Second Opinion on a Motherboard Failure

[MrUnknown]

So, as the title says, I am wanting to run this by everyone to see what you guys think.

So, I get a HP e9000 Desktop computer in saying it started bluescreening and crashing after being on for a while. Eventually it would get past the starting BIOS screen. Client knows a little about computers and decided to clean it out, they said this fixed it for a short period but it started doing it again.

From the description they gave I was thinking some sort of overheating even though they cleaned it out already.

When I got it, I turned it on and it booted to the Desktop just fine with no issues. Decided to stress test it a bit and run a full scan from Norton on it. I am not sure of the exact amount of time it took, but it did crash and was hung on the HP BIOS screen. Rebooting did the same thing.

Seems hardware related so I remove everything from the motherboard excluding RAM and CPU. First boot, says it can't find any devices to boot to, so I plug in a DVDROM drive and the computer hangs on a detecting devices screen. I unplug it and it says it can not find a boot device. I plug in the other drive, which is a Blu-Ray drive, and the system hangs when detecting devices. Unplug the drive and try again and the system hangs on trying to detect a bootable device, telling me to insert a bootable disk and press any key. Plug in the HDD and get the same thing.

I tried a huge combination of inserting RAM sticks in the motherboard thinking they may have messed up somehow and tried my known good RAM with the same result: Pluggin in any device into the SATA ports causes it to hang on bootup when trying to find a device to boot to. I tried changing the boot order to just the HDD and even resetting the CMOS with the same effect.

I do notice the CMOS battery has heat marks on it and it looks like the North Bridge, maybe onboard graphics (I think this motherboard has a nVidia chipset, the chip in question has the nvidia name on it but I don't know what its function is, or even if motherboards have northbridges and southbridges anymore (I remember something about them going away))

This is the motherboard in question:
http://media.photobucket.com/image/HP 503098/Digital_Anny/M2N78-LA Violet6/m2n78v6600C.jpg?o=

A Pegatron (ASUS) MZN78-LA

The chip in question is under the silver heatsink in the front. On this layout, the CMOS battery is behind the RAM slots, but on the computer I am working on it is to the left of the silver heatsink and the heatsink is right next next to the SATA ports like it is in the image.

I removed that heatsink and found some dried up thermal glue/pad/whatever nasty crap OEMs use. I tried replacing it with some Arctic Silver, but the heatsink puts no actual pressure on the chip and it didn't spread.

When the system is cold, it will boot fine. I am thinking if I purchase a chipset fan for that specific chip, it will resolve the issues. I can't really find a replacement motherboard for this computer either, except one for $400, which I am not going to recommend to them.

Anyone else have anything else I can try? Or do you think I am right in my thinking?

 

[anonymous Mac Tech]

I removed that heatsink and found some dried up thermal glue/pad/whatever nasty crap OEMs use. I tried replacing it with some Arctic Silver, but the heatsink puts no actual pressure on the chip and it didn't spread.

This seems like a problem? I'd say first thing is there is a heat transfer issue if nothing else. Still doesn't fully explain why it doesn't detect any boot devices but reliable heat transfer needs to be addressed before moving on. I'm assuming the caps all at least visually check out?

 

[parker.casey]

I'd have to say the issue is ... HP. It's working exactly as they design. Look at a dv6000 or one of their other laptops, they have the same issue after 2+ years normally. HP is 100% crap.

 

[MrUnknown]

This seems like a problem? I'd say first thing is there is a heat transfer issue if nothing else. Still doesn't fully explain why it doesn't detect any boot devices but reliable heat transfer needs to be addressed before moving on. I'm assuming the caps all at least visually check out?

Caps all look perfect, the machine is only a bit over a year old.

Tomorrow I am going to pick up a chipset fan with a thermal pad to replace the heatsink on it. Then continue to test it to see what happens when it gets hot.

I am assuming this chip controls the SATA ports, among other things, so if it gets hot then it will start having problems. Without booting, I can't see anything else fail. The computer will also sometimes fail at the HP BIOS logo screen, not even getting to the point of trying to detect devices.

Saying this chip is the problem is only a guess. Everything else on the motherboard looks fine and I have ruled out something else causing it. It is simply a guess based on it looks like that chip is overheating and it only starts failing when it is hot.

I'd have to say the issue is ... HP. It's working exactly as they design. Look at a dv6000 or one of their other laptops, they have the same issue after 2+ years normally. HP is 100% crap.

hehe, I agree. This is a desktop however and I haven't seen one of them in for the same exact issues yet.

 

[NETWizz]

You did a remarkably good job troubleshooting; I think.

Many technicians do not remove everything but CPU & heatsink, Power Supply, stick of RAM, monitor, & keyboard to get a POST.

I am also impressed that you checked for bulging capacitors, which is what I was going to suggest until you said you did it.



Looks like the motherboard is bad, but I don't know if you have a Power Supply tester, but that is the next step simply because it is quick and it would be bad to replace a good motherboard if the problem were with the power supply... it is verification that you have eliminated everything but the motherboard.

If you have a known-good power supply, you could temporarily attach that and verify the system still crashes.

 

[Cybjun]

The week before last I had two HP towers with similar issues both used the same motherboard, both had failed motherboards and both were first gen. vista machines. tested with a new power supply, ram, USB dvd drive, new hard drive new cables, after a bios reset and hangs on the hp splash screen 9 out of 10 boots, and when trying to do a test load on the one time it did work, it crashed midway through. my cost on the board was more then it would cost to build a good general use PC. I think it would be good to just call it and recommend replacing it.

 

[gikstar]

I suspect that you will have to replace the motherboard.

That motherboard looks pretty standard to me. I would expect that you shouldn't have a problem finding some sort of HP replacement board, doesn't have to be the exact one, but make sure you get the I/O shield with it.

Currently I have a Gateway FX541X with a bad motherboard... too expensive to replace with the same one. I found a board with similar characteristics on eBay (new) fairly cheap that will do the trick. I had purchased this system for next to nothing and without a hard drive, so I will be doing a fresh install anyway. I'll be using a Gateway motherboard with the Gateway COA key.
Should have no problem with it.

 

[MrUnknown]

Thank you all for the replies even to a fairly dead thread.

Well, I have guaranteed that it is the particular chip on the motherboard that I was talking about. I put a fan on it and was able to almost do a complete virus scan on the computer before it shut down on itself.

I also found a post online with someone having the same exact problem and they just inserted a SATA PCI-X card and used that instead of the on-board SATA chips. However, I don't know, or really feel that this is a permament fix and the card I need is pretty expensive (I should be able to get a motherboard for the price). I am going to recommend a motherboard replacement as soon as I can find one for the thing so I can give a price.