PC issue - any ideas?
#1
Guest
Posts: n/a
PC issue - any ideas?
My PC keeps hanging for no apparent reason. It can hang as soon as you power it on, doesn't even get to the bios screen and no power to keyboard at this point and no error beeps and no video display, just hear the disks spin up and thats it. Secondly if it does boot it can crash at any time, even if idle and the processor is not doing anything yet there are no entries in the event log. It hangs to the point of no lights on the keyboard so can't give it the three finger salute, only power it off.
I seem to have solved the issue of it crashing at powering on by updating the bios, although why this was all of a sudden needed I don't know, it has been fine for the last 3 years on the original bios version! Plus its a b!tch to find updated bios files for a product that Intel no longer wishes to support
However I can not fix the crashing at any time issue. Its completely random as well, i've been using it for a few hours now and its not crashed yet other times it can crash within 30 seconds or a couple of minutes of logging on or it could run fine for 8 hours and then lock up. I would rebuild it but i'm concerned its not software related due to no event logs and the total lock up nature of the crashes and think it could be a memory issue, although if it is its difficult to pinpoint as sometimes you can use the PC fine for hours. Alternatively it could be the power supply spiking/dropping voltages to cause a lock up?
Its running XP Pro SP2 on a P4 2.4 with 1.5gb memory on an Intel D845ept2 motherboard, 500 watt psu and a Radeon 9800 pro video card, one DVD rewriter, a CD rewriter, one 60gb IDE drive, one 40gb IDE drive plus two network cards, one of which is wireless and a USB2.0 card. It also has a usb mouse running off it, a usb card reader and a usb plextor video capture box as well as mobile phone/psp/digital camera and pda plugged in occasionally (hence having the 500 watt psu).
Any ideas on any sort of diagnostic tools for trying to determine hardware faults? Software wise its not a virus or spyware as i'm running my work's corporate anti virus solution on it and Microsoft's Windows Defender.
(p.s. i was desktop support for 6 years so have tried quite a few things in trying to fix this issue but no luck so far, however i'm not a hardware guru like a lot of people on here)
I seem to have solved the issue of it crashing at powering on by updating the bios, although why this was all of a sudden needed I don't know, it has been fine for the last 3 years on the original bios version! Plus its a b!tch to find updated bios files for a product that Intel no longer wishes to support
However I can not fix the crashing at any time issue. Its completely random as well, i've been using it for a few hours now and its not crashed yet other times it can crash within 30 seconds or a couple of minutes of logging on or it could run fine for 8 hours and then lock up. I would rebuild it but i'm concerned its not software related due to no event logs and the total lock up nature of the crashes and think it could be a memory issue, although if it is its difficult to pinpoint as sometimes you can use the PC fine for hours. Alternatively it could be the power supply spiking/dropping voltages to cause a lock up?
Its running XP Pro SP2 on a P4 2.4 with 1.5gb memory on an Intel D845ept2 motherboard, 500 watt psu and a Radeon 9800 pro video card, one DVD rewriter, a CD rewriter, one 60gb IDE drive, one 40gb IDE drive plus two network cards, one of which is wireless and a USB2.0 card. It also has a usb mouse running off it, a usb card reader and a usb plextor video capture box as well as mobile phone/psp/digital camera and pda plugged in occasionally (hence having the 500 watt psu).
Any ideas on any sort of diagnostic tools for trying to determine hardware faults? Software wise its not a virus or spyware as i'm running my work's corporate anti virus solution on it and Microsoft's Windows Defender.
(p.s. i was desktop support for 6 years so have tried quite a few things in trying to fix this issue but no luck so far, however i'm not a hardware guru like a lot of people on here)
Last edited by Bravo2zero_sps; 08 October 2006 at 10:28 PM.
#2
I suspect either your RAM, PSU or CPU have gone. Most likely the PSU. Is it a el-cheapo, no-name version, or a decent make? This is quite a common mistake, some people spend £1,000's on their PC and £20 on a shoddy PSU - madness!!
I would start by running several tests....
