HD Overheating?

[Quadriflax]

What I thought might be my video card causing a loss of signal may actually be my HD overheating, but I'm not sure. Here's the situation. Sometimes when I open files my screen goes black. It seems to be completely random. For example, this last time I was trying to open a PDF file I d/led and it just went black. I tried opening another one on my D:\ drive and it also went black. Same thing happened when I tried to open any other type of file (all of which I was able to open prior to this crash). Yet I could open PDF files and everything else just fine on my C:\ drive. When opening the PDF file from the D:\ drive, I got the error message "cannot find file "filename" (or one of its components). Make sure the path and filename are correct and that all required libraries are available."

Upon reboot everything is working fine again as it was up to this random glitch. I've reinstalled Windows and don't think that's the problem. But this latest error (which I've never gotten before when this has occured, and only got when trying to open a PDF file) leads me to believe my D:\ is overheating. Does this sound like it could be the case? I ran the Maxtor diag tools (well, the first two quick tests anyway) and it said it was fine). Nothing feels extraordinaraly hot, from the outside anyway. Not sure what to make of this, but this error on the D: (and lack of error on the C🙂 make me think it's a HD and not a video card issue.

Thanks for any advice. Once again, sorry for the haste. I'm quite busy and I really don't need this right now...

Bonus question: I have two HDs and a floppy drive taking up all my 3.5" bays. Are there HD fans that wouldn't require me to move my drive to a 5.25" bay or to remove anything? The fans I've seen all require that I mount the drive in a larger bay or blow air down/up on the drive from an empty bay (meaning the fan takes up a whole bay, which I can't do since mine are all filled). There's gotta be a solution for people who have all their bays filled, right?

 

[Link Hylia]

could always just aim a fan at the hard drives, and make sure both are seperated, to allow airflow.

for the second Q:

http://www.becooling.com/becooling/harddrivefans.htm

 

[Taelon]

"I ran the Maxtor diag tools (well, the first two quick tests anyway)..."

I would _really_ recommend that you run a FULL drive test, not just the quick one. That IMHO was the one step you made the mistake of omitting that'll give you a definite answer. It could be as harmless as having to let the drive's firmware map out bad blocks (every modern drive has spare blocks it uses to replace bad blocks, which it keeps in an internal table that can be updated by running diagnostic tools such as yours).

If you have used your drive(s) for quite a while and this has never happened before, it stands to reason that overheating is an improbable cause...

 

[Quadriflax]

Well, I didn't have time to do the full tests. I'll do those probably when I get home from class today. I'd never seen this problem until I installed my new video card. Which is why I originally thought that was the problem. But this latest discovery of the D: drive being the only one effected by this problem makes me think it's the drive.

I don't know if it would be bad sectors though, since in most cases I can open the files fine. Let's say I open file1.pdf on my d: drive. I close the file, then 2 minutes later go to open it again and the screen goes black. I can't open ANY files on the drive, file1.pdf or otherwise--they all make the screen go black. If I reboot, file1.pdf still opens fine, as does everything else. No changes were made to the files. I can't see any patterns forming other than it only seems to effect the slave drive. If the tests don't yield any problems, what then? Could it be a bad cable or something? But I would think that would also effect my C: drive since they're one the same one.

This drive is actually a replacement for a drive I know crapped out (two errors reported by the diag tool). I got it no earier than August 02 IIRC (I posted about that problem here too, so that would tell). Maxtor sent me the new one since it was still under warranty. I'll run the final two tests (not the low level format, though) and see what comes up. If anyone has any further suggestions as to what can be causing this, please let me know! Thanks!

(oh, and yes, I have updated AV files)

 

[Curtis]

Are you sure that the slave drive, d:\, is actually the slave drive? Most HDs these days still require you to set the jumpers on the back of the unit to slave. Typicially they come from the manufacturer set to master. A bad cable is also a possibility - do a swap with another just to be sure. Also make sure the master drive is set to master.

I don't think bad sectors have anything to do with this. If a sector became bad after the file was stored on the drive, the file would cease to work even if you copied it over. If a sector became bad before the file was placed on the drive, it would not be placed there.

