Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations wOOdy-Soft on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Problem diagnosis: MB, CPU, other?

Status
Not open for further replies.

GEAK

Instructor
Feb 1, 2001
90
US
This is probably a pretty specific problem, not too sure whether I can get help (local techs appear to be clueless). I've got a dual Athlon MP 2000+ system riding on an Asus a7m266-d board, had it since it was a baby (about 4.5 years ago). About 8 mos back problems started: it would powerdown - no bsod, no warning, nothing - this happened within 4 mos of replacing DVD burner, within 2 mos of installing a second burner.

Techs I took it to said the 400W power supply was insufficient, I replaced it with 550W. Worked fine for a few weeks before the next unprompted shutdown. Shutdowns continued but less often. On occasional restart the bios wouldn't detect the HDs. After manual reset they'd be found - I suspected IDE controller but techs said HDs are the problem. After 3 mos of steadily increasing shutdowns it stopped booting - I could hear a chirping sound (from the drives?) during boot (definitely not beeps, I seriously doubt it was head crashing).

I replaced 1x40gig & 1x80gig with 1x160gig. Worked for less than a week. Then techs said video card (ATI AIW Radeon 7500) was problem. Replacing that with newer card (ATI 9200 SE) fixed it for another week. I've since put in a PCI IDE controller and it boot a half dozen times before problems came back. Last few times it was running I checked event log - nada. No indication of any problem.

For past 4 mos haven't been able to get it to boot. It'll post fine, detect all devices and then partway through boot sequence (loading XP Pro) it'll restart. Won't boot in safe mode, won't boot normal, won't boot.
I've thrown almost enough money at this machine that I could've bought another one.

Techs also indicated that they suspected problems were due to a power spike/power surge (I've had it on a UPS since new and UPS is fine). All the hardware that the techs said was faulty I've since tested in other systems and it all works fine. I won't take it back to the techs because (a) they seem as clueless as I am and (b) I've run out of money.

Unrelated issue is that I've been unemployed for ~8 mos except for part-time work. I'm working f/t now but for ~1/2 what I was earning before. I'd like to do video editing & dvd authoring for extra cash but my other systems (amd 3.3g laptop & 1g celeron) aren't up to the task.

Any suggestions (preferably inexpensive ones) would be greatly appreciated. I'd still rather get this thing up & running than have to shell out $3-4K for a new video workstation. I've seen the same board on ebay (10 in past few weeks) selling for 150-200USD (sometimes with CPUs at double the price). Can't afford that until my house sells (hopefully soon) and don't know whether that'll fix the prob anyway. But, the only thing thats still in the tower that isn't new is MB, CPU(x2), RAM (2x512) and NIC.

Suggestions?
 
Heat? Have you cleaned and put new paste on the CPU's? CPU fans slowing/dying? Do you have any way to measure temps? Like franklin says: memtest 86.
 
No I haven't checked memory. I hadn't really considered that as a possibility.

Yes I've checked the fans & CPU temps. The board reports temperatures & fan speeds and I regularly ran a prog that would display same. I replaced the original heatsinks & fans about 9 mos before the problems started cropping up. The newer fans ran slower than the originals but they were more efficient & quieter. I can't recall the actual numbers but I know the temps were low.

Incidentally, the prog (MBMonitor? I think) that I ran to monitor the temps was also set to automatically shut down the computer if the temp came within X degrees (adjustable) of critical. It would also write an error in the event log if shutdown was necessary. None were ever written so I'm assuming the problem wasn't due to heat. I also have 2 case fans to augment the cooling; one was original, the second was added later (before replacing the CPU fans). They're old but still in good working condition (as far as I can tell).

Something else I just remembered. The techs I took it to also suggested that it might be a heat problem. They claimed that they had better luck when it was warm. They suggested that I power up and go into bios setup and just let it sit there for 30 min, then exit and boot to Windows. I had no better luck trying their suggestion - for me warm or cool it was still unstable. I was never able to identify a pattern that I could tie to unprompted shut downs or successful boots.
 
