Somethings dead? MB, PSU, CPU?

Kenny007 · Joined: Oct 30, 2008 Posts: 4

I posted elsewhere with this, but not too much help so far, and you guys sounded awful knowledgeable in another thread that landed me here, so here goes.

Well, there's a first time for everything, so here goes. I generally don't ask for help on forums, I just pace around the room and swear and keep trying til I figure the issue out. Anyways, the issue...

I built this PC over a year and a half ago. Some of the major components are as follows

EVGA 680i SLI A1 Motherboard
E6600 CPU
4x 1GB Kingston HyperX 800MHZ DDR2 memory
1x EVGA 8800GTS w/ 640mb
150GB WD Raptor HD
700w OCZ GameXSteam PSU
Antec 900 Case

(Let me know if you need more specs)

Anyways, like any other day, I wanted to game, so on my machine went or so I thought. Didn't want to start though...so I looked in the case and my side fan came out of the panel some, so I reseated it. After one more failed try, on the PC went.

I browsed the net for about 10 minutes, then down the PC went again. Whenever the PC shuts off, it doesn't want to come back right away. A short wait later, and the PC is running again.

Now my concern is growing, as this is unheard of for my rig. I have had an overclock on it of 3.2GHz for the majority of it's life with no issue, so I began to suspect that. This time around I ran ORTHOS on a blended test for about a half hr with no issue, though apparently the moment I left the room, the PC shut off again....

So, for the first time in nearly 2 yrs, I decided to bring the PC back to default specs. Restored the BIOS to defaults, and upon selecting Save & Exit, the PC shut off again...what the hell....

Time to start the suspect part elimination. I took out memory in pairs; either way the PC failed. Tried just one...failed. Removed both Hard drives and started the PC. With nothing to boot to, it eventually tried to boot off the network. While searching, it shut off again....

Removed all optical drives, failed. Removed my sound card (only expansion card outside of the GTS), failed.

Was too late to continue, so I called it a night.

Now at work, I've pretty much boiled it down to PSU, CPU, or MB. Granted the issue screams of an overheating issue, but with the 7ish fans in the tower, PLUS the Tuniq Tower I have on my CPU (google it if you havent heard of it...best air cooler I've ever owned....was always 23c or less idle overclocked), heat was never an issue, which was confirmed in the short span my PC was on. Sadly, I no longer work at a PC repair shop like I did when I built it, so parts are scarce to switch out. Now, as corporate IT, this place has LESS to work with. In addition, I moved out of state, so friends to pawn off of are scarce.

My last intention is to test the PSU with my former unit, a P4 tower, to see if it shuts down the same. My only concern though is that the P4 unit wont be asking for even half the power my current rig did, so it may provide a false positive.

Having dealt with countless individuals PC issues, as well as built over 200 PCs myself, I'm leaning towards a motherboard. At the shop I used to work at, it always seemed that CPUs were generally the survivor when a motherboard or CPU were suspect; they just seemed more resilient. I should be able to RMA the board or the PSU if needed, but I can't be sure which it could be. Any help on this would be GREATLY appreciated, as I'm at wits end with this at work while I can't work on my rig at home.

Thanks again!

goretsky · Posted: Fri Oct 31, 2008 2:48 am Post subject:

Hello,

Based on your description, I am inclined to think the power supply first, but it could be an issue with the motherboard or the processor. A few thoughts, in no particular order:

Keep in mind that fans become less efficient overtime as their bearings wear out, lubricants dry up and they accumulate dust and debris, so it could be a cooling issue, although I don't think it would be a problem given the age of the computer unless the fans were very dirty or airflow was obstructed at the points of ingress or egress.

The Tuniq Tower is a fairly heavy CPU cooler, at 798 grams (~1.76 pounds). Is it possible the cooler itself has become loose or even deformed the motherboard so it is no longer making good contact with the processor's integrated heat spreader? I can envision a scenario that as the CPU heats up the IHS expands and contact surface area with the base of the cooler decreases. It might help to remove both it from the CPU and the mounting hardware from the motherboard, inspect them, and reassemble them. This would also be a good opportunity to replace the thermal compound..

Do any of the passive components on the motherboard such as the capacitors look damaged (bulging, leaking electrolytic fluid, et cetera)?

Regards,

Aryeh Goretsky

Kenny007 · Joined: Oct 30, 2008 Posts: 4

Thank you for the prompt reply.

Since my initial posting, I have disassembled the entire unit, cleaned everything inside and out, and reassembled. Sadly, the problem just got worse. Now, rather than staying on for a minute or so, it wont fire up more than half a second. Not even long enough for the fans to get up to speed.

