Roba122
asked on
Built new computer has problem need input
I have just built a new com[puter for my girlfriends mother. I built it with a all new components. It ran fine for the first few days and now its locking up and freezing. Its completely random. Its just recently started to re-booot itself too, theyre telling me, though I have only seen the freezing. Mouse, Keyboard, and Ctrl+Alt+Del dont respond have to push the power button to restart. So here is what I've done to trouble shoot it. I've updated the entire thing all the drivers OS the whole thing all the software is certified to work with windows except Nero Rom that hasnt even been used yet (its not using the whql drivers). I ran Memtest on it no errors. I used everest software from lavalys to check out everything and it all looks good.
configuration:
DVD Rom
CD-RW
80 GB HD
256 MB Ram
3.5 floppy
Motherboard/ W/ On board audio
ATI Rage 63 mb Card AGP
AMD Sempron 2800 Chip
*****235 Watt PS****
I realize the PS is a little small other than that I can only think that there is a motherboard problem. So does anyone have an opinion I think that the PS is too small but am unsure. Other than that the motherboard is the only thing I can think. The case is well ventilated though I could probably do to add another fan. The bios doesnt have any temp sensing and I cant figure out how I can find that so any input there would be helpful.
bottom line I ruled out memory and Video card
Drivers are good no BSOD or anything
The only thing that jumps out at me is the PS
If its not that than I think the MB
I'm adding a new fan though I dont think heat was a problem
So any input? I need to get this resolved soon cause everytime i go over there "so any idea about the PC"
configuration:
DVD Rom
CD-RW
80 GB HD
256 MB Ram
3.5 floppy
Motherboard/ W/ On board audio
ATI Rage 63 mb Card AGP
AMD Sempron 2800 Chip
*****235 Watt PS****
I realize the PS is a little small other than that I can only think that there is a motherboard problem. So does anyone have an opinion I think that the PS is too small but am unsure. Other than that the motherboard is the only thing I can think. The case is well ventilated though I could probably do to add another fan. The bios doesnt have any temp sensing and I cant figure out how I can find that so any input there would be helpful.
bottom line I ruled out memory and Video card
Drivers are good no BSOD or anything
The only thing that jumps out at me is the PS
If its not that than I think the MB
I'm adding a new fan though I dont think heat was a problem
So any input? I need to get this resolved soon cause everytime i go over there "so any idea about the PC"
One other thing to mention, Memtest is a good utility, but not always perfect. The only way to confirm 100% for sure, is to put RAM in from a know good system, and see if that makes a difference.
It could be a number of different problems:
- overheating cpu
- overheating video card
- faulty PCI card (especially network cards)
- bad RAM
- faulty power or inadequate power supply
- faulty motherboard
You should check the temperatures by putting your hand on the heatsinks and test if they are hot, since you can't detect it with Everest. You may not have seated the cpu heatsink flush with the cpu and it is overheating, in which case you need to adjust it. You also need a small amount of thermal paste (a small drop) between surfaces to improve thermal conductivity.
Remove all PCI cards to eliminate them as a source.
I know you ran memtest, but there's a small chance the RAM is faulty and still passed the test - only swapping in RAM that you know works is a sure method.
The power supply is too small and could be causing the problem, though I would think it would have shown up at the very start when you built the system.
When all else has been eliminated, the motherboard is left as a suspect and should be replaced, since it is new and could be faulty out of the box.
- overheating cpu
- overheating video card
- faulty PCI card (especially network cards)
- bad RAM
- faulty power or inadequate power supply
- faulty motherboard
You should check the temperatures by putting your hand on the heatsinks and test if they are hot, since you can't detect it with Everest. You may not have seated the cpu heatsink flush with the cpu and it is overheating, in which case you need to adjust it. You also need a small amount of thermal paste (a small drop) between surfaces to improve thermal conductivity.
Remove all PCI cards to eliminate them as a source.
I know you ran memtest, but there's a small chance the RAM is faulty and still passed the test - only swapping in RAM that you know works is a sure method.
The power supply is too small and could be causing the problem, though I would think it would have shown up at the very start when you built the system.
When all else has been eliminated, the motherboard is left as a suspect and should be replaced, since it is new and could be faulty out of the box.
make sure the CPU and the video card is seated properly. To test the power supply theory disconnect everything but the esentials, i.e.: motherboard, video and hard drive. If you still have the problem, more than likely it is not the PSU. Heat could be the problem, AMD is known for hot CPUs. Adding a fan to the case will not solve that problem. Make sure the CPU has a good fan and that it is properly installed.
