GA-X58A-U3DR won't post after working successfully

[cjhod1]

I'm having issues with my Gigabyte X58A-U3DR mobo not posting, even though it was working perfectly well last night. Specs are as follows:
- Gigabyte GA-X58A-U3DR (Rev 2.0) mobo
- Intel Core i7 950 (factory speeds)
- 6GB Corsair DDR3 1600mHz Dominator RAM (factory speeds)
- Corsair TX 750w PSU
- Gigabyte GeForce GTX470OC edition GPU (running factory setting)
- 3 x 1TB Seagate 7200rpm HDD's running RAID5
- Windows 7 Ultimate 64bit OS
- Pioneer 22x SATAII DVD-RW

The computer was operating perfectly well yesterday afternoon, for 3-4hours straight, no heat/voltage issues or error messages at all. I go to start the computer this morning and a range of funny things happened:
1) First sign of trouble was when it wouldn't post on cold start, so i turned off the power on the PSU, and at the plug in the wall, and restarted. Got to POST, then was asked for DISK BOOT DRIVE (even though OS was already installed)
2) Reset the computer to go into the RAID settings, only to find one of the HDD's had an "Error Detected" and hence the RAID was no longer effective.
3) I wiped all of the RAID setups, and re-configured the whole system to run RAID5 again (formatting the HDD's effectively)
4) When I went pressed restart to begin installation of the OS again, it wouldn't POST, or let me get to the BIOS or CMOS settings.

After the above I noticed the lights next to RAM Slot #5 (furthest from the CPU) had 2 x Orange and 2 x Green LED's, indicating something wasn't right. Then they just turned off all of a sudden, and I noticed the CPU Voltage and CPU Temp lights came on, even though the PC had only be on for probably 15 seconds.

I've tried the following:
- starting with only 1 RAM stick in (tried all 3 different sticks too), no luck.
- pressed the CLR_CMOS button numerous times and disconnected the PC from power for some time, no luck.
- re-seating the CPU (carefully) and looking to see if any of the CPU pins on the mobo were bent (which none of them looked out of place)

After trying all of the above, the CPU LED's are no longer coming on, but the RAM lights are still illuminating as above.

I honestly don't know why it is not working, seeing as the system worked perfectly fine last night, and after this most recent shut-down it just won't work now.

Any suggestions are most welcome.

[Dark Mantis]

Hi and welcome to the Gigabyte UK Forum.

Funny situation indeed. Though don't worry about the LEDs illuminating at startup, that is normal procedure and then they go off again after a few  seconds as POST progresses.

I would try replacing the motherboard battery, even though I don't really expect it to be that but at least it is out of the equation then. It is a CR2032.

Afterwards I would suggest just using one hard drive attached to the Intel ICH10R southbridge port.

Do a complete CMOS clear Remove the power cable from the mains supply and then press the power switch on the case for a few seconds just to drain any residual energy in the PSU capacitors. Once done remove the motherboard battery for at least ten minutes before replacing it. Next plug back into the mains supply and boot. You will now need to enter the BIOS by pressing DEL and load Opimised BIOS Defaults. Make any other changes to the BIOS settings to suit your self and then press F10 to save and exit.

Post again afterwards to let us know how you have progressed.

[cjhod1]

Thanks for the response. I won't have a chance to have a crack at this until Saturday morning (Aus time = 36hours from now). I will report back then.

As an additional question, is there a chance the PSU could be the culprit? Or are Corsair PSU's generally pretty reliable?

[Dark Mantis]

Yes no problem just report back after you have managed to have a look at it.

There is always a chance that it is the PSU and if you have the chance to swap it out for another one to test for that then I would suggest doing so. At least it will put your mind at rest over that scenario. Corsair are a very good make though, but like anything they have their problematic ones. As you will notice from my sig I also run a Corsair PSU.

[cjhod1]

Once again, thanks for the response. I have a mate with a near identical setup to mine (only difference is GPU (nVidia for me, ATi for him), CPU (950 for me, 930 for him) and PSU (750w Corsair for me, 650w Thermaltake for him) between our two systems). His has been working flawlessly for some time now. I will be mixing and matching parts (to eliminate variables) with him after trying the above fix.

I will report back once that has been done.