1 - Memtest.... http://www.memtest.org/#downiso
2 - SuperPi..... http://files.extremeoverclocking.com/file.php?f=36
3 - Prime95.... http://files.extremeoverclocking.com/file.php?f=103
Memtest will tell you without doubt if your memory is done. The other 2 are damn good for testing your CPU. Run a 1M run of SuperPi and then a Torture Test in Prime95. If you pass these, there's not a lot wrong with the CPU.
Then I would install and run a monitoring app, like either....
1 - PCWizard.... http://www.cpuid.com/pcwizard.php
2 - Everest.... http://www.majorgeeks.com/download4181.html
And use it to monitor your voltages. 10% above or below is fine, much outside that and you need a new PSU. Keep a close eye on +12v, +5v and +3.3v lines.
I would start by running several tests....
1 - Memtest.... http://www.memtest.org/#downiso
2 - SuperPi..... http://files.extremeoverclocking.com/file.php?f=36
3 - Prime95.... http://files.extremeoverclocking.com/file.php?f=103
Memtest will tell you without doubt if your memory is done. The other 2 are damn good for testing your CPU. Run a 1M run of SuperPi and then a Torture Test in Prime95. If you pass these, there's not a lot wrong with the CPU.
Then I would install and run a monitoring app, like either....
1 - PCWizard.... http://www.cpuid.com/pcwizard.php
2 - Everest.... http://www.majorgeeks.com/download4181.html
And use it to monitor your voltages. 10% above or below is fine, much outside that and you need a new PSU. Keep a close eye on +12v, +5v and +3.3v lines.
#3
Scooby Regular
Join Date: Apr 2004
Location: Cardiff
Posts: 1,928
Likes: 0
Received 0 Likes
on
0 Posts
#4
Guest
Posts: n/a
Thanks for the replies, and yes it is a £30 psu. Have active monitoring for the voltages and temps on the system board as it comes with it and there have been no alerts from them, in fact the fans are no where near stressed.
I am hedging my bets on memory as it hung in the bios last night so its definately not OS related. Added a 1gb Kingston module not so long ago, looks like i'm going to have to start testing with one module at a time to try and isolate which one it is.
I am hedging my bets on memory as it hung in the bios last night so its definately not OS related. Added a 1gb Kingston module not so long ago, looks like i'm going to have to start testing with one module at a time to try and isolate which one it is.
Last edited by Bravo2zero_sps; 09 October 2006 at 09:23 AM.
#5
Scooby Regular
iTrader: (1)
Join Date: Aug 2005
Location: Manchester ish
Posts: 18,547
Likes: 0
Received 0 Likes
on
0 Posts
try the graphics card, a work colleague has had very similar issues to yourself, also with a radeon 9800 pro. Sometimes it would hang at boot with nothing on the display (keyboard would light up and discs would spin up), othertimes it would boot up and could be stable for ages then just lock up or it could be minutes.
Luckily he had a nforce 2 (amd) msi mobo and it has a diagnostic connector that lights up (did actually say it was a floppy disc controller error though ), only found out it was the graphics card when we removed it and used the onboard vga
Luckily he had a nforce 2 (amd) msi mobo and it has a diagnostic connector that lights up (did actually say it was a floppy disc controller error though ), only found out it was the graphics card when we removed it and used the onboard vga
#6
Guest
Posts: n/a
Cheers guys, looks like i'll be needing a new motherboard Found two seperate issues. Firstly one of the memory banks is bad so it wasn't the module as when swapped it worked fine in the other bank. Problem is motherboard only has two memory banks so down to one now and down to 1gb memory. When a module is plugged into the 2nd bank it dies pretty much on every boot now. With the 2nd module empty it boots every time.
However the crashing didn't stop there. The second crashing appears to be the hard disks as when I run PCWizard and a benchmark on either hard disk it crashes. Every other component runs fine. As its on either disk again I am assuming this means its that controller on the motherboard. I have swapped out the ide cable as it was dodgey but still hangs on the disk benchmark.
