System Failure, Advice Needed

[Xtreme2damax]

I've been experiencing problems with my PC. I recently received a replacement PSU after RMA'ing the old PSU as I thought the PSU was responsible for my system rebooting and shutting down.

After replacing the PSU I am still continuing to experience shutdowns. I have now diagnosed the motherboard as the culprit. However I was to be completely sure the motherboard is responsible before replacing. Here's a description of the issues..

My system will either reboot or shut down unexpectedly, whether the system is idle or in use doesn't matter. The system reboots or shuts down anywhere between a few seconds since being powered on to a few hours.

After shutting down the system normally it will power itself back on most of the time so I have to hit the switch on the PSU.

The system will not post with the ram in slots 1,3,5 but it will post with the ram in slots 2,4,6. I've been experiencing this issue since I purchased this motherboard.

Sometimes after an unexpected shutdown the bios (CMOS) settings will be reset to the defaults. I've already tried changing the CMOS battery which didn't resolve this issue.

After an unexpected shutdown or reboot there will sometimes be graphical artifacts on the bios splash screen so I have to manually reboot the system to get rid of the artifacts.

Sometimes after an unexpected reboot or shutdown I'll have trouble powering on the system as in it the system struggles to power on. The system also will have trouble posting and hang with the cpu led on the motherboard lit. When I power on the system the cpu led on the motherboard will flicker before the system posts.

I do not think the CPU is the problem. Whenever the system is powered on running a CPU intensive application will not trigger a reboot. I've been monitoring temps and my cpu idle temp is between 38 Celsius - 42 Celsius even when the system is overclocked to 4.2 Ghz. My under load temps are anywhere between 50 Celsius to 70 Celsius. All my other components are relatively cool and not overheating. The system posts most of the time and when it stays powered on I can use the system normally.

Additional details: When I first installed the motherboard I accidentally dropped a thin strand of copper or metal wire in the PCI-e x16 2.0 slot that I thought I was able to get out. Another minor or major detail is that the clamp that attaches to the back of the motherboard to hold the cpu heatsink/fan in place might be shorting out the motherboard and killing my power supplies. I recently came to that conclusion, I originally thought that those were designed not to short out the motherboards.

Specifications: Core i7 950 (Currently stock speeds), Asus Sabertooth X58 motherboard, Corsair HX850 Semi-Modular PSU, 6GB Corsair DDR3-1600 Ram in Tri-Channel, EVGA Superclocked Geforce GTX 460.

Please recommend a plastic clamp for a socket 1366 motherboard that is compatible with a Thermaltake Frio as I do not want to risk shorting out another motherboard.

[jnelsoninjax]

Have you tried the stock cooler? Other then that the first thing I would think is the PSU, but seeing as you just replaced it and the other symptoms that you describe, I would due the following: Check that the RAM itself is not the issue, pull the motherboard out and visually inspect all the capacitors and other mount points, check all that all the fans are running and are free of dust, and as I suggested try going with the stock heatsink/cooler and reset BIOS to the defaults. If the issues continued after all this I would lean towards the motherboard.

[Xtreme2damax]

I'm sure neither the ram or CMOS battery are the issue. I've tried removing sticks of ram, seating and re-seating without any improvement. The system shuts down or reboots without any blue screens. The capacitors on the motherboard aren't standard capacitors, they aren't electrolytic capacitors so there would likely be no bulging.

I haven't tried the stock cooler yet and I am not sure that would make any difference. There is barely any dust in the system and all components are relatively clean, CMOS/Bios settings are at the default stock settings. I just want to be certain that the issue isn't something more major and costly such as the cpu. If replacing the motherboard doesn't solve my issues then I will likely be considering another brand for my psu if that is indeed the issue again.

[jnelsoninjax]

One other thought just came to me. it almost sounds like the PSU is not powerful enough to power the whole system. I know when I upgraded my GPU, I ended up purchasing a supplemental PSU to run the card. Though if I read what you posted the PSU is an 850 Watt, which should be more then enough to run the system. Do you happen to have another MB laying around that you can try the CPU in? You might be able to buy a cheap board and see if the system works, that way you would almost have a definitive answer, and if your careful enough, you might be able to return the board back to the store, if you do not end up needing it.

