonding

operations easier: the pieces are kept in

place up to complete polymerisation

Clinching in addiction to bonding offers

additional safety as an extreme protection in

the case of adhesive premature failure

32

Behaviour of crash boxes with

alternative joining solutsions

Is it possible to substitute spot-welds with

other joining systems directly in the common

constructive solutions?

Crash behaviour can be improved?

33

• Cleaning and surface preparation with sand paper• Degreasing with Loctite 7063• Activation with Loctite 7388• Application of Loctite 330 adhesive• NDT ultrasonic inspection

Bonded crash-box production

60 60

• Cleaning and surface preparation with sand paper• Chemical degrease• Mixing of components and application of CIBA araldite adhesive• NDT ultrasonic inspection

34

NDT procedure by Rossetto & Goglio(ref. XXIX AIAS et al.)

Ultrasonic NDT inspection

35

Stress analysis of the bonded

flanges15

30

30

60 15

sp.1

F

F F

F

0

40

80

120

160

200

-10 -8 -6 -4 -2 0

coordinate y [mm]

Von Mises equivalent stress [MPa]

adhesive layer thickness 0.2 mm

adhesive layer thickness 0.4 mm

Position (mm)

Equivalent stress (MPa)

36

0

5

10

15

20

25

-10 -8 -6 -4 -2 0 2 4 6

coordinate y [mm]

Von Mises equivalent stress [MPa]

flat fillet filling

round fillet filling

Improvements by filleting

Position (mm)

Equivalent stress (MPa)

37

Quasi-static crushing

Spot-welded crash-box

0

5

10

15

20

25

30

0 10 20 30 40 50 60 70 80 90 100

spostamento [mm]

Forza [kN]

saldatura a punti - trave N4016003



0

200

400

600

800

1000

1200

0 10 20 30 40 50 60 70 80 90 100

spostamento [mm]

Energia [J]




Spot-welded - test N4016003



Energia(J)

Stroke (mm)

Load (kN)

Stroke (mm)




38

0

5

10

15

20

25

30

35

40

45

0 10 20 30 40 50 60 70 80 90 100

spostamento [mm]

Forza [kN]

adesivo LOCTITE 330 - trave B1



0

200

400

600

800

1000

1200

0 10 20 30 40 50 60 70 80 90 100

spostamento [mm]

Energia [J]





Adhesively bonded crash-box

Bonded LOCTITE 330 - test B1



Energia(J)

Stroke (mm)

Load (kN)

Stroke (mm)




39


Comparisons (1/2)

0

200

400

600

800

1000

1200

0 10 20 30 40 50 60 70 80 90 100

spostamento [mm]

Energia [J]



adesivo CIBA araldite 2010 - trave B4

Stroke (mm)

Energy (J)

0

5

10

15

20

25

30

0 10 20 30 40 50 60 70 80 90 100

spostamento [mm]

Forza [kN]



adesivo CIBA araldite 2010 - trave B4

Stroke (mm)

Load (kN)


Bonded - LOCTITE 330 – test B1

Bonded - CIBA Araldite 2010 - test B4


Bonded - LOCTITE 330 – test B1

Bonded - CIBA Araldite 2010 - test B4

0

5

10

15

20

25

30

0 10 20 30 40 50 60 70 80 90 100

displacement [mm]

Force [kN]

spot weld - piece N4016003



adhesive LOCTITE 330 - piece B1



0

200

400

600

800

1000

1200

0 10 20 30 40 50 60 70 80 90 100

displacement [mm]

Energy [J]







Stroke (mm)

Energy (J)

Stroke (mm)

Load (kN)

40

Mean load (kN) Max load (kN)

B1 11.62 28.57

B3 10.80 27.62

B5 11.79 41.75

mean value 11.40 32.65

stand. dev. 0.53 7.90

LOCTITE 330 adhesiveLOCTITE 330LOCTITE 330 adhesiveadhesive

Spot-weldSpotSpot--weldweldMean load (kN) Max load (kN)

N1 8.09 20.92

N2 7.92 21.92

N3 7.57 21.41

mean value 7.86 21.42

stand. dev. 0.26 0.50


Comparisons (2/2)

41

•Made with an ΩΩΩΩ elements bonded to a flat plate (top-

hat section)

15

60

60 15

sp.1

Alternative solutions for

improved bonded box (1/2)

15

70

60 15

sp.110

• New shape with two ΩΩΩΩelements one (smaller)

inserted into the other

AA BB

42

• Bonding flanges as simple

substitution of spot-welds

•Made with two ΩΩΩΩ elements (double hat section)

15

30

30

60 15

sp.1

60

22

22

sp. 1

16

• Double U section

• Can be obtained by joining

two identical U elements

Single-lapjoint

Alternative solutions for

improved bonded box (2/2)

CC DD

43

Components characteristics

Crash-boxes length 300 mm

Sheet thickness 0.8 mm

Material DC04 (ex FeP04)

LOCTITE 330 adhesive

procedure:

cleaning and sanding (sand-paper P100)

Application of “cleaner” 7063 and activator

7388

Polymerisation for at least 3 days

44

Quasi-static tests

Load and energy curves

C shape: 2 folds followed by global instability and debonding

other sections: regular folding some debonding (except D shape)

0

10000

20000

30000

40000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

750

1500

2250

3000

Energia (J)

