Greetings fellow researchers!
I am experimenting with thick plate (9.5 mm) AA5083-H321 (non-heat treatable) and AA6061-T6511 (heat treatable) using the same FSW tool (tapered, threaded, three-flats). Several trials have confirmed that the process window seem to be very different, i.e. AA5083 has very narrow process window that can yield defect-free samples, while AA6061 has a large one. Even using a 0.056 mm/rev revolutionary pitch (slow, hot weld) in AA5083 FSW sometimes results in intermittent voids at the boundary between shoulder-driven flow and pin-driven flow. AA6061 can go with a much higher rev. pitch without a problem.
My first guess for the difference was flow stress, but I have looked at the flow stress from available data (Kaufman's handbook) of both alloys at elevated temperatures and it seems rather identical.
Any ideas/suggestions are greatly appreciated.
For your information, the tool shoulder and pin diameter is 15 mm and 6 mm respectively. A 2.5deg tilt angle was used.
Thank you in advance!
You may try increasing the plunge force. The width of the process window is determined by the plunging force value so that wider welding speed range may be used at higher force values. Holes arise due to insufficient flow of plasticized metal into a zone behind the tool. 5083 may have strong adhesion to the tool and therefore bad stirring and flow capacity.
We use 0.5 m/min welding speed at 550 rpm and 2600 kg force for welding 5083 5 mm thickness butt joints.
Thank you for the feedback Dr Sergei. I haven't considered initial plunge force at the moment, but will try to take that into account too.
The adhesion to the pin you mentioned may be the issue since there maybe be difference in deposition volume due to shoulder vs pin, hence the voids mainly formed at the shoulder/pin-driven boundary. Interesting insight!
Thank you again.
I think you should use wider shoulder (approx 20mm+) and tilt angle should be reduced to 2 deg max. How long is your pin?
Hello Dr Azman. Thank you for the suggestions. Yes, enlarging the shoulder is one of our idea to solve the problem. Our tool pin is 9.3mm (0.2mm less than plate thickness). However, since we were thinking of making a comparison between 5083 and 6061, enlarging the shoulder means more heat generation mainly at the top, which may widen the temperature gradient from the top to the bottom of the thick plate. In 5083 this may not have an effect, but 6061 will be affected.
We haven't considered changing the tilt angle and would be a valid parameter to check. Thanks!
- Badges
- Science topic
- Similar topics
- Biological Science
- Physiology
- Physiological Processes
- Stress