Possible Failure - Hard Drive?
Dear Fellow Enthusiast, or Current Occupant
System:
P4C 2.6 @ 3.3, Asus P4C800-E Deluxe, 2X256 Corsair PC3500XMS, 2XWD Raptor in RAID0 on ICH5R (with OS), 1XWD 40G JB, LiteOn CD, LG404 DVD RW, Sapphire 9800XT, Enermax 465 ps
What Happened:
I had just burnt a music CD for my daughter on the DVD RW, popped it out and stuck it into the CD to check it worked. Played about 1 min of the first song, hit the FF button in media player, and the world went dark.
Lost signal to monitor. Hit reset button, nothing. Mobo light still lit, drives and fans spinning. Monitor still working fine when switched to other computer thru KVM.
Powered down P4 system at the power supply, sat and stewed in my own juices for 20 min thinking about how to go about diagnosing.
Turned power back on at PS, powered up and *bingo* machine fires up and drives the monitor BUT "click...click...click" for the first 5-7 minutes.
Ruh roh, keto sez, hard drive dying.
Went to WD's website and downloaded both the Windows and the bootable floppy versions of their diagnostic utility. Did both 'quick' and 'extended' tests on all 3 drives, zero errors found. Hmmm. Clicking had stopped by the time I started tests.
Possible Extenuating Circumstances:
Computer sits on the floor by an open window with the case side panel off. It's been VERY cold here - Asus' monitoring utility won't read below 0C, I've been seeing "N/A" a lot in the case/mobo temp box this week. We had a snowstorm overnight, it's conceivable that some snow blew in the window and into the case, though there is no visible evidence to that effect. Dry as a bone on the case floor.
Further, it's been 18 of the past 24 hours of hard usage on the machine, 2 reformats plus lots of benching etc etc. This doesn't figger to be a factor imo but there you have it. It hasn't been physically changed or moved.
Questions:
-Temps don't really fluctuate, at least not quickly but at below 0 temps might I have some condensation forming inside one (or more) of the HD's?
-I wasn't close enough to the case to isolate exactly where the clicking was coming from. Anything else in there that goes "click...click...click"? I don't think so but...
-Any other HD diagnostics I should be trying out? Just ran a quick ATTO and WHOOPS the IDE 40G WD has some very bad funky write numbers at 1.0, 2.0 and 4.0, all the rest, and the RAID partition, look fine. But the drive with the suspicious numbers a) is only storage b) wasn't being accessed when the failure occured c) hasn't been used much during the above mentioned heavy usage over the past 24 hrs.
-What else might cause the monitor signal to fail, then recover after a 20 minute wait?
Yours truly,
Flummoxed in Edmonton :type:
System:
P4C 2.6 @ 3.3, Asus P4C800-E Deluxe, 2X256 Corsair PC3500XMS, 2XWD Raptor in RAID0 on ICH5R (with OS), 1XWD 40G JB, LiteOn CD, LG404 DVD RW, Sapphire 9800XT, Enermax 465 ps
What Happened:
I had just burnt a music CD for my daughter on the DVD RW, popped it out and stuck it into the CD to check it worked. Played about 1 min of the first song, hit the FF button in media player, and the world went dark.
Lost signal to monitor. Hit reset button, nothing. Mobo light still lit, drives and fans spinning. Monitor still working fine when switched to other computer thru KVM.Powered down P4 system at the power supply, sat and stewed in my own juices for 20 min thinking about how to go about diagnosing.
Turned power back on at PS, powered up and *bingo* machine fires up and drives the monitor BUT "click...click...click" for the first 5-7 minutes. Ruh roh, keto sez, hard drive dying. Went to WD's website and downloaded both the Windows and the bootable floppy versions of their diagnostic utility. Did both 'quick' and 'extended' tests on all 3 drives, zero errors found. Hmmm. Clicking had stopped by the time I started tests.Possible Extenuating Circumstances:
Computer sits on the floor by an open window with the case side panel off. It's been VERY cold here - Asus' monitoring utility won't read below 0C, I've been seeing "N/A" a lot in the case/mobo temp box this week. We had a snowstorm overnight, it's conceivable that some snow blew in the window and into the case, though there is no visible evidence to that effect. Dry as a bone on the case floor.
Further, it's been 18 of the past 24 hours of hard usage on the machine, 2 reformats plus lots of benching etc etc. This doesn't figger to be a factor imo but there you have it. It hasn't been physically changed or moved.
Questions:
-Temps don't really fluctuate, at least not quickly but at below 0 temps might I have some condensation forming inside one (or more) of the HD's?
-I wasn't close enough to the case to isolate exactly where the clicking was coming from. Anything else in there that goes "click...click...click"? I don't think so but...
