Can I call this a failed motherboard now?

[shamrin]

I really do hate these probable mobo failures, seems you can never identify them deductively unless you've got popped caps or something. Anyway...

This is an HP desktop running XP. The symptom is that it spontaneously reboots from Safe Mode and Normal Mode at boot time. No dumps.

• Tested RAM w/ Memtest - OK
• Swapped sticks, changed RAM sticks
• Ran Video RAM Test Boot disc - OK
• Switched from onboard graphics to a PCI-E card
• Tested HDD outside of computer using Seatools (could not use Seatools on host machine as keyboard and mouse would not respond)
• Pulled plug on optical drives
• Swapped PSU
• Boot to F4UBCD (XP environment) - No mouse cursor on screen, no keyboard worky
• Boot to Ubuntu - V. slow boot (10 mins) then very slow operation trying to open C: drive (gave up waiting)
• Boot to MRI-type test environment, runs for 1 hour then freezes when tech tries to change view on HDD test screen
• Boot to ERD environment - Fails whilst loading Windows (progress bars freeze)

I think we're down to the motherboard but it just won't fail in some satisfying way that I can ID. Any further suggestions or should I just put a fork in it?

 

[Mr.Mike]

Have you checked the BIOS settings? It's difficult to think about anything else without a little more information. What model HP?

 

[BDNTech]

Try this

assuming the hdd checked out ok change the cable for the hdd and reset the bios by taking out the cmos battery holding the power button in for 10secs and putting the battery back in.

 

[shamrin]

assuming the hdd checked out ok change the cable for the hdd and reset the bios by taking out the cmos battery holding the power button in for 10secs and putting the battery back in.

Good suggestions. I thought briefly of both of these but I couldn't figure out how they related to my problem. Anyway, have now tried them both, alas to no avail.

Subsequently I tried booting it to the installed Recovery Console (black screen) then to an XP boot disc (USB keyboard stops working in install environment, machine runs super-slow with PS/2 keyboard). It's the flipping motherboard.

 

[PCX]

Its really hard not to say its failing isn't it? How long did you test the memory for and what did you use to test the MOBO?

 

[shamrin]

Its really hard not to say its failing isn't it? How long did you test the memory for and what did you use to test the MOBO?

The only (practical) way I know of to identify a motherboard problem is by induction (test every other component and if they are all OK, it must be the motherboard, or as Sherlock Holmes said, "...when you have eliminated the impossible, whatever remains, however improbable, must be the truth..."). Alas, this method is probabilistic and you can never know for sure if you are right.

I only ran through a couple of iterations on the RAM test, but I switched to different RAM as well and caught the same issues.

 

[katz]

Did you test the CMOS batt. voltage? I've seen low voltage batt. cause flaky things...

 

[indyvirusguy]

I would say it is the motherboard.

 

[crashtech]

I guess the only thing left to do is yank it out of the case and test it on the bench, since the next step is to change it anyway. Once in a while the way it is mounted, or a foreign object caught under the board can cause a problem, although it's very unlikely.

 

[altrenda]

The only (practical) way I know of to identify a motherboard problem is by induction (test every other component and if they are all OK, it must be the motherboard, or as Sherlock Holmes said, "...when you have eliminated the impossible, whatever remains, however improbable, must be the truth..."). Alas, this method is probabilistic and you can never know for sure if you are right.

troubleshooting words to live by.

 

[Painless]

My question: is it worth the time and effort to carry on investigating. The cost of a new mobo or possibly even a completely new machine is probably less or at best similar to your cost.

The only reason I would do this amount of diagnosing / troubleshooting is if I had time on my hands and wanted to practice my diagnostic skills.

 

[PCX]

I guess the only thing left to do is yank it out of the case and test it on the bench, since the next step is to change it anyway. Once in a while the way it is mounted, or a foreign object caught under the board can cause a problem, although it's very unlikely.

Is funny you mention that. We actually had a customer come in with a custom built that they did themselves and the motherboard was shorting out. Come to find out they put different size standoffs in the case and in the wrong places. Needless to say, the board was warped and fried as a result . . . I can't imagine how it lasted for a month . . . .?

 

[Tepin]

I hate diagnosing failed motherboards without any failed tests or visible damage. I would still try a different HDD and SATA cable/s. I've worked on systems with similar behavior and had disks that passed diagnostics when they are offline, but just fail to perform in real world environments.

The downside about electrical problems as that almost any component can cause the same symptoms.

I agree that it looks like a failed board though. One overlooked symptom to the failed component is heat. Open up the case and touch things inside. Things with heatsinks on them are the most obvious things to start at. If you react to touching the Southbridge heatsink the same way you do to a soldering iron that's not normal.

 

[BigFrog_Computing]

Bad hard drive is my reaction to the OP. If Windows can't read it, and Linux can't read it, that's a pretty good sign.

What happens if you boot a Linux live CD without the HDD connected? Can you use USB, surf the web, etc, all normally?

 

[crashtech]

From the OP:

Tested HDD outside of computer using Seatools

I've not used Seatools in a long time, but a bad HD presumably would have exhibited some problems during this phase.

 

[Cambridge PC Support]

my question: Is it worth the time and effort to carry on investigating. The cost of a new mobo or possibly even a completely new machine is probably less or at best similar to your cost.

The only reason i would do this amount of diagnosing / troubleshooting is if i had time on my hands and wanted to practice my diagnostic skills.

+1



(.........................)

 

[Painless]

Thank you Kev.

John

 

[crashtech]

There is merit to the thought that it might be wasted time to go further, but I can also understand just wanting to know for the sheer satisfaction of having definitively solved the problem.

After a second look at the OP, I wonder if the temporary installation of a disk controller card might shed some light on things. I've had many instances of onboard IDE controllers failing, though I have not seen this as much with newer SATA setups.