PSI - Issue 23

C. Skotarek et al. / Procedia Structural Integrity 23 (2019) 463–468 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

466

4

attached to the plating of the hole via the cold-welded zone. Crack initiation becomes very likely due to the porous nature of the cold-welded zone. This scenario can be checked by performing an interrupted tensile test. Fig. 5a shows the corresponding displacement curve. The micro-tensile device was stopped at a macro-displacement level of 0.033mm (s. Fig. 5a) and the hole was filled with glue (Delo CA 2905). Then the pin and the hole were cut using Leica TXP target preparation. Fig. 5b shows that the cold-welded zone has cracked from both sides, and that the crack is barely opened. This implies that the electrical performance of the connector is anticipated to be still quite good at this stage of the damage process [Hannel et al.], but that gradual degradation will occur due to oxidation processes taking place in cracked area.

0.010

0.005

0.000

0

0.2

0.4

0.6

0.8

1

-0.005

-0.010

pin tip opening [mm]

-0.015

strain [%]

a) pin tip opening for interrupted test

b) hair-line crack in cold-welded zone

Fig. 5 Damage in cold-welded zone at the onset of pin opening

3. Simulation model A simulation model was set up with the purpose of simulating the contact stresses during the insertion process and of estimating the amount of damage in the contact zone at a given load level. The pin was scanned in a high resolution computer tomograph (CT; Zeiss XRadia Versa 520). Both the exact physical dimensions of each pin and the surface topography were determined and converted to a stl-file, which then, in turn, was imported as an input file to the FE code ABAQUS. The insertion process was simulated in a dynamic simulation with an insertion rate of 50 mm/min. Fig. 6 shows typical insertion diagrams and the resulting contact stress in the area where the cold-welded zone will develop. The comparatively high forces necessary in the initial part of the insertion process are related to the deformation of the pin and the hole, where the plateau value corresponds to stationary sliding conditions. The average value of the contact stress amounts to about -100 MPa with the exact value depending on the detailed physical dimensions. The cold-welded zone was approximated by a cohesive contact approach [ABAQUS] upon completion of the insertion process. In this approach, there is no specific element layer attributed to the contact zone, but all elements of the pin and the bore hole touching each other take part in the adhesive zone. Hence, the cold-welded zone has zero thickness in the simulation model in contrast to reality, but its shape is directly determined by the physical dimensions of the bore-hole and the pin in agreement with a real press-fit system. Fig. 7 illustrates how the damage process is incorporated in the cohesive zone model. There are two damage variables, one characterizing the onset of the damage process (“damage initiation”), and one defining the amount of decohesion (“damage evolution”). If the equivalent stress exceeds a certain critical value, then the damage initiation variable attains a value of one, and the damage evolution process starts in terms of surface separation. If the displacement of the nodes of two elements in the contact lies above a pre-defined value, then the damage evolution process is completed and the corresponding damage parameter reaches its maximum value of 1, i.e. this part of crack cohesive zone is supposed to be cracked. Parameter identification of a cohesive zone model is a problem which has recently attracted the attention of many researchers (e.g. [Skec], [Lelias et al.], [Albuquerque et al.], [Xu et al.]). In the study presented here, the parameters were determined via the pin-tip opening curve as shown in Fig. 5a. Fig. 8 indicates that the mechanical behavior of the system can be reproduced quite well with the model described above.