Now i've got the problem of either living what appears to be infrequent crashes, seem to happen when the PC has been on a long time or get a new motherboard but haven't got a clue what to get making sure its compatible with what I currently have or how to swap it out.
Any ideas on disk controller checks? Going to swap the hard drives to the ide port the cd/dvd drives are on and see what happens...
However the crashing didn't stop there. The second crashing appears to be the hard disks as when I run PCWizard and a benchmark on either hard disk it crashes. Every other component runs fine. As its on either disk again I am assuming this means its that controller on the motherboard. I have swapped out the ide cable as it was dodgey but still hangs on the disk benchmark.
Now i've got the problem of either living what appears to be infrequent crashes, seem to happen when the PC has been on a long time or get a new motherboard but haven't got a clue what to get making sure its compatible with what I currently have or how to swap it out.
Any ideas on disk controller checks? Going to swap the hard drives to the ide port the cd/dvd drives are on and see what happens...
Trending Topics
#8
Guest
Posts: n/a
Thanks D16GER, unfortunately my theories were a little premature. Whatever the issue is its intermittent which is why I thought I had got to the bottom of it
Its just taken 2 hours to get it to boot again. Eventually coming back on after removing the module from bank 1 even though i'd been removing both slots, disks, pci cards, changing the video card, the lot.
Not sure where to go now as its virtually impossible to pinpoint when the issue keeps coming and going.
What's the order of components initialising when you power a pc on? At what point should USB devices power on? The USB mouse isn't even lighting up when the PC crashes at power on, but when it does power on the machine boots. So whatever initialises before USB power is where its crashing.
This is going to bug me now as its a friggin nightmare Don't want to spend money on a new power supply or an other component if its not confirmed its that first.
Its just taken 2 hours to get it to boot again. Eventually coming back on after removing the module from bank 1 even though i'd been removing both slots, disks, pci cards, changing the video card, the lot.
Not sure where to go now as its virtually impossible to pinpoint when the issue keeps coming and going.
What's the order of components initialising when you power a pc on? At what point should USB devices power on? The USB mouse isn't even lighting up when the PC crashes at power on, but when it does power on the machine boots. So whatever initialises before USB power is where its crashing.
This is going to bug me now as its a friggin nightmare Don't want to spend money on a new power supply or an other component if its not confirmed its that first.
#9
Scooby Regular
Join Date: Oct 2003
Location: Runway two seven right.
Posts: 6,652
Likes: 0
Received 0 Likes
on
0 Posts
Try a back to basics repair. Remove anything non essential to get the PC to boot. If that works, reinstall 1 thing at a time, and stop when it becomes unstable. Bear in mind you may have more than 1 faulty component, so dont chuck it all back in at once if you find something!
#10
Guest
Posts: n/a
Thanks, thats what I have been sort of doing for the last 2+ hours. Like I say video card, pci cards, dvd/cd rom drives, all been disconnected.
It definately seems to centre around the memory banks, its not one bank though as the way I got it to boot this time was to remove the module from bank 1 which was the one working earlier and put it in bank 2, the one that wasn't working earlier. Doesn't matter which module it is, either the 512kb or the 1gb work, just not at the same time like they did when I added the 1gb back in April.
Perhaps there is a good reason Intel no longer support the motherboard I have, because its sh!te
It definately seems to centre around the memory banks, its not one bank though as the way I got it to boot this time was to remove the module from bank 1 which was the one working earlier and put it in bank 2, the one that wasn't working earlier. Doesn't matter which module it is, either the 512kb or the 1gb work, just not at the same time like they did when I added the 1gb back in April.
Perhaps there is a good reason Intel no longer support the motherboard I have, because its sh!te
#11
Guest
Posts: n/a
The memory post is the first thing to initialise at power on isn't it? The thing is if it was a memory issue I would have expected beep codes but there aren't any, its just sitting there with no video display, no bios splash screen, just the disks spinning up. If I remove the memory completely then I get beep codes.
Oooh I hate computers
Oooh I hate computers
#13
Guest
Posts: n/a
Unfortuantely not. Well I have but its not powerfull enough to run everything, the original psu that came in it was something like 120 watt which is why I got a 500 watt one to replace it with.
