My question is do you think the problem with chain 2 is the capacitor that the 63rd chip? Would you do something different that re-soldering if you faced these errors?
I assume you're talking about one of the large electrolytic caps? I doubt it would cause the issue you are seeing. If it was one of the caps in the boost circuit that generates the 1.8V signal for the last voltage domain, maybe the regulator could go unstable, so that would mess up all the chips in the last voltage domain though.
Also, just re-soldering those chips is a bit of a shot in the dark. The problem could be those chips, or the chips after, or sometimes even just another chip in the same voltage domain.
Thanks wndsnb for responding. You are right. it is the large 330mf and 30v cap just besides, what I think is, the 63rd chip. Incidentally this board is finding only 62 chip. I encircled the cap and the 63rd chip in black oval in the picture that I borrowed from Zeus manual.
https://imgur.com/a/RaQXtuSAnother strange thing is after another reboot the error message changed a bit
"Chain 2 only find 0 asic, will power off hash board 2"
compared to last time
"Chain 2 only find 62 asic, will power off hash board 2"
So why did it find 62 Asic last time but 0 this time? Outside temperature is sub zero after a round of snow and I don't have heating in the place where the miner is so excess temperature is ruled out. This is the board where I reinstated the capacitor. i agree with you that the problem may be in chip # 62, 61, 63 or in the entire last column 61 through 65 which is the voltage domain if I am not wrong. I agree that just re-soldering is a shot in the dark but I lack expertise and probably equipment to do anything more.
Chain 0 is now saying
"chain 0 get hashrate_reg_counter 63, require 65, failed times 1: ooooo ooooo ooooo ooooo ooooo ooooo ooooo ooooo ooooo ooooo ooooo ooooo xxooo"
compared to last time
chain 0 get hashrate_reg_counter 58, require 65, failed times 1: ooooo ooooo ooooo ooooo ooooo ooooo ooooo ooooo ooooo ooooo ooooo oooxx xxxxx"
This board seem to have done slightly better. If I am reading it right last time chips #59-65 failed but this time only chip#61 and 62 failed. Am I reading it right?