Many factors play a role in the overall outcome of PCB assembly, especially when very small components are involved. Broadly speaking, tombstoning is a problem with very small 2-pin components (0201 caps and such) and we've filtered out the following challenges:
- proper paste application
- component placement
- designing for "good" thermal properties of the PCB with regard to tombstoning
- moon phase
It seems that the effect of moon phase is least understood and researched, so I'm now designing a small PCB that is cheap, not useless (if the assembly process was successful), and - most importantly - contains some 0201 components.
There was a little pause because we had holidays and went on vacation so I missed the last full moon (20220318), but there are plenty of parts left.
Soldered Nr. 7 today - 20220320 - 18:27
One resistor (the 360R one) didn't sink into the solder on one side, resulting in an open connection. I'll generously count this as the first tombstone in this experiment:
Well it doesn't have to be a design that's already done. Designing a custom PCB for this project wouldn't be a problem. I'd then simply order a batch of 25 or so (that would already be enough for half a year)
True, like I said, just what popped into my head. The other thought I had was to pass the boards around to various people who can do the test at various times of the month. Could help get better statistics and possibly control for skill level.
Replying to your top level comment because I can't reply to the other one.
Passing the boards around might be fun, otoh if I just publish the design everyone can add their results to the pile.
I'm now leaning towards a PCB with no actual function that just has some 0201 footprints in two different styles (more and less prone to tombstoning) and with different thermal surroundings. I'd place a mix of caps and resistors, maybe 16 per PCB. That's plenty to get me dizzy for a week until the next is due.
Maybe you need a rubber chicken on your bench? :)