I can't try it outside the case as I don't want to take it all apart to drop the motherboard out, and knowing my luck i'd kill the motherboard trying.
What's making you think there could be a short somewhere? The randomness of the fault?
I thought about trying to get a spare psu from work to test but I doubt they will be powerfull enough as they are HP Evo workstations with everything onboard so no need for a powerful psu. I will check tomorrow though to see what power they are.
I can't try it outside the case as I don't want to take it all apart to drop the motherboard out, and knowing my luck i'd kill the motherboard trying.
What's making you think there could be a short somewhere? The randomness of the fault?
I thought about trying to get a spare psu from work to test but I doubt they will be powerfull enough as they are HP Evo workstations with everything onboard so no need for a powerful psu. I will check tomorrow though to see what power they are.
#14
TBH wattage doesn't mean diddly squat. I have a 480w in my main and a 300w in my other system.
The 300w is almost as powerful as the 480w, due to the efficiency of it's design.
In fact I have seen 600w PSU's that don't pack the punch my 300w does.
Look for the amps it provides on the individual rails, that is the key. Make sure it is strong on the +12 and +5v rails.
You will probably find the old PSU was quite powerful, even though it has very low wattage.
I suggest you disconnect EVERYTHING bar the following....
1 stick of RAM
GFX card
1 HDD
1 Optical drive
...then see if you still get the crashes. If you do, try the old PSU, it should power that few components, even if for just long enough to diagnose the fault.
I really am leaning towards the PSU being at fault now. Especially given the randomness of the errors. Can you tell me the readout of the +12v, +5v and +3.3v lines?
The 300w is almost as powerful as the 480w, due to the efficiency of it's design.
In fact I have seen 600w PSU's that don't pack the punch my 300w does.
Look for the amps it provides on the individual rails, that is the key. Make sure it is strong on the +12 and +5v rails.
You will probably find the old PSU was quite powerful, even though it has very low wattage.
I suggest you disconnect EVERYTHING bar the following....
1 stick of RAM
GFX card
1 HDD
1 Optical drive
...then see if you still get the crashes. If you do, try the old PSU, it should power that few components, even if for just long enough to diagnose the fault.
I really am leaning towards the PSU being at fault now. Especially given the randomness of the errors. Can you tell me the readout of the +12v, +5v and +3.3v lines?
Last edited by D16GER; 10 October 2006 at 11:56 PM.
#16
Scooby Regular
iTrader: (1)
Join Date: Aug 2005
Location: Manchester ish
Posts: 18,547
Likes: 0
Received 0 Likes
on
0 Posts
If you arent even getting a display at initial boot then I would disconnect all power, turn the pc on (obviously it wont but it will discharge the capacitors) and then reset the cmos and try again.
#17
Guest
Posts: n/a
Will this be ok for a replacement psu for a P4 system?
450W X-power ATX PSU
At least I can take it back if its not the psu thats causing the issue.
450W X-power ATX PSU
At least I can take it back if its not the psu thats causing the issue.
#18
Does sound psu related, have seen this before, remove any hardware you dont need as it may be dragging the available current down, bit like in Apollo 13, the key here was you said hard disks, so at least two and probably a couple of optical drives, might just be too much for the PSU ?
Had it at work, machine kept failing at the same time each night, went through it numerous times, it seemed centered on the backups but nothing had changed, then we realised some extra disks had been added, the failure was due the backups kicking in, powering the tape drives up at the same time as all the hard disks, this tipped the psu over the edge and the thing conked and then rebooted, all scripted so it came back up.
We realised and added extra power supply capacity and it was fine after that.
Turn off anything you aren't using in the bios, like floppy controllers, serial and parallel ports, onboard sound, onboard vga - if present.
Just run with your boot drive.
Is it cooling properly, it's a P4 it will slow down before it crashes (Athlons just bun if too hot), so does it get sluggish before it stops ?