The Master Boot Record may also be damaged on the D:\ drive. In DOS/WIN9x days, you could fix this with the command "fdisk /mbr". I'm not sure what the procedure is for newer Windows OSes. Can you go to the system information window and tell me what "type" the HDs are listed as? Usually they will be type 47 or something. If it's tyoe 00, then you have a problem with the MBR I think...

 

[Quadriflax]

Curtis: Yes, I'm sure the drive jumpers are set correctly on both drives. PowerMax tools verify the drive is set up correctly as well. In Device Manager they are both listed as Type47.

I performed the full disk scan from PowerMax, and it crashed the first time at LBA 35772160 of 78177792, if that means anything. But the second time it completed and said it was fine. This was a "factory recertification" test. Could the crash be a signal of a problem in itself?

There is one other test that remains (aside from low-level format) and that's the "burn-in-test" but I have no idea what it is. It says I can select from 1-60 cycles to perform and that each cycle will take 20-30 minutes. I'm not sure what this test does, nor how many cycles it would take to verify the drive is in working condition. If someone can explain it to me I would appreciate it.

I'll definately be checking out the cable. It might be a dud. I'm sure I have a spare one around here somewhere, but I'll have to find the damn thing first. It might be old, does that matter as far as speed is concerned? Or are all IDE cables pretty much the same?

I've been posting to newsgroups and contacting tech support, so I'm all over on this one. Mostly because I'm not sure what the real problem is. Someone called it a "black screen of death," but the article they passed on to me was about a specific Intel monitoring program, not anything I use.

I don't think I'll buy a fan quite yet, beacause as Taelon suggests it's highly unlikely the drive is overheating. I'm not running a 24/7 FTP or downloading huge files when it happens. I'm so damn confused as to what can be causing this. It could be almost anything at this point. Not finding anything wrong with the drive gives me mixed feelings. If it was the drive I could just get yet another replacement. But now I'm not at all sure what the hell is wrong with it.

Given all this, what would you experts say is more likely the problem? The video card or the HD? Anyone think my impending XP installation (including disk formats) will have an effect on this? I'm really looking to that to solve this problem. I just have to get through the term first so I have time to work with it.

Damn computers. Damn them straight to hell.

But bless all ye who art attepting to help me.

 

[Gear]

Not all IDE cables are the same. I don't know the exact specifications; but there are cables for the different transfer rates of devices (cdroms, hds,etc). The ones wich come with the motherboard and the hds usually work fine with them.

Also check if you are plugging the cable correctly, the master drive should go at the end plug and the slave at the mid plug (but I don't know if it affects performance at all, or its just for 'cable select').

Using the primary IDE channel for hds and secondary IDE channel for cdroms (= use different cables for each kind of device) helps, too.

I have no clue about the results of your tests; but haven't you considered the possibility of a virus attack?

 

[Gear]

Originally posted by Quadriflax@Jan. 21 2003, 7:03 pm

But bless all ye who art attepting to help me.

It is more interesting than reply to the endless "how to burn a mode1/mode2 saturn game" topics

 

[Taelon]

Dude.... you said this all started when you put in a new video card.

Has it ever occurred to you that the power supply may not be able to keep up with increased power requirements (AGP cards nowadays suck a lot of wattage) and that the HD could be the victim of that?

 

[Quadriflax]

Has it ever occurred to you that the power supply may not be able to keep up with increased power requirements (AGP cards nowadays suck a lot of wattage) and that the HD could be the victim of that?

Yep, but I have a new Antec 400W PSU, so I don't think that's the problem. It should be more than enough to power my system.

Dude.... you said this all started when you put in a new video card

Yes, it did, but that could just be a coincidence. I've been posting to NVIDIA newsgroups and no one has an answer there either, so I don't know. It very well may still be the video card, but the last error message I got made me think it might be the HD. It could be the damn NIC or the sound card at this point for all I know, but nothing points to that being the case.

Not all IDE cables are the same. I don't know the exact specifications; but there are cables for the different transfer rates of devices (cdroms, hds,etc). The ones wich come with the motherboard and the hds usually work fine with them.