[Xtreme2damax]

I do not have any spare parts lying around although I wish I did now. I bought the motherboard at Newegg.com and was going to RMA it last year but I foolishly peeled off the sticker on the PCI slot that had the serial number on it, ruining the sticker in the process because I didn't feel like removing the motherboard just to get to the serial number. I'm not sure if I would be able to RMA the board without the sticker with the serial number in tact and there is no serial number on the box that the motherboard shipped in.

[jnelsoninjax]

All I can say is contact the manufacturer and ask about the RMA process and see what they tell you!

[Xtreme2damax]

I replaced the motherboard today, FedEx came early with my shipment. Ran to the store to pick up some wipes and isopropyl alcohol, pulled the the heatsink/fan, cpu, cleaned off the old thermal grease, gave my heatsink a bath to remove caked on dust, put the new cpu in, applied new thermal grease, attached heatsink/fan, installed motherboard and hooked everything up. Waited most of the day to get my system up and running as I had to wait for the heatsink to dry completely which took a few hours.

Glad to have everything running again, so far so good.

[jnelsoninjax]

Cool! Hope that the board was the whole issue!

[Xtreme2damax]

Yeah I hope so as that would be a big relief to me. I have a question, what would the average idle temp and load temp be for a system overclocked to 4.2 Ghz on air cooling? On idle I am getting between 38 Clesius to 45 Celsius and on load I am getting between 50 Celsius to 65 Celsius. Also is it normal for the cpu led on my board to flicker when booting the system?

[jnelsoninjax]

I would say the temps are fine, my AMD is at 54 C, atm, and it's only air cooled. As far as the CPU LED, I am not sure what you are referring to.

[Ntrstd]

Any temp under 65C should be fine. I say this because the capacitors are usually rated at 65C.

Overclocked systems will spontaneously reboot because of timing issues. This happens a lot with games because they usually run the computer at 100% load regardless of clock speed.

The system will not post with the ram in slots 1,3,5 but it will post with the ram in slots 2,4,6. I've been experiencing this issue since I purchased this motherboard.

I think this is because of banked RAM and terminating connections. So the 6 slots are divided into 3 banks. Slots 1-2, 3-4, 5-6. To electrically terminate each bank, there has to be RAM in the second slot of a bank. So if there is an odd number of RAM sticks, that's slots 2, 4, 6. Or maybe 1, 2, 4.

Any more RAM would go into slots 1, 3, 5. Or maybe you would use slots 1, 2, 3, 4. Many mb are really made like this.

As you have implied, the only way to find out about this is to read the mb manual and experiment.

We had a brand of computer with 80% failure rate of the DVD burners. The computers would just turn themselves off. Replacing the burners stopped the unexpected shutdowns and reboots.

Also look for power interruptions, do your lights flicker often?

My guess is it's a RAM compatibility issue.

[Xtreme2damax]

Any temp under 65C should be fine. I say this because the capacitors are usually rated at 65C.

Overclocked systems will spontaneously reboot because of timing issues. This happens a lot with games because they usually run the computer at 100% load regardless of clock speed.

I think this is because of banked RAM and terminating connections. So the 6 slots are divided into 3 banks. Slots 1-2, 3-4, 5-6. To electrically terminate each bank, there has to be RAM in the second slot of a bank. So if there is an odd number of RAM sticks, that's slots 2, 4, 6. Or maybe 1, 2, 4.

Any more RAM would go into slots 1, 3, 5. Or maybe you would use slots 1, 2, 3, 4. Many mb are really made like this.

As you have implied, the only way to find out about this is to read the mb manual and experiment.

We had a brand of computer with 80% failure rate of the DVD burners. The computers would just turn themselves off. Replacing the burners stopped the unexpected shutdowns and reboots.

Also look for power interruptions, do your lights flicker often?

My guess is it's a RAM compatibility issue.

I read the manual and it appears that the slots for ram are as follows:

A2A1B2B1C2C1

Slots A1, B1 and C1 are the recommended/required slots for tri-channel operation which is why the system wouldn't post with the ram in slots A2, B2 and C2. Since replacing the motherboard I have not had a shutdown yet, the system is more stable, fast. Also everything powers on without hesitation and when I shut down the system actually stays off instead of powering back on it's own. No issues with the CMOS being reset to defaults, still testing everything for stability and to see if the issue is truly gone.

[Xtreme2damax]

Here is the link to the Asus site for my motherboard:

http://www.asus.com/Motherboards/Intel_Socket_1366/SABERTOOTH_X58/#overview

I just want to know, preferably from other owners of this particular motherboard if the red cpu led flickering when powering on the system is normal.