How many of the Bitmain APW3's have you tried out? Through my testing I have 1 remote connected to a APW3 and one now at my place connected to an APW3 and not having issues with them.
What way are you testing the APW3 to determine they are coming up short? I have been testing by running the units with them so wanted to what tests you were using to get this. Just trying to compare as we seem to be getting different results.
I have personally purchased and have been using 30 of the Bitmain PSUs with no issues so far. The rest of my miners have been on rental PSUs. I would say that over the time of 9 months is a very good track record.
the s-7 uses 50 to 80 watts less then the s-9.
many symptoms described on the s-9 were lessened by down clocking freq just a bit. all anecdotal info.
all your 30 plus are on s-7 correct? how many are on the s-9.
mind you if you have climate control for you psu's (i.e. ambient under 80f) you will get better results.
a psu right at its real world limit will be aid by a cooler temp.
And as I said i am basing all info on what I read here. as I do not have a single antminer psu.
this graph clearly shows high ambient temps will hurt the antimatter psu.
a 5c temp shift drops amps from 133 to 117.
and if you really look at the temp graph the temp scaling is deceptive.
look at 20 c look at 50 c then look at 55 c it is not scaled correctly.
to the point that it is done to deceive your eyes.
(graph above)
I have also had good luck with the APW3's by bitmain. I am testing by using them by running S9's and having no issues there. One is a older one that ran a S7 for quite a while and now is powering a S9 and again not any issues.
I would hold off on saying it comes up short personally. I just have not experienced this and I realize I'm one person but at this point I have used 2 separate S9's and 2 APW3's. So I would get some test's with hand's on them before deciding that the APW3's are coming up short. And I'm just one person so I cannot speak for everyone but they seem to work for me fine.