I think the one I'm using is the one that came with my new motherboard. I should probably just buy a new one since I don't know what's what any more. Also, everything is hooked up right. I followed the diagram it came with, so it's all good. And yep, the HDs are on the primary and the CD drives are on the secondary.

I have no clue about the results of your tests; but haven't you considered the possibility of a virus attack?

I have NAV, with updated AV files, so I think I'm pretty safe there. I suppose it could be, but I don't d/l a whole lot of files from shady sources. I'm quite careful about it. But, if it is, the format I'm planning should hopefully fix it.

I've also noticed that usually when I scandisk I get an error that says Windows is not reporting the free disk space correctly. This happens with both drives quite frequently. It might just be the result of me having to reset all the damn time (or at least often enough), but could that be indicitive of a problem? Scandisk can fix it, but it's still an error.

I'll try to locate another IDE cable and see what happens. I might have to buy one to make sure it's a good one though. Thanks again.

 

[Taelon]

*brainstorming*

New power supply.... New videocard.... HD problem.... Windows burning and crashing on some occasions when opening files....

Possible electromagnetic interference?

Power cable from PSU to HD ok? +5V as well as +12V? No intermittent contact?

Possible bus errors? PCI conflicts between videocard <-> CPU <-> IDE during bus transfers? DMA setting having possible effect on this? PCI/AGP IRQ conflicts?

Is the Windows desktop completely stable when using full hardware acceleration supported by the videocard? Window blitting etc. OK? Possible *true* moment of freeze when opening PDF file: opening new window?

*end of brainstorm*

 

[Quadriflax]

Possible electromagnetic interference?

What could cause this? How would I be able to tell? I've never had any issues before, but I did get some new neighbors recently. Maybe they're up to something over there...

Power cable from PSU to HD ok? +5V as well as +12V? No intermittent contact?

+5V, +12V?! Intermittent contact? Remember, I'm a moron. Last time I opened the case I didn't see anything else touching the drive. Also, I did change the plug for power to make sure it wasn't that. I made sure it was plugged in all the way too.

Possible bus errors? PCI conflicts between videocard <-> CPU <-> IDE during bus transfers? DMA setting having possible effect on this? PCI/AGP IRQ conflicts?

See, this I'm not sure about either. I've suspected conflicts all along since everything was working okay up until the new video card. But there are no conflicts reported in Device Manager. I posted a question about potential IRQ conflicts here:

http://litespeedcomputers.com/sx....;t=7299

and got the impression that there really wasn't anything wrong with the way things are. But, to be safe, when I sit down to install XP I'll disable COM1 and my LPT1 since I don't use either. That'll free up two IRQs. I'll also leave only my video card in there on first boot, then add everything one card at a time to make sure they get assigned okay. I don't know anything about BUS errors though.

s the Windows desktop completely stable when using full hardware acceleration supported by the videocard? Window blitting etc. OK? Possible *true* moment of freeze when opening PDF file: opening new window?

Hmm, could be. But then why only when opening files on the D: drive? That adds a whole new complexity to this stupid problem. If things were universal I would blame the video card. Although last time my brother plugged in his camera through the USB port, the monitor went off. It also used to happen sometimes when I would turn on my USB printer. That's what made me think there was a conflict between the video card and the USB ports. And, as the other thread shows, there's two USB things on IRQ5 with the video card. I've posted that config around newsgroups and sent it to BFG Tech support (no reply yet), and no one gave me a definate "yep, there's the problem." So I still have no clue.

Another thing I've noticed is that Windows wants to scan the D: a lot more often when Windows fails to shutdown correctly. Usually it wanted to scan the C: then the D:, but now it starts with the D:. But, no errors reported by PowerMax, so it's not the drive. Or is it? Either way I don't think I can get an RMA from Maxtor without an error code. I still haven't done the burn-test, but only because I don't know what it is and I don't know how many cycles are necessary.

Oh, also, everything seems to work okay as far as 3D acceleration goes. RTCW works fine. I have gotten kicked back to the desktop before, but that's really rare. Though I don't play that often. Also, Medal of Honor worked fine, as did Spearhead, until the 2.11 patch. Then it started crashing and freezing my computer every time I tried to load a multiplayer game (but single player was okay, IIRC).

