Another dead HDD within a couple months?

[Addikt]

Hey guys,

I have a few concerns about my current setup. I originally had a WD 74GB raptor installed as my system drive, however it died a couple of months ago and I'm in the process of RMAing the drive. It started with the drive being detected as having errors on it and wouldn't load the OS. Then the drive finally just died and wouldn't boot. Then I re-installed a WD Caviar Black 1TB as my system drive and I'm getting the same problem. It's started to fail to detect the drive when I restart sometimes, and I've been getting an odd message saying that my Over-voltage has failed. Could it be possible that the voltages supplied to these drives are too high and damaging the circuitry? Or could it just be a coincidence that I've had two drives die in the last couple months?

 

[Mr Fox]

Originally posted by: Addikt
Hey guys,

I have a few concerns about my current setup. I originally had a WD 74GB raptor installed as my system drive, however it died a couple of months ago and I'm in the process of RMAing the drive. It started with the drive being detected as having errors on it and wouldn't load the OS. Then the drive finally just died and wouldn't boot. Then I re-installed a WD Caviar Black 1TB as my system drive and I'm getting the same problem. It's started to fail to detect the drive when I restart sometimes, and I've been getting an odd message saying that my Over-voltage has failed. Could it be possible that the voltages supplied to these drives are too high and damaging the circuitry? Or could it just be a coincidence that I've had two drives die in the last couple months?

Excessive Hard Drive Failure is a direct Symptom of a crappy PSU...........

 

[Modelworks]

Get a volt meter and find out. Don't trust software readings. Any multimeter will work , usually can find one for around $15

 

[Addikt]

So I get a voltmeter and test for what exactly? Also, how do you test an SATA power connector? I have an OCZ ModStream 700W PSU which I RMAed about a year back because my GameXstream died on me. Is the PSU really to blame? Why are the other drives not affected only the system drive?

Also, before I had this PSU running 6 drives and now I only have 4 in the case with 2 1TB drives and 2 500GB drives. Before it seemed to power the 6 drives with no problems.

Is this power supply garbage? Am I going to end up screaming my head off at some OCZ rep because I lost a bunch of my data?

I think what's concerning me most is that it's only the system drive that seems to be affected, none of the other ones. Granted, the system drive is the one that is under constant use.

 

[RebateMonger]

Originally posted by: Addikt
Could it be possible that the voltages supplied to these drives are too high and damaging the circuitry? Or could it just be a coincidence that I've had two drives die in the last couple months?

Yes.

The easy way to check PSU voltages is with the latest digital power supply testers. They all seem to be made by the same company, with minor feature changes, and the "high-end" ones (which I'd recommend) cost around $20-$30. Unlike earlier analog testers that mostly used Red/Green lights, these will show both under- and over-voltages and have sockets for things like SATA power connectors.

Power Supply Testers

You really can't predict how long hard drives will last. There will be early failures and you certainly could hit a couple of those if you are "unlucky".

 

[Emulex]

vibrations can hurt drives too. like having a subwoofer pound next to the pc.

google the dude screaming at a raid set with iometer running and you can see how the drives slow down when he screams.

 

[Addikt]

I don't think it's vibration that's the problem, there isn't really anything around the case to cause that much vibration.

I do have one new thing to report though. I was checking my PSU to see if it was overheating and it felt pretty cool to the touch and was blowing out cool air. Then as I had my ear by the case I could hear my HDD make a cracking sound and then I could hear it spin down like you can hear a CD spin down in a CD-ROM drive. Sure enough, a few seconds later the system crashed and BSODed. Could the drive just be a lemon? Why would it just suddenly shut off. I think the PSU might in fact be the culprit.

I looked up the PSU testers and a lot of people have been saying that they are unreliable. I think I'll just splurge and try one, don't know how much help it's going to be though. The other option is just forget it and put in a new PSU altogether. To me it really sounds like the PSU is crapping out on me. This was the first BSOD but before that the computer would just randomly restart. Power surges I suppose?

 

[PandaBear]

If the HD is mounted to a case, then it is very hard to have enough vibration to kill it in 2 months. Your PS is most likely the cause of the problem. HD is usually designed to be good for up to 60C (that's how large OEM qualify them before using them in large volume).

If the voltage spike is enough to crash the HD's controller but not long enough to park the head on a ramp, or only enough to disturb the servo control, you can have a sudden jerk of the disk or arm movement. This can cause head crashes that scrape out a small section of the media.

If you can borrow a PS, or even buy one, and swap it out for a few months and see if the problem goes away, that will be good.

 

[RebateMonger]

You can also use Speedfan or other monitoring program to monitor the voltages (measured by the BIOS) to see if they're going out of spec. You can look at the variance of the voltages to see if they are abnormally distributed.

 

[Addikt]

Thanks for the replies guys. A friend of mine told me about that voltage monitoring software and from what I've heard on these forums in the past, the consensus is that these tools are inaccurate.

@PandaBear
That's what I was thinking as well, because the same thing happened to the previous hard drive. I didn't hear any of the sounds it was making but I would get random crashes and now you can't even format the drive, so perhaps the boot sector finally took a hit from the heads. So far I haven't found any evidence of data corruption, but I'd rather be safe than sorry. I think I'll end up calling OCZ and see what they can do, then if they don't help me out I'll just buy a new PSU and RMA this drive with the other one. Worst case, I'll end up with another PSU and can use the old one in the file server that I'm building.

Thanks to everyone who replied for the advice, sounds like a lot of us agree that the PSU is to blame.

 

[RebateMonger]

Originally posted by: Addikt
Thanks for the replies guys. A friend of mine told me about that voltage monitoring software and from what I've heard on these forums in the past, the consensus is that these tools are inaccurate.

If you are looking for variations in voltage, whether the BIOS voltage readings are "inaccurate" doesn't really matter. It won't make any difference if the absolute voltages are off by 0.5 volts or whatever. What you care about is if the measured voltages are swinging plus or minus three volts during the day when they should only be swinging 0.2 volts.

 

[Addikt]

Do you know how to monitor and log that information though? I can't watch my system every second of every day to see when it happens.

 

[RebateMonger]

Originally posted by: Addikt
Do you know how to monitor and log that information though? I can't watch my system every second of every day to see when it happens.

It records the signals in a log file and creates a graph for you of time versus voltages. You'd have to read the manual to set it up to give you what you want.

 

[Quiksilver]

Originally posted by: Addikt
So I get a voltmeter and test for what exactly? Also, how do you test an SATA power connector? I have an OCZ ModStream 700W PSU which I RMAed about a year back because my GameXstream died on me. Is the PSU really to blame? Why are the other drives not affected only the system drive?

How-to read power supply voltages.

We really don't know if the PSU is to blame until you've measured the voltages and posted them here.

As far as why the other drives unaffected, could just simply be because your other drives may not be as sensitive to voltage regulation issues.

Hell, it may not even be your PSU but rather some bad luck...

 

[Addikt]

Originally posted by: RebateMonger

Originally posted by: Addikt
Do you know how to monitor and log that information though? I can't watch my system every second of every day to see when it happens.

It records the signals in a log file and creates a graph for you of time versus voltages. You'd have to read the manual to set it up to give you what you want.

Do you know how to do this though? So far I have found the best tool to be Everest but I don't think it's possible. I've tried logging the information, but I don't get a graph instead it gives me separate HTML log files with a table of all the voltages. It makes it very hard to track fluctuations.

I've been searching google and the help files and I can't find a way to do what you've explained here.