The k6-2 board works fine with an other PSU, but not with the enlight ?

Correct - I ripped out a Silencer 370 out of another computer and no problems at all - very strange. The K6-2 board must be drawing more off one component of the supply than the P3 board - only thing I can think of.

And to boot I have two of the enlights - it doesn't seem to like either one

well, thats pretty funny. I do not ever had this brand of psu.
It is possible as you say that the cubx has better filtering or tolerance.
Personally i would open the PSU and inspect it for leaking/bulging caps. That is what i always do when i get a used PSU, regardless if it works or not.

Are your two enlight psu the same model ?

I was thinking about opening the supplies myself. I am not sure if they are the same model or not.
I got them with some enlight cases I bought some time ago. I later upgraded the power supplies and kept those as spares.
I've searched the internet and quite a few people seem to say enlight supplies are, shall we say this nicely, low-end.

Could be a sign of bad caps and a lower performing PSU showing through?

Check the psu for bad caps. I have an enlight thats 11 years old an it works like a champ

Actually I was looking at the motherboard caps. They are Tayeh, which is apparently a bad brand. None of them appear to be bulging or leaking, but I'll bet they are just bad enough to make the board more susceptible to a borderline power supply.

Possibly... Im not an electronic expert but i change bad caps pretty often and i "fixed" many things this way. I usually rely on visual inspection but it is well know that caps can go bad without obvious exterior signs. Ive always wondered about how to approach this problem, as very often you cannot confirm a cap is bad without removing it from circuit... So far with computers i limit myself to physical evidences and it seems to be good enough. I would still check inside those PSU first thing if i were you.

Ive been repairing consoles too and it gets pretty interesting as in the console scene, problems are usually better known and more documented since every machine is the same and uses pretty much same layout and parts. Im thinking for instance about PC-engine Turbo Duo popping sounds that need changing some specific caps that do not necessarily look bad... PCE machines are particularly cursed about bad caps due to the time period and origin of parts it used... Some folks at PCENGINEFX stuck with issues actually made full recaps of their machine (30-ish caps) to once and for all get rid of their issues.

All this goes to show that, over time, caps aging can be a big source of malfunction and its not always easy to diagnose...

While your specific problem is actually pretty benign, you are obviously a curious person and your post touches an important issue for all of us collectors of old hardware. I support further talking on the subject.

MaxWar wrote:

Possibly... Im not an electronic expert but i change bad caps pretty often and i "fixed" many things this way. I usually rely on […]
Show full quote

Possibly... Im not an electronic expert but i change bad caps pretty often and i "fixed" many things this way. I usually rely on visual inspection but it is well know that caps can go bad without obvious exterior signs. Ive always wondered about how to approach this problem, as very often you cannot confirm a cap is bad without removing it from circuit... So far with computers i limit myself to physical evidences and it seems to be good enough. I would still check inside those PSU first thing if i were you.

Ive been repairing consoles too and it gets pretty interesting as in the console scene, problems are usually better known and more documented since every machine is the same and uses pretty much same layout and parts. Im thinking for instance about PC-engine Turbo Duo popping sounds that need changing some specific caps that do not necessarily look bad... PCE machines are particularly cursed about bad caps due to the time period and origin of parts it used... Some folks at PCENGINEFX stuck with issues actually made full recaps of their machine (30-ish caps) to once and for all get rid of their issues.

All this goes to show that, over time, caps aging can be a big source of malfunction and its not always easy to diagnose...

While your specific problem is actually pretty benign, you are obviously a curious person and your post touches an important issue for all of us collectors of old hardware. I support further talking on the subject.

If you happen to have an oscilloscope, here's a cheap and easy way to test a capacitor's ESR:
http://kripton2035.free.fr/analog%20esr/esr-scope.html

It's not very accurate but gives a good indication of whether the cap is fried or not. And because it only puts a very small voltage across the capacitor (< 0.05V), it can even be used to measure in-circuit. 😀