Recently the WD5000AAKS hard drive in my P4 HTPC has started having trouble spinning up or staying spun up. Basically, sometimes when I try and boot/resume it I can hear it spin up then down again continuously and obviously this causes it to be unable to complete booting and leads to BSODs, etc. I think it first happened when I added an IDE DVD, so I disconnected that and it stopped happening for a while but happened again when I tried to resume from standby today.
I've moved the HD to my PhenomII machine and it seems OK there, so I guess I've got a problem with either my PSU or motherboard. I've been using a EF28 DC-DC board with a Dell DA-2 brick for about a year now (with the brick directly powering the P4 socket) and not had any problems until recently but I guess I'll have to get hold of a normal PSU and see if that makes any difference.
What I'm wondering is if it's possible for a motherboard fault to be causing this strange behaviour or if it can only be the PSU.
Could HD not spinning up be motherboard problem?
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
-
dukla2000
- *Lifetime Patron*
- Posts: 1465
- Joined: Sun Mar 09, 2003 12:27 pm
- Location: Reading.England.EU
I'ld vote for the PSU. It could be the mobo if you have a delayed start parameter set in the BIOS and the BIOS has gone mad (or indecisive whether the drive should start or not). But highly unlikely.
If you disconnect the data cable, and leave the power cable plugged in, do you get the same symptoms? (Probably only booting to BIOS.) If so definitely the psu.
If you disconnect the data cable, and leave the power cable plugged in, do you get the same symptoms? (Probably only booting to BIOS.) If so definitely the psu.
Hi dukla2000
Just reporting back that I've fixed the problem. After following your suggestion and eliminating the motherboard, I realised I'd daisy-chained a couple of molex extensions to the one coming from the DC board so that I could connect both the HD and DVD. I guess I made it too long and caused voltage drop or something, as I've removed the extensions now and everything's working fine again
Just reporting back that I've fixed the problem. After following your suggestion and eliminating the motherboard, I realised I'd daisy-chained a couple of molex extensions to the one coming from the DC board so that I could connect both the HD and DVD. I guess I made it too long and caused voltage drop or something, as I've removed the extensions now and everything's working fine again
This problem has reared it's ugly head again unfortunately.
Both times it's happened, the HD has become corrupt and connecting it to my other PC and running the Western Digitals diagnostics, it's failed the Quick Test. After doing the Extended Test, which takes a few hours, the Quick Test passes OK.
Most of the files remain OK, but quite a few important system files on the C: partition were corrupted and so I had to restore XP from an image.
I've recently replaced my FX5600 graphics card with a 2600XT, which I think is meant to use less power, even though it has an external power connecter. I've not tried it for 3D gaming anyway, only used it for TV/Video with MediaPortal, so it shouldn't be drawing much. I did have the external power connecter going to the EF28 DC-DC board, but I've now rewired that straight to the DA-12 AC-DC brick. I didn't have any problems for about a week though, with it connected to the EF28.
The other thing I'd done shortly before the latest crash is put a PCI modem in and I also was using a LPT PIC programmer, which would only draw about 40ma. That didn't actually work, and on measuring the voltage on the LPT data pins, it was reading 0.81v when high, when it should be 3.3v or 5v, so something's dodgy there.
I've got about a month's warranty left on the EF28, so I really ought to find some way to properly test it so that I can get it replaced if it's dodgy. I've got about 6 months warranty left on the HD, but I'm dubious that there's anything wrong with that, as it always works fine after the problem occurs if I connect it to my other PC.
Or maybe the the motherboard's at fault and perhaps causing the EF28 to act up, resulting in the HD problems?
Both times it's happened, the HD has become corrupt and connecting it to my other PC and running the Western Digitals diagnostics, it's failed the Quick Test. After doing the Extended Test, which takes a few hours, the Quick Test passes OK.
Most of the files remain OK, but quite a few important system files on the C: partition were corrupted and so I had to restore XP from an image.
I've recently replaced my FX5600 graphics card with a 2600XT, which I think is meant to use less power, even though it has an external power connecter. I've not tried it for 3D gaming anyway, only used it for TV/Video with MediaPortal, so it shouldn't be drawing much. I did have the external power connecter going to the EF28 DC-DC board, but I've now rewired that straight to the DA-12 AC-DC brick. I didn't have any problems for about a week though, with it connected to the EF28.
