Ok OK OK, I may have a wimpy power supply

[Lemon law]

To make a long story short, my wife bought a computer system that was rock steady
for 1.5 years. Suddenly it became squirrely until the OS became un bootable.

I could repair the OS using the original windows installs disks, but chkdsk kept showing new errors
after new errors as the OS lost the capacity to boot to safe mode. A test of the hard drive showed it was toast. So I bought a new HDD drive, reinstalled the OS and programs, and now its rock stable again.

But question, is this problem, at its root, a power supply question? And more importantly how to tell.

The power supply is pretty wimpy, only 15 amps on the 12 volt rail. But then again, my wife is not powering much. One HDD drive, a floppy drive, one DVD/RW
drive, and a dual core cpu and motherboard. No add in video card or case lights.

But on the more disturbing side, I have voltage monitoring software double checked as accurate using a VOM on the molex connectors. That bottom line is that the voltages are all south of specified voltages and have been since day one.
But not so far south of specified to trigger a +/- 5% Ansi values yet. Although usual 12 volt rail reported voltages are about 11.6 V, on occasion it can fall to 11.3 Volts.

So question, can that be the root cause of a premature HDD failure?

And even if I know a better power supply might be needed, I need something slim line in the vertical dimension, there is little clearance between the power supply and the CPU as it is. Anything thicker will not fit.

 

[nickbits]

How did the HD fail? But I doubt the ps caused the failure.

 

[Lemon law]

Originally posted by: nickbits
How did the HD fail? But I doubt the ps caused the failure.

----------------------------------------------------------------------------------
Lots of read write errors that gradually made the drive unbootable. It could be repaired
by using the original windows install disk, but still the repair did not last while the ability
go into safe mode was totally lost. The other symptom was, after five or 10 minutes of
running windows, the computer would totally lock up, no mouse, no ctr alt delete, no
nothing. And the only way to shut down was to cut power with a hard reset.

Now that I have a new Hard drive, the problems have totally gone away after doing nothing else.

 

[deimos3428]

Hmm, good question.

Like any other motor, an HDD motor draws a fixed amount of power*. So if the voltage drops, the current goes up. If this causes the rated amps for the device to be exceeded the result is excess heat, which could damage the motor or shorten it's useful life. The degree to which any such damage occurs would depend on the severity and length of exposure, naturally.

As the concept of "excess" heat is relative to predefined tolerances/ratings, to some minor extent this damage is ongoing and simply included in what we think of as "normal" wear -- the far more likely cause of bad sectors.

---
*Assuming it's not idle and fully spun up. Yes, I'm greatly simplifying here.

 

[john3850]

Did the PC have Vista and a limited amount of ram.
If so the page file would be running most of the time.
Heavy use of a page file = short life of your hard drive.

 

[Lemon law]

To answer John3850, the computer ran Windows XP home and has a Gig of ram. And multiple runs of memtest86 failed to get a single error out of memory. Temperature monitoring of HDD temps show temps in the mid 30's C before and after failure.

Just seems that 1.5 years is early for a HDD failure. So I have to suspect somewhat low but still close to in spec's 12 voltages as a possible suspect.

In terms of computer usage, my wife is more the little old lady from Passadina type, almost 100% of the time, the cpu is idling.