PDA

View Full Version : Just killed my second Masscool 460W


schifter
01-14-2007, 03:33 AM
So my first Masscool (Sytrin/Nexterm) unit failed when my 7900GS died and I put my 7800GT in the machine (system specs below). It would run for a few minutes under load, then shut down. Turn it back on, and it would work again for a while. The 12V line stayed steady at 12.12V no matter what. I did this a few times while trying to troubleshoot the problem, until the power supply wouldn't come back on.

Ordered a second one, while I waited for the vendor to tell me if the first one was under warranty. 50 bucks, I figured it was worth a shot.

The new one lasted about three and a half minutes before it shut down. Just like the first one, the lines were perfectly stable. I took the 7800GT out and put in a 6600GT, because at this point I'm sure the 7800GT is overloading the unit. Turned it back on, and watched the magic smoke roll out the vents on the side.

So... This power supply tested good when Jonny tested it, but it sure doesn't like me. Two bad units? I'm not sure about that. A friend of mine ordered one on my recommendation after I got the first one, and his lasted about three months.

System specs:
Athlon64 Venice core at 2.5GHz / 1.6V
DFI nf4 Ultra-D
2x512MB BH5 DIMMS at 3.13V
2 7200rpm hard drives
1 optical drive
3 case fans
7800GT volt-modded to 1.5V, heavily overclocked, about equivalent to a 256MB GTX

Looking at these specs, I can't see any way I could be using 32A of 12V. Now, if this thing has true independent rails, and the only thing on rail 2 is that power connector for the refrigeration unit, I could see the problem.

On top of this, I still haven't gotten a response to my warranty question. I'm about to sacrifice the possible warranty on the first one to see how it failed.

In the meantime, I suppose I'll be getting something different. The Corsair HX620 looks good. I didn't really want to spend that much, but at this point overkill seems like a good idea.

madmat
01-14-2007, 05:48 AM
Sounds to me like your PC has other problems that were causing the PSU to shut down on protection of some sort. I'd say that once you get a new power supply you'll either have it keep shutting down or it'll have enough ass behind it to take out whatever it was that killed the Nextherm.

I hope I'm wrong. Either way good luck.

schifter
01-14-2007, 06:34 AM
The machine ran fine with the 7900GS in it. After the first time it shut down, I put in the 6600GT and ran it for a couple days.

The 7800GT came straight out of another machine and into this one, as in, it was working in the other box, then 10 minutes later this one was misbehaving.

I can't think of anything to blame except the power supply. I suppose I'll just take one of my Enermax units out of another machine and put in there to test. I just really didn't want to take another machine apart.

Before someone mentions me running things out of spec, it did the same thing with BIOS defaults. The only thing that was running over its rated speed or voltage at that point was the video card.

madmat
01-14-2007, 06:51 AM
The 7900GS should be about the same draw as the 7800GT and far more than a 7600GT, this is why I think that the 7900GS might have damaged something else when it died.

Either way, good luck and I hope you get it figured out.

jonnyGURU
01-14-2007, 08:26 AM
There's no way that's a power supply issue. Not two in a row like that. The way the scenario is playing out in my head is, there's a short and the PSU is trying to trip to protect itself and the rest of the computer. Eventually, the short was quicker than the protection circuitry and the PSU blew up.

Sounds like when you installed the video card, maybe you crossed some of the contacts in the PCI-e slot and shorted out the board or some other damage occured to the motherboard during the transplant of the video card.

If the PSU isn't on it's way back, I'd try to bench test it again with a paper-clip or PSU tester to see if it comes on. If it "tripped" sitting off the build for a couple minutes will definitely let it reset.

If the first PSU is already on it's way back, I'm skeptical about trying the Enermax at this point. Wouldn't want you to be on your third non-working PSU when that one blows up.

schifter
01-14-2007, 12:18 PM
If the PSU isn't on it's way back, I'd try to bench test it again with a paper-clip or PSU tester to see if it comes on. If it "tripped" sitting off the build for a couple minutes will definitely let it reset.


I tried that with the first one when it wouldn't come back on. It's really dead.

The second one didn't come back on after the first time it shut down.

The thing works perfectly with the 6600GT in it. That says to me it's not the motherboard, processor, or memory. When the 7900GS died (It had the memory problems lots Samsung-memory 7900 cards did. I can't see how that could've damaged something.), it took a week to get a new card, and it ran on the 6600GT the whole week.

I can't think of any way anything could be damaged in such a way that a 6600GT works perfectly, but a 7800GT doesn't.

jonnyGURU
01-14-2007, 01:18 PM
You can short the video card and not the motherboard.

Also, keep in mind that the 6600GT doesn't use the 6-pin PCI-e connector. The 7900 does. That's another variable.

schifter
01-14-2007, 02:51 PM
Okay, so the machine has been sitting on my desk running on the Enermax with the 7800GT for a little over half an hour now, running primes and rthdribl.