It sounds like there is a heat issue there, is there thermal paste on the cpu/cooler ?
try a basic graphics card
Just get a new mb, they arent that dear nowadays, Ebay may turn something up, I have just retired an Athlon Machine to the loft with an Epox 8KHA board that has been on for 5 years without a single problem that wasn't virus related. My 2.8 P4 doesnt seem any quicker than the AMD and I think thats partially down to the no name motherboard in the Intel machine, Intel may well do reference designs but the overclockers always use stuff like Abit, Asus etc, basically anything that is built to get overclocked will be over engineered for use at standard clock speeds.
Had it at work, machine kept failing at the same time each night, went through it numerous times, it seemed centered on the backups but nothing had changed, then we realised some extra disks had been added, the failure was due the backups kicking in, powering the tape drives up at the same time as all the hard disks, this tipped the psu over the edge and the thing conked and then rebooted, all scripted so it came back up.
We realised and added extra power supply capacity and it was fine after that.
Turn off anything you aren't using in the bios, like floppy controllers, serial and parallel ports, onboard sound, onboard vga - if present.
Just run with your boot drive.
Is it cooling properly, it's a P4 it will slow down before it crashes (Athlons just bun if too hot), so does it get sluggish before it stops ?
It sounds like there is a heat issue there, is there thermal paste on the cpu/cooler ?
try a basic graphics card
Just get a new mb, they arent that dear nowadays, Ebay may turn something up, I have just retired an Athlon Machine to the loft with an Epox 8KHA board that has been on for 5 years without a single problem that wasn't virus related. My 2.8 P4 doesnt seem any quicker than the AMD and I think thats partially down to the no name motherboard in the Intel machine, Intel may well do reference designs but the overclockers always use stuff like Abit, Asus etc, basically anything that is built to get overclocked will be over engineered for use at standard clock speeds.
#19
Guest
Posts: n/a
Thanks for the reply J4CKO. However I have tried an older video card that didn't need its own power supply, disconnected all drives but the bootable drive etc and it still did it. There is no heat issue, the onboard sensors report everything is running about 38C and the fans aren't anywhere near stressed/capacity and could step up significantly if needed. Also there is no slow down before the freeze, everythign is running fine then all of a sudden the whole systems freezes and has to be powered off, no repsonse from anything apart from the power button.
I still can't get my head round how swapping the memory from one bank to another and then having to do it back again fixed the issue each time it failed to boot at power on.
Its all working fine at the moment but if I power it off and put the 2nd memory module in the whole process will start all over again although it will still freeze if I run a benchmark from PCWizard on either hard drive, all other components return their results fine.
I still can't get my head round how swapping the memory from one bank to another and then having to do it back again fixed the issue each time it failed to boot at power on.
Its all working fine at the moment but if I power it off and put the 2nd memory module in the whole process will start all over again although it will still freeze if I run a benchmark from PCWizard on either hard drive, all other components return their results fine.
#20
Guest
Posts: n/a
OK swapped the power supplies over, turns out my old one is 300 watt so might be OK. Its a chieftec and the one I took out is a Magna 500 watt. Which make is better?
Here is the voltages for the Chieftech, I take it the tiny differences in voltages are irrelevant?
Here is the voltages for the Chieftech, I take it the tiny differences in voltages are irrelevant?
#21
The Chieftec I would imagine should be better. Look at the labels on the side of both. You should see ratings for +12v, +5v and +3.3v. These will be in amps, eg: +12v - 18a
Whichever has more amps will obviously provide more power.
Also look at the total power available. Should be on the same label.
Oh and yes, those small changes are irrelevant.
Just out of interest, can you not see the +12v readout? If not, can you run Everest and check it? It is quite important.
EDIT: - almost forgot, a quick test of a quality PSU is weight. If it's really light and tinny chances are it's crap.
Whichever has more amps will obviously provide more power.
Also look at the total power available. Should be on the same label.
Oh and yes, those small changes are irrelevant.
Just out of interest, can you not see the +12v readout? If not, can you run Everest and check it? It is quite important.