ASKER
I put everything together with thermal paste and it is defientely seated well I put my hand on the heatsink and it is not hot at all and that had been after it had been running for a while. Also there is big honkin fan over the heatsink, Its loud enough theyre complaining about the noise. I've gone through 3 video card changes and no difference and I have no PCI cards the only card I have is video and its in the agp slot. When I built it I ran 3 different sticks of ram with no difference and before I did that and was still waiting for parts I put the ram that was in this computer in a different one to trouble shoot it. I'm almost 100% its not the ram between that and the mem test. So I'm pretty much down to PS and MB. So what you guys still think?
Running a power supply at, or near its maximum rating will cause it to fail or reduce its efficiency. In your case, the weak power supply might have been running near its capacity which would not immediately affect it. But over the course of a few hours or days it could no longer work at peak efficiency (supply enough power), causing erratic behavior. Calculate how much power you need, then add another 100 watts to be safe. Here's a chart to help you calculate your requirements.
http://www.tomshardware.com/howto/20021021/powersupplies-02.html
http://www.tomshardware.com/howto/20021021/powersupplies-02.html
The power supply is easy to test, if you have spares - just swap a compatible one in (even if you have to borrow your own, for the sake of your relationship).
Based on your comments, I would still lean towards the PSU. If after changing that, then as Callandor suggested, sounds like bad MB.
Just in case you did not know, AMD processors are more sensitive to power fluctuations. It is very critical that you have sufficient power to your processor, memory, video, internal devices, and external periphrials.
Your PC may be running fine, but if an application is started that requires CPU power, Video Rendering, and HD data flow, your PS is insufficient, especially with the number of power hungry devices you have plugged in. I suggest you take a look at powersupplies that specify "AMD Approved".
Also, in regards to concerns about ventilation, a better power supply can reduce the amount of heat that is created as well as push more air out through the back, improving your system temperature.
Your PC may be running fine, but if an application is started that requires CPU power, Video Rendering, and HD data flow, your PS is insufficient, especially with the number of power hungry devices you have plugged in. I suggest you take a look at powersupplies that specify "AMD Approved".
Also, in regards to concerns about ventilation, a better power supply can reduce the amount of heat that is created as well as push more air out through the back, improving your system temperature.
I would also suggest that to simply test the PSU issue. Disconnect the floppy, CDRW/DVD rom drive(s) and run the system using minimal hardware. If you can run the more comples apps on this machine and not have the same issue, you can be almost certian the PSU is too small. The CPU requires 62w without the Mobo (approx 25w) or drives (35w for dvd&cdrw)+ HDD (60-120w) and floppy, plus ram (50w+), plus mouse, keyboard, USB, NIC and Video. If the PSU offers only 235w, this is likely the issue.
Again, just disconnect all drives but HDD and you should see an improvement. If so, replace the PSU with 350w or better.
Again, just disconnect all drives but HDD and you should see an improvement. If so, replace the PSU with 350w or better.
Your problem is probably with bad memory or a bad device. The reboot is due to a blue screen of death. The lockup indicates a hardware problem (and could be the result of a driver) I usually test using a Knoppix boot cd and see how stable it runs then. If it crashes or locks up, then you can count on hardware problem. If it runs fine then it could point to an XP problem.
My best guess is a bad memory stick or memory timing problem due to aggressive BIOS settings. Tune it down and test.
My best guess is a bad memory stick or memory timing problem due to aggressive BIOS settings. Tune it down and test.
This is a crap shoot, but are you using a logitech wirelss keyboard? I know you said "new components" and I only had this on an early version wireless keyboard, but it is worth a shot. I had completely random lockup and occaiosnal BSoD and after going through 3 v cards, 4 sets of ram 2 power supplies, etc. I figured out it was the keyboard. (note that I use various logitech wireless keyobards today and have 0 problems)
On another occasion I had lightening damage and I found that the power supply, motherboard and ram cross contaminated each other. I had to replace all three to get it the random crashing to stop.
Have you reformated to ensure it is not a Windows install problem?
On another occasion I had lightening damage and I found that the power supply, motherboard and ram cross contaminated each other. I had to replace all three to get it the random crashing to stop.
Have you reformated to ensure it is not a Windows install problem?
I have anothher theroy, has your GF's electrical outlet been checked?
I dont' mean to be rude, but if she lives in a chap appartment in a slioghtly run down neighborhood, the actual power to the computer could have spikes and dips.
If you aer willing, buy a UPD "uninterruptable power supply" and test how that works what the computer is pluged into that.
I dont' mean to be rude, but if she lives in a chap appartment in a slioghtly run down neighborhood, the actual power to the computer could have spikes and dips.
If you aer willing, buy a UPD "uninterruptable power supply" and test how that works what the computer is pluged into that.
I don't know about anyone else, but I'm interested in the points. He never did try a new PSU as I suggested. (along with others)
I concur with eccs19. The Question and Answer should be added to the knowledge base with the resolution being to replace the PSU.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Everet Home edition should give you the CPU temps. See what it tells you.