GEAK
Are these new heatsink/fan units fitted the correct way around? recess in the base positioned over the raised cam box of the CPU socket. Please don't just dismiss this idea, I've seen so many over the years fitted incorrectly and even after technicians have sworn they were on the correct way.

Correct application of heatsink paste!

Some early motherboards struggled picking up an RPM signal from newer slower fan types (the original fans likely span at around 5,000rpm where as the replacements are likely just over half the speed) some older motherboards simply would shut down straight away or randomly with these slower coolers.

Lastly, again on slightly older motherboards, there was an international problem with a whole production run of capacitors, this was due to unstable electrolyte formulae and was responsible for leaking capacitors in all sorts of electrical equipment.
These bad capacitors effected many makes and models of motherboards (I personally replaced a Leadtek Nforce2 motherboard only last week with 10 bad capacitors)
This link shows for what to look for:


Martin





We like members to GIVE and not just TAKE.
Participate and help others.
 
You know motherboards do ware out. However, often it is a buildup of dust on the CPU Coolers or the Chipset Coolers. Try taking the CPU Coolers apart, cleaning them and putting them back together. Take the coolers off and completely clean the fins. Sometimes dust accumulates that you can not readily see unless you remove the CPU Fan and look underneath. Dust can be a killer of computers/servers. Replacing your furnace/air conditioner filter/s can help cut the dust down inside your house. If the computer case has a filter look at it too. Make sure the fans keep moving after the motherboard boots up. You will probably need to remove the case cover and run it for a while to check this.

RAM can go bad also.

If the motherboard has worn out problems may be intermittent. Some server programs require up to 4 Gigs of RAM so make sure you have enough. The power requirements are the most sever on bootup when all your drives boot up at the same time.

You may have problems with other devices like network adapters.

You sure the power supply you have is good? Did you test it?

If you do not like my post feel free to point out your opinion or my errors.
 
Here's a cheaper alternative. Another thing that often causes pc's to reboot frequently (and actually the other thing I would have tried besides the power supply first and I'm questioning why the tech's you took it to did not) is your CMOS battery. You can get one for about $3 at walmart or radio shack. The number you want is CR2032. Takes you not even five minutes to replace and if it doesn't work, hey, you're out a gallon of gas...hardy har :)
 
I had it apart again yesterday, this time specifically looking for bad caps (now that I know what to look for) and there didn't seem to be anything wrong with any of them. I didn't get in there with a magnifying glass but from what I could tell they were all pristine (no tell-tale bulges or electrolyte puddles).

I had another look at the fans & heat sinks, they too appeared clean and (mostly) dust free. If I had time and enough steel wool I would've pulled and cleaned them anyway - I'll leave that for the weekend.
.
.
.
"steel wool" HAH! No, I'm not (usually) *that* stupid.

I'll swap out the CMOS battery as suggested and see if I can get the RAM tested somewhere - I think it's safe to assume that I shouldn't test it on that computer and I'm pretty sure that the RAM in it can't be put into any of my other computers.

If that doesn't work then I may take it to another shop (a new one opened up in the neighborhood about a year ago). Maybe they'll have better luck diagnosing it.

Thanks again everybody.
 
I am having a problem with video I guess. I just installed a new hard drive and everything worked fine. I closed everything up turned it on for about the 25th time and I could see everything fine but it was green. Is this a video card problem or motherboard problem. I have been having problems with the motherboard starting up that is why I changed the hard drive. It is a msi k7n2g and I haven't gotten the raid drivers installed yet. Could this have something to do with it as it is onboard raid and also onboard video
 
Since this is an ongoing thread, it would be better to start a new one covering your problem TCS.
But yes, it sounds like either a video chip porblem or a memory problem, as memory is probably shared.

Ed Fair
Give the wrong symptoms, get the wrong solutions.
 
TCS...this could be your monitor as well. If you have a friend who could bring theirs over, try swapping out another monitor and see if this fixes it. If it does, then you'll know your monitor crapped out. If it doesn't fix it, I would think it's probably your video card and you should invest in a new one. These sorts of symptoms really don't sound like they'd have anything to do with your harddrive, and if they did, it's likely they'd have started right after you put the new one in.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top