EDIT: - almost forgot, a quick test of a quality PSU is weight. If it's really light and tinny chances are it's crap.
Last edited by D16GER; 12 October 2006 at 10:13 PM.
#22
Guest
Posts: n/a
Thanks D16GER, the Magna although the bigger PSU feels lighter and tinny where as the Chieftec feels more solid.
The onboard voltage and temp monitor doesn't report for 12v for some reason, those pics above are all it provides.
Also running Everest and looking under the Sensor category there is no readout for 12v. Why is there no 12v readout when it is a 12v ATX power supply?
Oh and since putting the Chieftec power supply back in the PC is running both memory modules and seems to be OK. The first time I switched it on with the Chieftec power supply in it crashed before the BIOS so not 100% convinced its fixed but since then it hasn't frozen up
The onboard voltage and temp monitor doesn't report for 12v for some reason, those pics above are all it provides.
Also running Everest and looking under the Sensor category there is no readout for 12v. Why is there no 12v readout when it is a 12v ATX power supply?
Oh and since putting the Chieftec power supply back in the PC is running both memory modules and seems to be OK. The first time I switched it on with the Chieftec power supply in it crashed before the BIOS so not 100% convinced its fixed but since then it hasn't frozen up
#23
Don't worry about it not reporting the +12v. That's just the way the sensors are installed on the motherboard. Some support monitoring the +12, some don't. My AV8 board doesn't report it either.
The rest of your voltages seem to be fine. Only thing would slightly concern me is the Maxtor HDD temp. 43c seems quite high. Although, given that most HDD's max temp is around 60c, it should be ok.
Glad to hear all is running well atm. Hope it stays that way.
The rest of your voltages seem to be fine. Only thing would slightly concern me is the Maxtor HDD temp. 43c seems quite high. Although, given that most HDD's max temp is around 60c, it should be ok.
Glad to hear all is running well atm. Hope it stays that way.
#24
Guest
Posts: n/a
Thanks for all your help I was wondering about the HD temp, especially as the system will still freeze if I run a benchmark on the HD's. They run fine under windows with no slow down with read/write access etc so don't think they are on their way out. I am hoping its more a case of Windows needing to be reinstalled and not hardware issues.
If the crashing/freezing up on booting returns then I am taking a guess that its going to be the motherboard thats on its way out as both BIOS has been updated and power supply replaced and also both memory modules checked they are OK.
If the crashing/freezing up on booting returns then I am taking a guess that its going to be the motherboard thats on its way out as both BIOS has been updated and power supply replaced and also both memory modules checked they are OK.
#25
You could also try giving the memory a small increase in voltage. Fo example, if it is set at 1.6v, raise it to 1.7, that may help stabilise it.
Also, with regard to the HDD's, have you ran the manufacturer's utilities yet? They will tell you the health status of the drives.
Also, with regard to the HDD's, have you ran the manufacturer's utilities yet? They will tell you the health status of the drives.
#26
Guest
Posts: n/a
Not yet as it took me until late last night getting it back up and running by finding my old PSU and putting it back in and testing. I will run the utilities tonight to see what state they are in.
How would I go about safely increasing the voltage to the memory?
How would I go about safely increasing the voltage to the memory?
#27
In the bios, you may or may not have the option, depending on the motherboard. If you have it, it's simply a case of raising it by 0.1v.
If it's not there, then possibly it can be adjusted by a dipswitch on the motherboard itself, but I doubt it.
What mobo is it? I can have a look for you.
EDIT - see it on the Everest screen above, will look now
If it's not there, then possibly it can be adjusted by a dipswitch on the motherboard itself, but I doubt it.
What mobo is it? I can have a look for you.
EDIT - see it on the Everest screen above, will look now
#29
Guest
Posts: n/a
Yep sounds about right, its **** but I didn't spec the PC when I bought it any further than CPU, minimum number of pci slots and midi tower case. I then added what I wanted but never factored in the motherboard functionality.
Thanks again for your help and will run those hard drive tests tonight.
Thanks again for your help and will run those hard drive tests tonight.