Now, I did pull the board out of the case to check the mounting and make sure I didn't have the board shorted to the case somewhere, but, as far as I can see, I didn't.

So, I think I'm going to assume there was a bad batch of power supplies. The serials on the two I have are 51 units apart. The one that died on my friend probably came out of the same batch, since he got it in between the two I got.

The vendor's web page says they only offer a 14-day warranty, and I can't find support info on Sytrin's page, because the page is broken. So I suppose I'll open the older unit and see where it failed.

jonnyGURU
01-14-2007, 03:15 PM
I hate that "bad batch" theory.

Quality control in the factory is better than allowing more than 50 bad units out in a shot.

14 days? That reseller sucks. Too bad. But before you open it:

http://www.masscool.com/masscool/warranty.asp

schifter
01-14-2007, 03:16 PM
I don't see anything obviously wrong on the board in the unit. Though when I take readings across the power switch, I get 50V, which seems a little odd to me.


EDIT: Yeah, too late on opening it. ;)

Spectre
01-14-2007, 04:32 PM
I hate that "bad batch" theory.

Quality control in the factory is better than allowing more than 50 bad units out in a shot.

14 days? That reseller sucks. Too bad. But before you open it:

http://www.masscool.com/masscool/warranty.asp

Yeah that has to be one of the more overplayed cards in existance. Manufacturers have their QC protocols worked out very well and they know exactly how to sample to keep things in check......missing one is very likely....but missing dozens or more...not really.

schifter
01-14-2007, 04:42 PM
I'll be glad to accept any other explanation for this. I don't really believe that a company like SevenTeam would let that many bad units out of the factory either, but I'm pretty much at a loss for an explanation for two dead power supplies, especially when one died within three minutes of use. Not just shut down, but smoked a component.

That, and the Enermax is still running the system under load, two and a half hours later.

Currently trying to decide between the Silverstone OP650 and the Corsair 620HX.

EDIT: Jonny, you want this one I voided the warranty on? I'll be glad to ship it to you. :)

EDIT #2: Went with the Silverstone.

piscian18
01-16-2007, 02:34 AM
This is shifters friend.

Mine blew on these specs. Keep in mind I'm not at home so I don't have exact details.

Asrock Dual-Sata II 1.5 bios
athlon venice core 3200+ 2400mhz
BFG 7900GT OC
PNY verto 3200 2x512 400mhz
Seagate 120gig sata I
Nec 1300A dvd+rw
3 80 mm fans

blew about 2 months ago while changing resolution on CS:S. Ironically it's been running fine since on a lower end superflower 500w jonny reviewed. I RMA'd the masscool with Gogofan.com but, they just sent me another one off the shelf rather than sending it back to masscool so I'm not in a big hurry to stick it back in my machine. The superflower isn't meant be permanent. Probably pick up one of these http://svc.com/rs-500-asaa.html.

schifter
01-19-2007, 12:44 PM
So, here's the conclusion:

Got the Silverstone OP650 yesterday. It's been running the system for over 24 hours under load now.

Masscool offered me RMAs on both of their units, even when I told them I opened one of them. I'm not sure what I'm going to do with two of those Masscool power supplies, but oh well.

GalvanizedYankee
01-19-2007, 01:13 PM
RMA the Masscool units...Then sell'em. ;)

If you can see the UL number on the OP-650's sticker, would you please post it here.
It starts with the letter 'E'.

schifter
01-27-2007, 01:57 PM
There is no UL number on the outside of the OP-650, as far as I can tell, nor is it in the manual.

MrWicked1968
01-27-2007, 02:19 PM
the Olympia is supposedly built entirely in house, so even if it did have a UL file number, it would likely point back to Silverstone.

GalvanizedYankee
01-27-2007, 02:34 PM
I would doubt most seriously that they are built in-house. They might be assembled and tested in-house but even that I doubt. I doubt PCP&C does any in-house building...of any retail unit.
A test unit might have been built, then sent to Taiwan with instructions to build these. That I can believe.

My contact at Enhance has been in meetings all last week and has not returned my calls. His secretary said to call him on Monday. I want to know if the OP650 is Enhance built but we will know when one is opened up.

jennifer could not provide a UL number for her supplier either. ;)

jonnyGURU
01-27-2007, 03:00 PM
I have one of the OP-650's and I can confirm a complete lack of UL #.

I haven't opened her up yet because I'm too busy getting the 1kW Shootout up by this afternoon and the Etasis 850W done for tomorrow, but this week I will have it opened up and on the load tester.

I have a hunch it's made in Enhance's new Taiwan factory, but I can't be too sure w/o some sort of confirmation/proof.


A test unit might have been built, then sent to Taiwan with instructions to build these. That I can believe.

Well actually, Silverstone is based in Taiwan, and they did hire away a couple engineers from Etasis, so all of this would have happened on the little island of Formosa. :D

Bbq
01-27-2007, 03:40 PM
1kw shootout.. me likey