Still a very irritating problem. I'm really, really hoping that XP will solve it and not just crash or something on load. Thanks for continuing to help though!

 

[Taelon]

It does sound like disabling unneeded devices (COM, LPT) in the BIOS, followed by a complete reformat of ALL drives and fresh install of XP, is indeed the way to go.

One final thing you could try first, though, is to simply take the videocard out ... run your system with whatever video you had before ... see if D: still craps out .....

 

[Curtis]

I still think it's possible that your Master Boot Record is damaged. I don't know if this will help. It's a subpage from here.

 

[Taelon]

It's difficult to fathom how a damaged MBR could lead to intermittent errors of the sort he's experiencing...

If anything, it would simply render D: unusable/not recognized by Windows...

 

[Curtis]

Difficult to fathom indeed...but nothing else has worked yet. ???

 

[Xavier]

im sorry is pdf causing you system to crash or is this just an example ?

What size are the harddrives and what operating system are you using ?

What items in device manager or giving you errors ?

Sorry if youve answered any of these already .

 

[ExCyber]

I thought I had replied to this, but I must have hit preview or something...

To sum up the recommendations I thought I had already made

:

- Disable any "active protection" features of your antivirus software and see if the problem goes away. Some implementations of these features cause Bad Things to happen, especially if the antivirus program is older than the operating system it's running on.

- Run at least one full pass of Memtest86 to make sure your memory subsystem is generally working okay.

- Check your mainboard/system manufacturer's site to see if there is a BIOS update relevant to this problem (but don't update your BIOS unless there's some indication that it will fix the problem).

- Have windows use a generic VGA/VESA driver instead of a native chipset driver and see if the problem goes away. Misbehaving video drivers (drivers in general, but AGP adds a bit of an edge) can cause system lockups, especially if some other component in your system is only marginally working within spec.

Hope this helps.

 

[Quadriflax]

im sorry is pdf causing you system to crash or is this just an example ?

It's just an example. Every file is affected when it decides to do this, but only on the D: drive.

What size are the harddrives and what operating system are you using ?

Win98 (not SE). Western Digital 15 GB (C🙂 and Maxtor 40 GB (D🙂

What items in device manager or giving you errors ?

No errors reported in device manager.

Disable any "active protection" features of your antivirus software and see if the problem goes away.

I never had a problem with this before. It is a fairly old Norton AV though. I'll upgrade when I try XP.

Run at least one full pass of Memtest86 to make sure your memory subsystem is generally working okay.

I've used this before when I suspected a bad stick. It didn't really seem to do anything, at least not that I understood. I ran the test (which took forever) and it never reported anything. It just started over again when it was done. Maybe I did something wrong.

Check your mainboard/system manufacturer's site to see if there is a BIOS update relevant to this problem (but don't update your BIOS unless there's some indication that it will fix the problem).

There is an update available, but it doesn't say anything about AGP fixes. I know there's a misidentification issue with this board that it would fix (my 1700+ is reported as an 1800+). I was going to flash it, but if you say I shouldn't, then I probably shouldn't.

Have windows use a generic VGA/VESA driver instead of a native chipset driver and see if the problem goes away. Misbehaving video drivers (drivers in general, but AGP adds a bit of an edge) can cause system lockups, especially if some other component in your system is only marginally working within spec.

Eh? I'm not sure what this would mean, or what drivers I should try. I've only tried nVIDIA detonator drivers up until now. I'm not sure what you're telling me to do. (sorry... me slow)

I sorta have my hands in a bind. I want to try these things and let you all know what effect they have to help narrow down the problem, but I don't have the time to fiddle with it. I'm just trying to narrow it down so I have a list of things to try when I finally break from school in a month. Really appreciate the help so far, thanks!

 

[Curtis]

You're using Win98? If it is the MBR just go to DOS and type "fdisk /mbr" (no quotes). Don't do this from within Win98, and make sure anything important is saved on the other drive.

If there is nothing wrong, it won't do anything at all, but it may fix your problem. I thought you were using XP already.