sudden system shutdown

[colosssus]

Well i decided to try a few "simpler" methods before doing an outside build. Unfortunately, the wireless card seems to have nothing to do with it, as even completely uninstalled, the system still crashes as always. However!! some progress has been made, i tried using a different video card, a geforce 210. and sure enough NO PROBLEMS! It ran the benchmark tests to completion, albeit terribly slow in comparison. I tried using different versions of directx, older drivers for the card, no dice. I talked with the EVGA guys and i turned off the automatic restart feature on windows....still shut down right away. I was told it must be a short between the videocard and the motherboard, and would require an outside build and alot of testing to figure out exactly what is going on. Frankly, it's just not worth it at this point. i've spent all night every night for a week troubleshooting, two weekends, and a ton of money. i'm just going to go with an ati card which is really what i wanted to do in the first place, but couldn't with my old motherboard. If by some stroke of incredible bad luck my problem remains......well i dunno at that point. I did by the way try running it with just the absolute minimum components, and reformatted and reinstalled windows. I even tried running the benchmark without ANY driver update whatsoever. Ram was all doublechecked. The only thing i noticed was both the ram, and cpu frequencies were lower then they were supposed to be. ie the ram was set at 1066mhz, instead of 1600 as it should have been, and the cpu is still set at 2.6ghz, instead of the 2.8 it should be. I changed the ram, but left the cpu where it was. I'm thinking of switching to a 5770 for the video card as it's about the same price, Any advice there? Thanks for all the help you guys did give me, it's certainly appreciated.

 

UPDATE: After a little more research, i found out my psu only has 1 12v rail. Now i have tried powering only the motherboard, videocard, hardrive, and either one, or two fans...and it still powered down. so maybe this psu, and my original psu could not handle the power demands of the videocard, although being a 1000w you'd think for sure it should be able to. The corsair that we tried i'm 99% sure was defective off the bat, as it could not even power up the system properly, ie. the fans on my cpu cooler weren't getting power, and neither were any of the case fans. So, lol again, i think i'll try a dirfferent powersuppy....a modular thermaltake 1000w cpu with 2 12vrails. This being the 4th power supply unit by god if it still powers off i WILL switch video cards. I can't say how thankful i am my wife works for a computer supply store, and has been able to swap things out at as needed.

Edited August 10, 2010 by colosssus

[l1o2l]

Lucky you, your wife works at Tigerdirect.

 

Just wondering, what were the amps on the previous power supplies?

[ir_cow]

Lucky you, your wife works at Tigerdirect.

 

haha that might be the whole problem, tigerdirect sells junk and think more than half the people here have been burned with fake rebates. anyways that was a rant but for the PSU a Corsair 850w may have a single 12v but its one massive line and no way can it NOT handle that setup even overclocked. i did see you were saying putting in a old card worked fine and that card was in your old build so don't let Evga try to blame the board i think your cards bad.

 

 

lastly have you ever tried a different wall outlet? i know it sounds random but maybe the whole problem is the power level its putting and your breaker is turning off. just a thought.

[wevsspot]

Colusssus - the Corsair TX and HX series power supplies are some of the best on the market right now. I know for a fact that the 850W Corsair will power your system and more without issues. I'm running the 750TX and it powers a butt load of gear without issue and I'm at about the max on that unit.

 

I hate to see you continuing to spend money without truly knowing what the cause of the random shutdowns are.

 

I can understand and appreciate your frustration, but the last thing I would do is drop more money on another power supply because I really don't believe that is the problem. Especially if you're still using the Corsair power supply.

 

The HD 5770 is a good card at a fair price if you want to give an ATi card a try. Or if you wanted to spend a little more the new nVidia GTX 460 cards are outstanding.

 

My previous recommendation remains the same. I'd still build outside the case, minimum hardware. Continue using the Corsair 850TX power supply, but slap another video card in and run your tests.

 

In order to get your ram running at it's rated speed you'll need to make some BIOS adjustments. But we can get to that once you've got the random shutdown issue resolved.

 

Regards,

 

JW

[ebarone]

Its entirely possible that your PSU's have been bad the whole time. In my experience PSU's rank number one in frequency of DOA, and I've never heard of a "random shutdown" problem that wasnt the fault of the PSU, be it inadequate power or just a busted unit. That being said, I strongly encourage you to remove it from the case and build the system. Not only will it speed along your swapping of components, but it will once and for all remove any shred of a chance of there being a short between either your board, video card, or PSU and the case itself, which would instantly cause a shutdown. It will also help you to avoid any human errors, such as forgetting to plug in the auxillary power plug on the motherboard, since everything will be plainly visible without a lot of backache.