-Any other HD diagnostics I should be trying out? Just ran a quick ATTO and WHOOPS the IDE 40G WD has some very bad funky write numbers at 1.0, 2.0 and 4.0, all the rest, and the RAID partition, look fine. But the drive with the suspicious numbers a) is only storage b) wasn't being accessed when the failure occured c) hasn't been used much during the above mentioned heavy usage over the past 24 hrs.
-What else might cause the monitor signal to fail, then recover after a 20 minute wait?
Yours truly,
Flummoxed in Edmonton :type:
0
Comments
As far as I remember, HDDs are vaccuums inside. If that's right, there couldn't be condensation. However, if there's a bad seal, then it is possible that there is condensation. But this possibility is highly unlikely.
CD-ROMs can click, when they're on very low spinning RPMs they tend to click. This is also unlikely.
As for what to troubleshoot:
BIOS battery? Memory? PSU?
I'm pulling straws here.
A stream of air didnt blow around a loose cable or something into a fan to make it click? Is that possible?
Otherwise, say it gets above 0 C and drive breathes in some moisture. At much below 0 C, the moisture FREEZES (-14.2 C is 32 F or freezing). That is also one reason why in cold climate conditions folks are told to let their HDs shipped in sit out of box at room tmep for 24 hours, partly to let the drives get up to normal size (platters, arms, spindle rings, they are metal core or all metal in WDs for IDE drives up to and through UDMA 100) and partly so any droplets that froze in side can first unfreeze and second the drive can breathe and low humidity will drop the water vapor content down to something tolerable. The drives, since they DO have breather holes, do not much like over 80% humidity, non-condensing, for operations. In theory 90% non-condensing is upper limit, 80% my safe limit down here in FL.
Yes, you can get small ICE crystals with a drive frozen, and a temp drop too far can leave data and arms not in synchrony and data not found. Try minimum 5 C environment, yes, suggest 52-55 F minimum temp for drive when operatings, when shipping they can take a lot lower but need to be brought gradually up to temp before first use.
Same old rules still hold true, IDE and SCSI drives are not true vacuum pack inside, need to pressure and temp and humidity equalize within certain temp\humidity ranges to run right. MIGHT be a good idea to force XP to run a disk error check.... Looks like DATA and not media errors with those results.
John.
Thrax, good ideas on the PSU and even better, bios battery - not sure at what temps it would start to slow down but afaik batteries do function 'less well' at lower temps, forgot about that. The memory tho? You just like saying it
Marc, it definitely was an internal component, not a (for example) molex end flapping around. I'm leaning more and more towards the CD drive, wasn't aware they got noisy.
Quick metric lesson for John: 0C = 32F. That is to say, my case temp has read below freezing point for most of the past week. 100C = 212F boiling point. Further, -40C = -40F but that's the only crossover point relative to the 2 scales. Body temp 98.6F = 37C, "normal" room temp (keep in mind, the writer is in Canada) 68F = 20C. 50F = 10C, which in the spring is about when we start to see hardier people walking around without a coat or jacket on - it's quite warm after being acclimated to low low below freezing temps for several months.
Hadn't thought about platter shrinking due to low temps, excellent point. Going to go force an XP data scan now, thank you.
Let's look at this logically again. The "clicking" can be easily diagnosed by 1) CPS (Clicks per second) and 2) POC (pitch of clicks). First a fan where one of the blades is catching will rapidly increase the CPS as the computer first starts. Those clicks will either continue or diminish as the airflow "propellers" the fan blades away from the box spokes. Fan blade clicks also sound very "plasticy" compared to a hard drive click. They are very distintive from each other. (POC)
It could very well be that the hard drive is dying. I had a Maxtor that was doing the very same thing. HDD tests would not show errors but, when I phoned to RMA, the online tech confirmed that the hard drive was most likely going to die. The exact term he used escapes me at this moment but it was something that wasn't widely published but Maxtor was more than happy to replace the drive no questions asked.
The hard drive clicking was also intermittent so I would strongly suspect your hard drive.
and shovel out that snow.
Ok, I used following formula:
Fahrenheit to Celsius: C= 5/9*(F -32)
Celsius to Fahrenheit: F = C * 9/5 +32
Plug 45 F into that equation: 5/9*(45-32) = 5/9*13 = 7.2 C
So, to get the crossover point, we have 5/9(? - 32) = 0; 32 F (5/9 * 0=0)
John.
It wouldn't suprise me if you get an error code. My WD 100gb JB drive clicked on me once and then it just died... I ran the data lite tool and it coded out 210 which ment for me the drive was toast.
I think that as long as you know you have enough power to run all of the deveices in your case and you have good connections if you hear a click and your sys dies, I think you can pretty much look at the HD.
"g"