The other thing I'd done shortly before the latest crash is put a PCI modem in and I also was using a LPT PIC programmer, which would only draw about 40ma. That didn't actually work, and on measuring the voltage on the LPT data pins, it was reading 0.81v when high, when it should be 3.3v or 5v, so something's dodgy there.
I've got about a month's warranty left on the EF28, so I really ought to find some way to properly test it so that I can get it replaced if it's dodgy. I've got about 6 months warranty left on the HD, but I'm dubious that there's anything wrong with that, as it always works fine after the problem occurs if I connect it to my other PC.
Or maybe the the motherboard's at fault and perhaps causing the EF28 to act up, resulting in the HD problems?
Well my Pentium4 PC's WD5000 was acting up again, so I got a replacement WD6400 sent as a warranty swap.
After cloning the WD5000 (which was working just about) to the WD6400 and also trying copying a working True Image to it, the WD6400 was failing to boot as well, with errors about missing files and BSOD before I even got to the XP boot screen. I then tried installing XP from scratch using my other PC, and that failed to format during the Text Mode setup, with an error about a faulty or damaged disk.
After using the WD Data Lifeguard utility to get the WD6400 to a state where I could wipe it with DBAN, I got another replacement WD6400. This worked for a day before also failing to boot, making clicking noises and giving me just a black screen for ages, before showing the white strip of the booting from hibernation screen, but no text and not progressing any further. It may have got a bit further had I left it for a few hours but clearly it wasn't working properly. I connected a new Samsung 500IJ 2.5" disk which I'd copied the XP install to earlier, and that worked fine. I then connected the WD6400 back and that still didn't work.
I then ran the WD Data Lifeguard (DOS) utility and the Quick Test failed with error code 0007. I then did an Extended Test and that got about 3/4 of the way through the drive before failing, saying "Too Many Errors: Code 0025".
I connected the WD6400 to my other PhenomII PC and after getting it to recognise it, I tried to access the partitions but this either failed completely or just chugged for ages and my PC became non-responsive. I'll try running the Extended Test again on the drive using my PhenomII PC tomorrow, but I'm not hopeful.
I think it unlikely that I've been sent two duff replacements, so I'm wondering if my Pentium4 machine is somehow killing drives. Is this possible? If so, I guess I'd better replace the motherboard and PSU as I've got no way of knowing which is responsible.
After cloning the WD5000 (which was working just about) to the WD6400 and also trying copying a working True Image to it, the WD6400 was failing to boot as well, with errors about missing files and BSOD before I even got to the XP boot screen. I then tried installing XP from scratch using my other PC, and that failed to format during the Text Mode setup, with an error about a faulty or damaged disk.
After using the WD Data Lifeguard utility to get the WD6400 to a state where I could wipe it with DBAN, I got another replacement WD6400. This worked for a day before also failing to boot, making clicking noises and giving me just a black screen for ages, before showing the white strip of the booting from hibernation screen, but no text and not progressing any further. It may have got a bit further had I left it for a few hours but clearly it wasn't working properly. I connected a new Samsung 500IJ 2.5" disk which I'd copied the XP install to earlier, and that worked fine. I then connected the WD6400 back and that still didn't work.
I then ran the WD Data Lifeguard (DOS) utility and the Quick Test failed with error code 0007. I then did an Extended Test and that got about 3/4 of the way through the drive before failing, saying "Too Many Errors: Code 0025".
I connected the WD6400 to my other PhenomII PC and after getting it to recognise it, I tried to access the partitions but this either failed completely or just chugged for ages and my PC became non-responsive. I'll try running the Extended Test again on the drive using my PhenomII PC tomorrow, but I'm not hopeful.
I think it unlikely that I've been sent two duff replacements, so I'm wondering if my Pentium4 machine is somehow killing drives. Is this possible? If so, I guess I'd better replace the motherboard and PSU as I've got no way of knowing which is responsible.
Well I ran WD Data Lifeguard (Windows) on the drive connected to my PhenomII PC and it got to sector 639481855 of 1250263728 before saying "Too many bad sectors detected".
It says the SMART data is OK, but I notice three attributes have got 1 in the Warranty Column. Does that indicate anything useful?
I should mention that I was running the PC for a few days with an old 10GB IDE HD for a few days without any problems before installing the second replacement WD6400, so I don't know if that helps rule out a motherboard/PSU problem.
It says the SMART data is OK, but I notice three attributes have got 1 in the Warranty Column. Does that indicate anything useful?