Given the young age of the board, I've begun an expedited RMA process on it, though I still can't be sure its the root cause. If it comes back and still doesn't address the issue, I'll see what I can do about switching out, or returning the PSU. I put the PSU in my P4 tower for a while, which I thought was working last I used it, but it would not boot past POST, though it did stay on at least. In retrospect, maybe another bad sign leading to the PSU, though I have had intermitant issues with that unit, hence the build of this new one in the first place...

I've also ordered a generic cooler, if only for ease of installation during testing, and more thermal compound. I had 2 tubes of it while working yesterday, but it must have been old, as it was harder to spread than normal.

I hope it is one of those two items, so I can avoid buying another processor, and ESPECIALLY avoid buying another video card. Despite it's 2 yr age, it's price hasn't dropped as much as I'd like it to...

drwho07 · Joined: Nov 29, 2007 Posts: 1546 Location: Central FL, USA

Kenny,

Just from your description, it sounds more like a heat related problem than anything else.
Are your fans on the CPU and Video card clean, and turning freely?

Look closely at the motherboard....... then look at the electrolytic capacitors, those little cans with the shiny tops. Are any of them bulging, split open or leaking a brown or yellow substance? If they are failing, you should just replace the motherboard.

If you've removed the power from all the drives and this PC still shuts down, I'd not be looking too strongly at the PSU.
But in my case, I'd jack in a substitute PSU, just for testing purposes.

The only time I've ever seen a PC shut down due to the CPU, was when the cooling on the CPU had failed. The chips themselves have so many safe-guards built in, that they almost never fail.
However, overclocking will cause a CPU to run hotter than normal, which, I suppose, could shorten its life.
At least once a year I take out my CPU cooler and wash the heatsink and thoroughly clean and then oil the fan.

Do some more troubleshooting before you start buying new parts.

Troubleshooting and maintenance is cheap.....buying parts is NOT.

Good Luck,
The Doctor Cool

Kenny007 · Joined: Oct 30, 2008 Posts: 4

Oops...I did forget to mention in my previous post that I had looked at the capacitors... I do know what to look for in those; I've seen one fail in front of me! Made quite a pop!

Shortly before I decided to begin the RMA, I did have next to nothing in the unit while trying to power up; the CPU and cooler, PSU, and board. No memory, drives or extra fans. Still powered down quickly.

While I'm not ruling out an OC related problem with the CPU... I'd be surprised if that were the case. After my cooling efforts and addition of the huge cooler, my temps on average were generally lower than stock, or at worst, on par. Voltage addition was minimal too. I wish I had another 775 processor laying around like I used to have access to...or even a spare PSU that could power my rig.

I'm not terribly opposed to buying another CPU either, at least they have come down in price considerably....I just don't want to buy a bunch of things I dont need. Hopefully the board is to blame and the RMA is the right procedure...otherwise I'll go mad.

bern · Joined: Mar 12, 2007 Posts: 1432 Location: ann arbor

drwho07 · Joined: Nov 29, 2007 Posts: 1546 Location: Central FL, USA

That 'memory' thing jumped out at me too.

For 'most' systems to even boot up, you must have the CPU and fan,
ram memory, a video card (unless the video is on the motherboard,
and a keyboard. Mouse is optional and not required to boot the system.
A hard drive and/or CD drives is NOT required.
Try booting to a floppy drive if you have one.

Good Luck,
The Doctor Cool

goretsky

Hello,

Do you have (or have access to) an ATX power supply tester? It can tell you if the power supply is putting out a PWR GOOD signal and if the voltages are with acceptable ranges.

Regards,

Aryeh Goretsky

Kenny007 · Joined: Oct 30, 2008 Posts: 4

An update for those nice enough to have been providing suggestions.

All my testing led me to a motherboard, seeing that the PSU was powering another unit of mine and I figured I'd eliminated all other easily figured options. So, I paid a small fee and cross shipped my board...

Nothing. Not only was the failure of this switchout disheartening, but the fact that my original board was now obligated to go back, and that it appeared to be in better shape than its replacement, was a bummer. But back it went.

Next in line was a PSU change. OCZ has always been great with RMA's, but wanted me to try a few things first. All they're testing "suggested" the PSU was fine, but I had my doubts, and bought a new one. After installation, the PC fired up. High 5!

However, Windows was not pleased with the various crashings it had suffered over the past week or two, and a CHKDSK was ran, as well as a few drivers reinstalled. I managed to correct all the issues that stemmed from this, save one.

My DVDRW and DVD drives, both SATA, do not work now. The DVDRW is detected by the BIOS, but will not read any media within, and the DVD one just isnt seen at all. I updated this new board's BIOS to the current one, and tried various cablings.

Any ideas?

goretsky

Hello,

Have you replace the battery which powers the CMOS/RTC memory? On most modern motherboards, it is a +3VDC CR-2032 lithium coin cell, which can be found at most places which sell batteries for a few dollars.

Regards,

Aryeh Goretsky