Re: ATX psu help
Well, i think that connecting two diodes in parallel (I mean, connecting two diodes rated at 0.5A, to get a new one rated at 1A) is a bad idea, too. (Damn, i hate to be the "guy of the bad news")
The problem is that, a little diference in the parameter "Is", can produce a big diference in the resulting Id when the diodes are in parallel and at the same voltaje (Remember that Id=Is*(e^(Vd/n*Vt)-1), where Is ≈ 10^-15). One of the diodes can get, more current than the others, just because they are a little different.
You can try it with two 1N4148, one 5V source, one resistor and a digital multimeter.
I don't think this problem will have an easy and cheap solution ... if you're working with little currents, maybe you can try (I mean the parallel diodes, not the paralleled PSU), but i think that 100A is a lot.
Can't you break the system in some separated parts?
Maybe if you explain what are you trying to do, someone can help you.
Regards, Diego.
Added after 26 minutes:
After some time searching in internet trying to help, i found this: **broken link removed**.
I think you need a mix between Figure 32, and Figure 33, but i'm not sure this will work for your particular case. Maybe you can try with a little load to see how it work, and slowly increment it, to see what happens. Please be very carefull!
Regards, Diego.