 

One time I put a standoff in the wrong screwhole on the case, so it came in contact with a solder point on my motherboard... the entire system would not work. Bottom line is, you always want to remove as many variables as you possibly can when trying to track down an intermittent problem like the one you're having, because you just never know what it could be.

[Syngensmyth]

I don't know how to find this, but it could be a cracked, weak or otherwise damaged trace on the motherboard itself. The heat builds up and expansion "disconnects" the trace (or whatever).

It would be interesting to see if XP has the same problem. I assume it would, but I have had driver conflicts cause sudden shutdowns. Truth be told, I would have abandoned the motherboard long ago so you are a patient person. Sudden shutdowns can be a mind bender, as you are finding.

 

I'm with Wev on the outside build.

Edited August 10, 2010 by Syngensmyth

[bp9801]

I don't know how to find this, but it could be a cracked, weak or otherwise damaged trace on the motherboard itself. The heat builds up and expansion "disconnects" the trace (or whatever).

It would be interesting to see if XP has the same problem. I assume it would, but I have had driver conflicts cause sudden shutdowns. Truth be told, I would have abandoned the motherboard long ago so you are a patient person. Sudden shutdowns can be a mind bender, as you are finding.

 

I'm with Wev on the outside build.

 

Its the same problem affecting two different motherboards.

[cjloki]

send all your stuff to me and i'll build it for you cuz near as i can tell you're jinxed....the only common denominator is operator headspace ! i won't charge you a red cent and you can pm me for the mailing addy

[ir_cow]

i have to say alot of people missed the whole point, by this point the OP already swapped out everything but the video card. i could see people leaning towards a bad board or bad PSU but all this was already going on which is why the OP ended up switching each part 1 at a time . wevsspot has a good point and do bare min, when the OP did it worked fine with a generic card. anyways i've had computers turn off suddenly for overclocks, ram compatibly, video cards and PSUs when put on stress tests so saying only psu cause system to turn off isn't 100% correct. yes i had my share of PSU problems from under-voltage but not normal use or a good PSU.

[bp9801]

i have to say alot of people missed the whole point, by this point the OP already swapped out everything but the video card. i could see people leaning towards a bad board or bad PSU but all this was already going on which is why the OP ended up switching each part 1 at a time . wevsspot has a good point and do bare min, when the OP did it worked fine with a generic card. anyways i've had computers turn off suddenly for overclocks, ram compatibly, video cards and PSUs when put on stress tests so saying only psu cause system to turn off isn't 100% correct. yes i had my share of PSU problems from under-voltage but not normal use or a good PSU.

 

He was still using the same wireless card as well, which could have caused the problem. Yes, some people may think its the PSU when honestly, I don't see how it could be considering the problem has persisted over two different PSUs. The two constants have been the wireless card and the video card. Rule one out, then the other, then both, and then we'll see. Since the video card was taken out for a different one and the problem went away, then the video card could be the culprit. I highly doubt a 1000W PSU couldn't power a simple GTX 465 so it seems that the card was the problem. I say to RMA the video card to EVGA and stick with the GeForce 210. Let the system run for a while with that card, say a few days, and see what happens.

[colosssus]

Well i finally solved it! it WAS the psu the whole time! I switched the psu and wouldn't you know, it frigging works! Now why anyone would make a 1000w psu with apparently pretty craptacular limits on the 12v rail is beyond me. And also, why the 800w psu from a different manufacturer also couldn't cut it is beyond me. The corsair was a defective unit...which just compounded the confusion that much more. So i guess lesson learned here is not only is the maximum wattage important in a psu, but also the number of rails..and i would imagine the power going to them. all in all a very expensive, time consuming lesson, but i don't mind being the guinea pig if it means some other poor guy doesn't suffer the same fate. Right now everything seems to be working fine, or at least it's stable. There is one thing that bugs me though, when i power up, it posts, shuts down, and restarts..then loads everything fine. At this point i'm very willing to live with that at least for a week or so while my poor knees and back get a chance to recover from being bent over on hardwood floors every night! Thanks again to all your help and suggestions. ^_^

[bp9801]

Well I stand corrected. At least its all fixed!

 

Btw, the number of rails doesn't mean anything imo. Its how many amps it can handle. A single 12v rail is a good unit provided it has enough amps to handle everything you're doing. Similarly, a multiple 12v rail can also be good provided you don't overload a rail with too much. Multiple rail units can handle 20-30 amps per rail so you have to be careful to not go over that if you are running say, dual video cards off one rail. A single 12v rail unit won't have that problem while a multiple one will.

 

I'd RMA the PSUs and get new units though.