I should mention that I was running the PC for a few days with an old 10GB IDE HD for a few days without any problems before installing the second replacement WD6400, so I don't know if that helps rule out a motherboard/PSU problem.
Which one was the culprit in the end? The PSU ? or the mobo?
I think I have the same problem :
My friend here has a very strange HDD problem. He has a system based on the A8V Asus mainboard and had 2 HDDs (1TB SAMSUNG F3 and 750GB Seagate) that are connected to the Promise 378 SATA controller as the VIA Chipset's own SATA capabilities are not recognizing these 2 drives.
To avoid prolonging the boot sequence (with SATA BIOS and all), the OS is installed on a third IDE drive that is connected to the IDE port on the mainboard.
The drive was a WD 320GB AAKB and it worked really well for the last 2 years but ... around a week ago, when I've installed the 1 TB drive ... in a day or two ... the system started to manifest HDD problems like this :
1ST) The sound is wobbling and not playing right ...
2ND) Programs start moving very slow and sometimes end up in BOSD
3RD) While booting, windows reports file system problems and recovers various files
In the end, the 320GB IDE was found to be defective reporting Error 0007 in Western Digital's own diagnostic software.
OK .. so another HDD right ?
He bought a new 160GB IDE HDD .. not bigger because he had the SATA drives for information storage and this IDE drive was only for the OS.
Installed the drive ... everything was fine but in 2 days, the same sequence of events, as described above, appeared and the system is not booting anymore, the boot sequence ends up in BSOD.
Where's the problem?
What's killing the IDE HDDs in this system ?
Its pretty clear that the Psu just can't deliver sufficient power. Maybe it was near the limit already earlier but its Max power has decreased due to age and makes other components fail. I'd say forget DC/DC and just get a regular Psu unless you actually prefer fixing problems to actually using the system for its intended purpose.
I think that was directed at me rather then East17.lm wrote:Its pretty clear that the Psu just can't deliver sufficient power. Maybe it was near the limit already earlier but its Max power has decreased due to age and makes other components fail. I'd say forget DC/DC and just get a regular Psu unless you actually prefer fixing problems to actually using the system for its intended purpose.
You're probably right, although the DC board is only powering the motherboard (and the PCI tuner card and the AGP slot power of the 2600XT) and the single HD, with the PCI-E power connector and the P4 power connecter wired straight to the Dell brick, so you'd think it would be able to cope.
Anyway, I guess I'll have to drop £60 on a Nexus Value 430 (I've already got one of those in my new PC and they're so quiet they negate the point in using a DC-DC board) and hope that fixes the problem and another HD doesn't get screwed up by some motherboard problem.
I connected the current WD6400 to my new PC and it won't even run the WD Quick or Extended tests now as it say's there's too many SMART errors (ECC corrected -50 and relocated sect -75) although I was able to do both a Quick and Full erase, which seems strange as that suggests it's able to write to the entire drive (although I guess it doesn't try to write to mapped bad blocks). I then tried creating 2 25GB primary partitions and two Logical partitions in an Extended area (393GB and 153GB) and the primary partitions formatted OK but the 393GB partition got stuck about 38% in (no error, just stopped progressing).
I'll need to replace the motherboard at some point as the current one doesn't even have HDMI and so can't support TruHD, but I haven't even got an AV receiver yet and money's a bit tight at the moment, so if I can just replace the PSU for now that would be good.
I might ask my Dad to have a look at the DC board some time as he's quite good with electronics and might be able to find and fix the fault and then maybe I can sell it. I've noticed some coil whine although I'm not sure if that's coming from the DC-DC board or the motherboard, but if it's the former the Nexus Value will actually be quieter as it doesn't have any of that.
Well I ended up buying a new motherboard and AMD II X4 630 processor.
I tested the DC board with it, but it still had problems so you were quite right about where the problem was. I'm temporarily using a (rather noisy) 450W PSU until I can get a decent quiet one.
I did buy a Nexus Value 430W but it had a ticking fan so I returned it. Actually, after using the DC board, I couldn't deal with all the cables coming out of the Nexus, so I'm going to have to get a modular PSU now
I tested the DC board with it, but it still had problems so you were quite right about where the problem was. I'm temporarily using a (rather noisy) 450W PSU until I can get a decent quiet one.
I did buy a Nexus Value 430W but it had a ticking fan so I returned it. Actually, after using the DC board, I couldn't deal with all the cables coming out of the Nexus, so I'm going to have to get a modular PSU now