Apevia WIN-500XSPX recap

[Wrog]

So I had some time this weekend to start putting together another HTPC for the kid's playroom, and thought to myself.. "Hey! I'll recap the power supply in advance so we won't have to touch this for a loooong time". It's an Apevia WIN-500XSPX that came in a Apevia X-MASTER-BK/500 HTPC case. So I went through and remove all 21, thinking that it'll be a quick one. As I go to start putting the new ones in (combination of FC/FM series, and a few BCN custom Samxon GD/RS), I run across two of the small ones that don't have their polarities marked on the PCB. No problem, I think to myself -- I have another one of these boxes out in the living room that I can reference... I open up the power supply (same HTPC case, same power supply model) and the insides are completely different! Help! I can make an educated guess, but would rather be 100% sure and avoid any fireworks.

I've attached some pics of the unit and marked up one that have caps near the SMPS controller (a KA7500C, datasheet attached). C212 in the top left seems to be fairly obvious. The top lead connects with the top lead on the cap to its left (C213), which is connected to the screwhole (ground). I'm assuming that the negative lead of C212 should connect to ground (top hole)? For the cap south of the SMPS controller (C208), the top lead connects to pins 13/14/16 of the controller (13/14 are tied together) and the bottom lead connects to pin 4 of the controller. It seems to me that the positive lead goes towards the controller -- is this correct?

What do you guys think of the unit? Safe to do a Q9400/4GB with a 2TB drive and quad tuner for an extended period of time?

[c_hegge]

Re: Apevia WIN-500XSPX recap

jumpered input filtering! you need to fix that. For C213, the outer leg is almost definately negative. I don't know about the other one. Having said that, though, I would not recommend that PSU for a quad core. It looks like it might do 300W or so, but that might not be enough for your system.

[momaka]

Re: Apevia WIN-500XSPX recap

For C213, the outer leg is almost definately negative.

Sounds right. Just follow the wire labeled "RC" and see which pin it goes to on the ATX connector. If it connects to the PS-ON (pin 16 on a 24-pin ATX connector, or pin 14 on a 20-pin ATX connector) then the cap's negative leg should definitely go to ground/case.

Not sure about the other capacitor (C208) either.
----
However, let me just say that this PSU is nowhere near good for a Q9400. It's actually quite crappy.
Not only is the input filtering totally lacking like c_hegge mentioned, but look at the size of the rectifying diodes on the secondary heatsink - they are way too undersized! There's absolutely no way this PSU will do 500W! I'd venture to say, not even 300W. You'll be lucky if you pull 200W before one of the rectifiers blows out. And I'm talking about 200W combined, so that's even less for the 12v rail, which is what most modern systems use these days.

Another problem is the huge loading resistors on the secondary side (10 Ohm on the 3.3v rail, 150 Ohm on the 12v rail, and another on the 5v rail I can't tell what it is). Those will cook you new caps. Remove them should you want to use this PSU in any computer.

With that said, I wouldn't use this PSU in anything modern. At best, it's maybe okay for a Pentium 3 or an older Athlon/Duron processor.
Sorry to sound so negative, but this PSU is just not good.

[Th3_uN1Qu3]

Re: Apevia WIN-500XSPX recap

I see everybody saying that the input rectifiers will blow up. Trust me, they won't. Even in PSUs with the 4-diode treatment - blown primary diodes are very rare, most often the switching transistors or 12v rectifiers fail. Also i only see Samxon and Meritek on the secondary - those are good caps. The output toroid is of decent size and there's pi filters too. The only thing i'd do is remove the minimum load resistors and it'll be fine.

[goodpsusearch]

Re: Apevia WIN-500XSPX recap

With that said, I wouldn't use this PSU in anything modern. At best, it's maybe okay for a Pentium 3 or an older Athlon/Duron processor.

It could be good for p4 1.5ghz - 2ghz SDRAM/DDR1 too.

But only after adding a proper transient filter and making sure that the 12V rectifier is 16A or more.

[Th3_uN1Qu3]

Re: Apevia WIN-500XSPX recap

The Q9400 has a TDP of 95 watts. That's on par with a 3GHz Pentium 4 btw. In a HTPC application hence no power hungry video card, that power supply will do fine.

[Wrog]

Re: Apevia WIN-500XSPX recap

Excellent, thanks everyone -- looks like I have some homework to do. I'll post back when I've digested enough to continue.

[momaka]

Re: Apevia WIN-500XSPX recap

I see everybody saying that the input rectifiers will blow up.