Vel. 15 mm/min

A

0

10000

20000

30000

40000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

750

1500

2250

3000

Energia (J)

Vel. 15 mm/min

B

0

10000

20000

30000

40000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

750

1500

2250

3000

Energia (J)

Vel. 15 mm/min

C

0

10000

20000

30000

40000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

750

1500

2250

3000

Energia (J)

Vel. 15 mm/min

D

Energy (J)

Stroke (mm)

Load (N)

Energy (J)

Stroke (mm)

Load (N)

Energy (J)

Stroke (mm)

Load (N)

Energy (J)

Stroke (mm)

Load (N)

45

Quasi-static tests

C shape: 2 folds followed by global instability and

debonding

other sections: regular folding some debonding

(except D shape)

46

Medium-speed tests


0

10000

20000

30000

40000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

750

1500

2250

3000

Energia (J)

Vel. 480 mm/min

0

10000

20000

30000

40000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

750

1500

2250

3000

Energia (J)

Vel. 480 mm/min

0

10000

20000

30000

40000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

750

1500

2250

3000

Energia (J)

Vel. 480 mm/min

0

10000

20000

30000

40000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

750

1500

2250

3000

Energia (J)

Vel. 480 mm/min

A B

C D

C shape: 2 folds followed by global instability and debonding

other sections: regular folding some debonding (except D shape)

Energy (J)

Stroke (mm)

Load (N)

Energy (J)

Stroke (mm)

Load (N)

Energy (J)

Stroke (mm)

Load (N)

Energy (J)

Stroke (mm)

Load (N)

47

Medium-speed tests

C shape: 2 folds followed by global instability and

debonding

other sections: regular folding some debonding

(except D shape)

48

0

10000

20000

30000

40000

50000

60000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

200

400

600

800

1000

1200

Energia (J)

Vel. 6.5 m/s

C

Impact tests (6 m/s)


0

10000

20000

30000

40000

50000

60000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

200

400

600

800

1000

1200

Energia (J)

Vel. 6.5 m/s

0

10000

20000

30000

40000

50000

60000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

200

400

600

800

1000

1200

Energia (J)

Vel. 6.5 m/s

0

10000

20000

30000

40000

50000

60000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

200

400

600

800

1000

1200

Energia (J)

Vel. 6.5 m/s

A B

D

A, B, C shapes: irregular debonding in the crushed part

D shape: no debonding

Energy (J)

Stroke (mm)

Load (N)

Energy (J)

Stroke (mm)

Load (N)

Energy (J)

Stroke (mm)

Load (N)

Energy (J)

Stroke (mm)

Load (N)

49



D shape: no debonding

50



0

10000

20000

30000

40000

50000

60000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

400

800

1200

1600

2000

2400

Energia (J)

Vel. 9 m/s

0

10000

20000

30000

40000

50000

60000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

400

800

1200

1600

2000

2400

Energia (J)

Vel. 9 m/s

0

10000

20000

30000

40000

50000

60000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

400

800

1200

1600

2000

2400

Energia (J)

Vel. 9 m/s

picco 62.85 kN

0

10000

20000

30000

40000

50000

60000

0 50 100 150 200

Schiacciamento (mm)

Carico (N)

0

400

800

1200

1600

2000

2400

Energia (J)

Vel. 9 m/s

A B

C D


D shape: complete regular folding, no debonding at all

Energy (J)

Stroke (mm)

Load (N)

Energy (J)

Stroke (mm)

Load (N)

Energy (J)

Stroke (mm)

Load (N)

Energy (J)

Stroke (mm)

Load (N)

51



D shape: complete regular folding, no debonding at all

52

0

10000

20000

30000

40000

50000

60000

Sez. A Sez. B Sez. C Sez. D

Carico (N) 15 mm/min

480 mm/min

6.5 m/s

9 m/s

Valore massimo

A, B shapes: maximum load in the average (≈30 kN), good energy absorption characteristics (2÷2.7 kJ)

C shape: high maximum load (35 kN), low energy absorption

(1.6-0.7 kJ), complete debonding and global instability

D shape: lowest maximum load (24-28 kN), good energy

absorption (1.6-1.8 kJ), regular folding

Quasi-static and medium-speed tests

Comparison of the different

shapes: low speedMaximum load(kN)

53

0

10000

20000

30000

40000

50000

60000

Sez. A Sez. B Sez. C Sez. D

Carico (N) 15 mm/min

480 mm/min

6.5 m/s

9 m/s

Valore massimo

C shape: high maximum load (53-63 kN), low compression (50-

104 mm);

B shape: low maximum load (39 kN) and crushing (57-115 mm);

A shape: average maximum load (38-51 kN), high crushing

(111-184 mm)

D shape: quite low maximum load (39-44 kN), high crushing

(83-181 mm), regular folding

Comparison of the different

shapes: high speedMaximum load(kN)

54

Conclusions

Similar results both from low and medium speed and

impact tests

Adhesive bonding is a good solution also for energy

absorption during crash

Sensitive improvements by means of suitable (but

simple) geometrical modification of more common

shapes used for spot-welded structures:

- C shape: bonded sections normal to sides → bad

design

- D shape: bonded sections parallel to sides →optimal design

Documents

Comparison of different joining techniques in a ... · PDF file1 Comparison of different joining techniques in a crashworthiness perspective Massimiliano Avalle –Politecnico di Torino