Interesting 400W CompUSA PSU - Wintech OEM

[Pentium4]

Gonna try and spice up the PSU section a little. Haven't been posting much this whole year

So I got this super unique and interesting PSU in the scrap bin at the local shop!

I have never seen anything like this thing before...so I'm trying to figure out. I know CompUSA had used Wintech but all I've ever seen is those stupid Leadman ones! This one looks honestly rated. The thing is pretty well built.

First of all, what's going on with this thing? Basically two identical boards stacked??

It seems that the top one is for 12V only, but it also seems to have a 5VSB circuit on it too...I've seen two board PSUs before but this one literally seems like two PSU! Both boards have full input filter with 4 Y caps, 1 X cap, 2 coils, NTC, 2 MOV, fuse, BR, two 200V caps, two 13009...

Has a total of 4 30A schottky rectifiers. Only one is 60 reverse voltage so that one must be for the 12V.

I haven't plugged it in to test yet, but it was in the "tested good" section. Not that I fully trust it Is it bad that one of the heatsinks are touching one of the long screws?

The Yate Loon fan is failed and the Globe Fan spins fine but was very dry and so I added some oil to it. Let me know what you guys think, I think I'm going to fix it up!

[stj]

Re: Interesting 400W CompUSA PSU - Wintech OEM

does it actually have any airflow?

[goodpsusearch]

Re: Interesting 400W CompUSA PSU - Wintech OEM

Interesting! I hope the heatsinks are not electrically conductive because they are used from both switching transistors and secondary rectifiers.

Seems well built. Would definitely try to fix it.

[ReeceyBurger123]

Re: Interesting 400W CompUSA PSU - Wintech OEM

Wow that's a strange design there, seems decent though. Deffo worth recapping and fixing up

[Pentium4]

Re: Interesting 400W CompUSA PSU - Wintech OEM

does it actually have any airflow?

Guessing not much, since it seems to have a fan controller! Since the boards seem to be in parallel, is this half bridge or full bridge topology?

Interesting! I hope the heatsinks are not electrically conductive because they are used from both switching transistors and secondary rectifiers.

Seems well built. Would definitely try to fix it.

It looks like the only way it can go back together is if the top heatsink is touching the screw. I guess I'll just wedge a piece of plastic between just to be safe. Yeah, it does seem like a nice PSU I doubt there are many CompUSA PSU still in operation. They've been defunct for almost 9 years now!

Wow that's a strange design there, seems decent though. Deffo worth recapping and fixing up

It's neat, isn't it?



Excited to see what momaka, Wester, and 370 have to say

[ReeceyBurger123]

Re: Interesting 400W CompUSA PSU - Wintech OEM

Haha, just get rid of the fuckyou caps and will be good for use

[Wester547]

Re: Interesting 400W CompUSA PSU - Wintech OEM

Well, since my name was mentioned...

It looks like the supply in question was produced in late 2001 in accordance to the datecodes on the Fairchild bipolar switchers (November-December 2001). I've never actually seen a design like this in such an old supply before either. It looks like two power supply PCBs (or two power supplies) were actually used in one case so as to achieve the 400W rating on the label. What also strikes me is the fact that +3.3V is not independently regulated in this unit - there is no extra toroidal coil to derive the +3.3V transformer pins from the +5V pins, and no toroidal coil for +3.3V either. That can't be good for voltage regulation and crossloading unless +3.3V is linear regulated (I'd hope not with those heatsinks). I also can't seem to spot a film coupling capacitor on the primary side - not good for the primary bipolar transistors. Disappointing to know the Yate Loon fan seized, but oh well, can't win them all. It should be fine if you lube it and maintenance it (IE restore the shaft to proper operation). I recommend getting rid of that conductive snot as well.

[Behemot]

Re: Interesting 400W CompUSA PSU - Wintech OEM

Suggest using combination of vaseline and oil, only using oil is not reliable enough from my experiency.

If both the units are independently regulated, than it may actually be good at least in terms of +12V vs. +3.3/+5V crossload. That means it may be possible to use it with newer, +12V heavy systems with little load on other rails.

[momaka]

Re: Interesting 400W CompUSA PSU - Wintech OEM

Excited to see what momaka, Wester, and 370 have to say

It's a piece of turd!
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Nah, I'm totally joking

Actually, this is probably one of the coolest PSU's I've seen in a long while (if not the coolest ).

First of all, what's going on with this thing? Basically two identical boards stacked??

It seems that the top one is for 12V only, but it also seems to have a 5VSB circuit on it too...I've seen two board PSUs before but this one literally seems like two PSU!

I think Behemot beat me to it, but...
Yes, one of the boards is for the 12V rail and -12V rail only. The other board is for 3.3V, 5V, and -5V rails.
The only rail I can't tell you where it is originating from is 5VSB. It's funny that both boards have a 5VSB circuit (though a 5VSB transformer is necessary, because it also generates a voltage for the PWM IC on each board).

Speaking of the 5VSB... I see 2-transistor design. yet. again. *sigh*
So you know the drill with the caps here - don't power it ON until 5VSB caps have been replaced.

Is it bad that one of the heatsinks are touching one of the long screws?

*Probably* not.
Since both the primary and secondary share a heatsink on each PSU board, my guess would be that the heatsink is grounded for safety. That way, if some component on the primary near or on the heatsink arcs-over to it, the arc will get shunted to ground.

What also strikes me is the fact that +3.3V is not independently regulated in this unit - there is no extra toroidal coil to derive the +3.3V transformer pins from the +5V pins, and no toroidal coil for +3.3V either. That can't be good for voltage regulation and crossloading unless +3.3V is linear regulated (I'd hope not with those heatsinks).

Yes, looks like the 3.3V rail is derived the same way as the 5V rail - i.e. each has its own rectifier and tap from the main transformer (so no mag-amp or linear regulator).

However, because this PSU splits the 12V rail away from the 3.3V and 5V rails, then the regulation on the 3.3V rail very likely won't suffer. In fact, as Behemot suggested, this PSU might do quite well in terms of cross-loading.

With a heavy 12V load, only the PSU board with the 12V rail will be loaded down. Because the controller on that board doesn't have a 3.3V and 5V rails to worry about, it can always regulate the 12V rail exactly according to the load.

Similarly, if this PSU is used in a 5V-heavy PC, then the 5V rail will be loaded the most. Thus, only the PSU board with the 3.3V and 5V rails will need to work harder. Now you may wonder if this will throw-off the 3.3V rail. But because the wire for the 5V rail is usually a continuation of the wire for the 3.3V rail inside the main transformer, then a heavy load that lowers the 5V rail will also lower the 3.3V rail proportionately. So if they both go down, the controller can step-up as much as needed, because it doesn't have to worry about a 12V rail going high from not being loaded.

So this power supply has an independently regulated design, as strange as it looks .

I also can't seem to spot a film coupling capacitor on the primary side - not good for the primary bipolar transistors.

I think it is behind one of the primary caps (you can see one of the primary caps being a little more forward than the other in this picture
Without it, this PSU would have committed a kaboom

Let me know what you guys think, I think I'm going to fix it up!

Yes, please do. Quirky PSUs like this are rare, especially nowadays with all of the boring SMD-everything.
Wintech for the WIN