Everybody? Who's everybody?
If you're referring to my post, I mentioned the rectifying diodes on the secondary heatsink, not the input diodes/bridge rectifier. As you can see, all of the secondary rectifiers in this PSU come in TO-220 package. This suggests low current output capability.

At the very least, the current capability of the 12v rectifier should be checked. Needs to be rated for 16A or more, as goodpsusearch said.

[Th3_uN1Qu3]

Re: Apevia WIN-500XSPX recap

Everybody? Who's everybody?

A number of people on different websites. Especially the guys at Hardware Secrets (an otherwise great resource) drive me nuts when they keep saying that on the supplies with 4-diode treatment the input rectifiers will be the first to blow up, when it's almost invariably the primary switchers which go out in their testing.

[japlytic]

Re: Apevia WIN-500XSPX recap

The mains filtering section looks like it has been skimped on, especially when non-approved Y-class capacitors have been used...

[Lobo]

Re: Apevia WIN-500XSPX recap

Wow, old thread. Why not give it a bump:p I have an Apevia 500W PSU that also came with an Apevia HTPC case. Same model number as this one... would you believe the board looks the same but the caps are different?
Of course I am having problems with mine as well. I'm still pretty noobish at this so I need a little help. I'm good at soldering, but don't know how to troubleshoot PSU's. Can someone point me in the direction of a PSU troubleshooting guide here (if one exists). Or should I just use this thread for assistance?

[momaka]

Re: Apevia WIN-500XSPX recap

Can someone point me in the direction of a PSU troubleshooting guide here (if one exists). Or should I just use this thread for assistance?

I don't recall ever seeing one (but I could be wrong). Usually it's done on a thread-to-thread basis. That said, I suggest you create you own thread so that it gets more attention. And don't forget to post pictures . Good to know what exactly we are working on (and in the event someone spots something troublesome that the others didn't ).

A number of people on different websites. Especially the guys at Hardware Secrets (an otherwise great resource) drive me nuts when they keep saying that on the supplies with 4-diode treatment the input rectifiers will be the first to blow up, when it's almost invariably the primary switchers which go out in their testing.

True.

[Pentium4]

Re: Apevia WIN-500XSPX recap

I would increase the capacitance of the input caps too (I don't know what the originals were though) What is the model of the rectifier on the 12V? Seem to be rare but I've seen 30A on the 12V in a TO-220 package

[Behemot]

Re: Apevia WIN-500XSPX recap

…would you believe the board looks the same but the caps are different?

Yes, those images are actually dhot after the recap took place.

Anyway, if you wanna Samxon RS 3300 uF/16 V caps, I have them

[cashkennedy]

Re: Apevia WIN-500XSPX recap

If your going to change the input caps and the output caps, why not just throw it away and buy a corsair cx430 for 19.99 after rebate. Not exactly the best supply, but better then the apevia will ever be. (Unless you live out of the US and cant order from newegg / get rebate)

[Pentium4]

Re: Apevia WIN-500XSPX recap

If your going to change the input caps and the output caps, why not just throw it away and buy a corsair cx430 for 19.99 after rebate. Not exactly the best supply, but better then the apevia will ever be. (Unless you live out of the US and cant order from newegg / get rebate)

His Apevia already has high quality capacitors, for a low end build like his why not use it? The CX430 is going to have garbage caps in it, so even in his system I bet the Apevia would outlive the Corsair before it would need a recap

[Behemot]

Re: Apevia WIN-500XSPX recap

His Apevia already has high quality capacitors

[Pentium4]

Re: Apevia WIN-500XSPX recap

Sorry I thought we were still talking about Wrog's Apevia

[cashkennedy]

Re: Apevia WIN-500XSPX recap

Wrog replaced the shitty caps in his with samxon / pannys . Who knows what was there to begin with. If you read his original post he took the pictures after redoing all the caps except the last 2 that didn't have polarity markings.

The corsair cx430 according to jonnyguru came with 100% samxon caps. So if your a fan of samxon you should be happy there. (samxon is pretty middle of the road though) The corsair also has rather low ripple, well within spec. About the only downside to the cx-430 is that they're lying about it being 80 plus bronze.

http://www.jonnyguru.com/modules.php...tory4&reid=214

[Pentium4]

Re: Apevia WIN-500XSPX recap

The CX430 is indeed a good PSU for its price, and still being 80 Plus is fine in my opinion. But the Samxon they use are GF, which are worse than middle of the road. Most Samxon are pretty good caps, they have 2 "dud" series as far as I know