Hard Drive Light Stays ON Continuously 1

[depawl]

Hello. Not sure if this is a hardware or software problem but here goes. This is an eMachines D4362, P4 3.0 GHz, 1 Ghz RAM, Windows XP Home SP2. A week or so ago I noticed that after the computer runs for an hour or so it starts locking up, and the hard drive light stays on continuously. Also after this occurs, it will not shut down (locks on "Windows is shutting down"). A manual restart corrects the problem, but it reoccurs after a period of time. What I have done so far:
Ran malware removal as per MajorGeeks Malware Forum. Found and removed a few items. All scans now come up clean. A Seagate diagnostics program indicated that the hard drive was bad, so I replaced it with a new one, the problem seems better (computer runs faster and doesn't lock up as much) but eventually the hard drive light stays on continuously, it slows down, and begins to lock up. I opened up the case, blew out some dust, removed the HS and CPU and reinstalled with thermal gel as per Arctic Silver instructions. Added a pci slot fan, ensured that all fans (have 3 now) are running ok. Monitored temperatures with Speedfan. CPU temperature normally runs in the mid 40's, occasionally may reach low to mid 50's.
When the hard drive light is on continuously I have looked at Task Manager. CPU usage is very low (0 to 2%). I have terminated all of the processes that I could, makes no difference. The only thing I can do is manually shut down (Using the power button).
At this point I am at a loss, other than to possibly reformat and reinstall the OS. I guess that would definitely rule out software issues.
Any advice would be most appreciated.
Thanks.

 

[Q6600]

First, have you truly ruled out software buy leaving the syetem in say, bios hardware management for a few hrs? (you may have but I didnt see that in the post)

 

[depawl]

No, actually I have not tried that but will and will post back.
Thanks.

 

[Lemon13]

sounds like a hardware malefunction, heat issue, i bet on mainboard but it could be psu/ram also.
good luck

 

[wahnula]

eMachines are notorious in this forum for bad motherboards and weak PSUs. Not saying that's what's happening here, just sticking in some more elements for consideration.

Also, some HDD lights will stay lit if they're wired backward, check this and see.

Tony

Users helping Users...

 

[G0AOZ]

And following on from Q6600's suggestion, how about running in Safe Mode and see if the same problem occurs...

Also remove any Internet connection, RJ45, USB modem, etc., in case some web activity is going on.

ROGER - G0AOZ.

 

[BadBigBen]

I've had similar issues, and what it turned out to be was that the HDD locked up (or lost connection) after which XP would go bonk and lock up everything, power off and letting it sit for a while would cure it...

problem is either with a bad SATA Connection, Controller, or cable... Switching out the controller helped for a while (3 months) and a switch of the cable only 3 weeks...

after reseating the PCI controller and placing it into another slot, all seems well (no lock ups for the past 3 months)...

to cut a long story short, check the cables (SATA, IDE, PWR molex, etc.), replace it, reseat the Controller (if it is an add on card) or place it into another PCI port...

Also, some HDD lights will stay lit if they're wired backward, check this and see.

your thinking FDD there Tony, if you can reverse the IDE cable most likely it will melt the cable and cause a fire (had this happen to me before with an IDE port with no keying)...

Ben

"If it works don't fix it! If it doesn't use a sledgehammer..."

 

[depawl]

Thanks to all who have replied. System appears to run ok in BIOS hardware management, but locks up in Safe Mode. At this point I'm leaning toward a bad motherboard, had some issues with the cpu fan connector awhile back. The only other thing I haven't tried was Ben's suggestion. I'll try a new IDE Cable today.
Thanks again
Dennis

 

[Freestone]

Check the motherboard capacitors for bulges or leaks.

http://en.wikipedia.org/wiki/Capacitor_plague

http://www.badcaps.net/

Run a memory test to rule that out.

http://www.memtest.org/

http://oca.microsoft.com/en/windiag.asp

 

[SimonDavies]

TBH if it had issues with safe mode but not BIOS I would be leaning away from motherboard and more towards hdd\controller, after all at the BIOS stage it's already running the system board.

One thing I would be tempted to do to rule out the board entirely would be to use a linux live dvd to boot to, this way the system loads up, uses the system resources and you get to test everything apart from the disk.

I would also check to see if the board supports S.M.A.R.T. disk monitoring, if so make sure it's enabled, you could simply find that the disk is on it's way out.

SimonD.

The real world is not about exam scores, it's about ability.

 

[bobbyforhire]

Check out Sandra from SiSoft. tell it to run a burn in test...it will let you know where your pc is crashing ( look at the test it's running when your pc crashes).

 

[depawl]

Update:
Further monitoring in BIOS indicated an error. Temps of cpu and system immediately shot to 112 degrees C and all voltages showed weird numbers.
Other info:
I have replaced HDD with a new one. Have also ran memtest - no errors shown. Have also tried a new PSU and new memory. Problem still persists. So at this point it appears to me to be a mother board issue, but I'll give the Sandra a try.
Thanks.

 

[wahnula]

Run another monitoring program like SpeedFan. Could be a problem with your sensors. Thanks for updating, we like that.

Tony

Users helping Users...

 

[smah]

I've seen something similar where the female terminal on the molex conector from the power supply to the hard drive was slightly stretched out. It would intermittently (heat related?) disconnect power to the hard drive & the system would lock up with the hard drive light steady on. If you have another power supply available, it would be worth testing for.

FWIW, in the situation described above, I carefully pried the terminal tightter (so that it wasn't so stretched out) and it's been fine for years. What I initially thought was a system with some majorly bad hardware, turned out to be just fine. It happens to be what I'm typing on right at the moment.

 

[depawl]

Forgot to mention that I have been running Speedfan. It works ok at startup or after reboot for a while, then at some point it loses connection with the sensors, and all temps read 0 degrees. One more reason I'm leaning toward a bad motherboard. Also, as posted above, I have tried a new power supply but it made no difference. thanks.