[cjhod1]

Ok - an update.

Yesterday morning I replaced the battery for the CMOS, held down the pwr button and clear CMOS button for a good 2-3mins, and then tried to power up the system (with only the CPU, 1 stick of Ram, GPU and PSU connected - sitting in the case).

The mobo posted, but went into a really weird BIOS loop, where it wouldn't get to the disk boot failure, just the initial post screen then restart. The mobo had recognised however it was having posting issues (gee, thanks  )

So I turned the power off, and tried to get it to post again, nothing. So I took the mobo (with the CPU, RAM still attached), grabbed my other ram sticks and PSU, and took the lot back to where I purchased it from for testing.

They put their own stick of RAM on, used my CPU, their own PSU, their own GPU, and ran it outside the case, and it posted first time, everytime. So I got them to put my RAM, GPU and use my PSU, posted first time, everytime (I was thinking WTF). So I put all the sticks of RAM on, posted everytime.

Bought the whole lot home, installing one piece at a time (back in the case), posting everytime. Long and behold, the whole system is back together now and didn't fault along the way of the install (i was testing it every new component that I added back to make sure it would still post, and then rebooted it about 4 or 5 times each test, just to be sure). The OS was installed back on, and the computer was running fine last night.

Go to start it up this morning, and I get a weird post beep - 1 long then 2 short - which according to the Gigabyte manual is Monitor of GPU error. So I try to restart it, same thing. I hit the power button on the case, and try and reboot, posts no problem, boots up the OS, and has been running flawlessly since.

Does this still sound like a mobo issue, or could the GPU legitimately be faulty? (I'm struggling to see how it could be the GPU, as it only does it very intermittently).

[Dark Mantis]

I would say that it points towards the motherboard.

Have you made sure that you haven't got the CPU heatsink screwed down too tight and distorting the board at all?

The other thing that is possible still is that the motherboard is shorting to the chassis, maybe by one of the securing screws.

[cjhod1]

I have reset the CPU cooler fan (and got them to do this when I was at CPL - where I bought the gear from).

The bit I don't get is why the system worked fine last night, then it does this when cold. It seems to me it is heat related, i.e. once the system is warm, its fine, otherwise it shorts/has a fault somewhere.

[Dark Mantis]

This is why I wondered if it could be a broken track or something shorting to earth.

[cjhod1]

This is why I wondered if it could be a broken track or something shorting to earth.

I will make sure nothing is fowling once the system has cooled down.

[Dark Mantis]

I had a motherboard a couple of weeks ago on the forum that was shorting to earth through the mounting screws. They are supposed to make an earth connection so there was obviously something wrong with the board. The only way to be sure is to take it out of the case and run it on some cardboard or something on the worktop.

[cjhod1]

I had a motherboard a couple of weeks ago on the forum that was shorting to earth through the mounting screws. They are supposed to make an earth connection so there was obviously something wrong with the board. The only way to be sure is to take it out of the case and run it on some cardboard or something on the worktop.

Yeh, the only problem being that the PC runs fine once its warm, so wouldn't the shorting issue exist the whole time (particularly if it was upsetting things that much).

[Dark Mantis]

Not necessarily as you said things expand when they get warm and it could be that as something is expanding it is breaking the connection to earth.

[cjhod1]

Update.

Took the complete PC back to CPL Notting Hill on Sunday afternoon after it had faulted previously that (wouldn't post, and was getting 1 long beep and 2 short beeps), and thank god it faulted for them whilst it was there. I left it with them to diagnose the fault, and I got a call the next day telling me the PSU 3v line was all over the shop.

They said they couldn't get it to post at all either, then changed to a new PSU, all fine. They tested the PSU and the voltage was dropping significantly on the 3v line, so they did a DOA on the PSU, gave me a new one, I tested it whilst at CPL, and I came home a happy boy.

2 days later and the computer has started everytime, even when cold (touch wood). So fingers crossed PSU appears to have been the issue.

Thanks to Dark Mantis for all the troubleshooting assistance.

Cheers (from one very happy customer who gets to enjoy his new system    )

[Dark Mantis]

I am happy for you that you finally got it sorted out as these things can be a pain in the a***!
Only to pleased to have been of assistance.