PC hardware problem, which component is bad?

Factual Questions
1

Ok, I finally got all the parts for my new PC, and put it together, but it has major problems.

Here is the hardware:

Asus A7V333 RAID
Athlon XP 1800+
256 MB PC2700 DDR
Samsung 40 GB ATA 100 (2)
300 W ADM approved power supply
Palit GeForce 4 Ti4200 128 MB
The first time I attempted to boot, the BIOS came up and asked me to select the Processor speed (1.1XX GHz, 1.53XX GHz, or Manual) No matter wich speed I chose, the voice system would say “System failed due to CPU overclocking.”

After reseating the processor a million times, I tried switching RAM slots. This fixed that problem, the machine was now able to boot from a WIN 98 boot disk.

I configured the RAID array, and tried to use FDISK to partition it (instructions say to do this). It repeatedly froze at anywhere from 1% to 50% while verifying drive integrity.

I decided to disconnect from the RAID and just plug one of the HDs into a regular IDE connector to see if the drive was ok. It partioned correctly, formatted 100% (on the 3rd try, it froze the first 2 times), started calcualting free space, and then I get

“Your Program has caused a divide overflow error.”

Now I get that same error when I try to do anything after booting from the boot disk. I also tried other boot disks, the other HD, always get the same error.

I tried running DocMemory, but it froze before the test was complete, no errors found.

The hardware monitor in the BIOS shows the processor temp as 104º F right after a format attempt.

Which component do I need to take back to the store? I don’t have any extra processors, RAM, or motherboards that I can use for testing (I only have old junk).

Thanks in advance.

2

  1. Check the voltages in the BIOS. See if the +5V, especially, is running lower than it should be (5.0v).

  2. What cooler are you using? Did you remove the thermal pad from the bottom of the heatsink? Are you using thermal paste?

  3. As I recall, your board has a “Palomino Support/CPU Overheat Protection” jumper. Is it enabled or disabled? IIRC, enabling this jumper disables CPU temperature measurement due to a bug, locking the CPU temp display to 42C, even though the CPU may be running much hotter.

  4. Is the motherboard contacting the case metal at all? There should be standoffs or studs ensuring that it doesn’t touch the metal case at any points, else it will short circuit.

  5. Finally, make sure the CPU is setup properly in the BIOS. It should be 11.5*133Mhz @ 1.75v.

Everything you’ve suggested sounds to me like CPU overheating, except the problem that was solved by moving the RAM. Try removing the heatsink, cleaning it and the CPU thoroughly until there is no trace of thermal gunk remaining, then applying a paper thin, even layer of thermal paste to the CPU core (blue square). If you don’t have any thermal paste, you can obtain a packet for $1 from any computer shop or Radio Shack. Reinstall the heatsink. Then, open the case and point a room fan into it. This should quite thoroughly eliminate cooling as a factor. If this doesn’t affect stability (if it becomes more stable, but doesn’t work completely, then you need to buy a better cooler), and the voltage readouts in the BIOS are fine, then the problem is either the RAM or the motherboard.

On preview, it also occurs to me that I have heard that Asus is having problems with the A7V333, to the extent of changing the warranty procedure to make it easier to replace the board. Something to check out if it turns out to be a faulty mobo.

3

  1. Check the voltages in the BIOS. See if the +5V, especially, is running lower than it should be (5.0v).

  2. What cooler are you using? Did you remove the thermal pad from the bottom of the heatsink? Are you using thermal paste?

  3. As I recall, your board has a “Palomino Support/CPU Overheat Protection” jumper. Is it enabled or disabled? IIRC, enabling this jumper disables CPU temperature measurement due to a bug, locking the CPU temp display to 42C, even though the CPU may be running much hotter.

  4. Is the motherboard contacting the case metal at all? There should be standoffs or studs ensuring that it doesn’t touch the metal case at any points, else it will short circuit.

  5. Finally, make sure the CPU is setup properly in the BIOS. It should be 11.5*133Mhz @ 1.75v.

Everything you’ve suggested sounds to me like CPU overheating, except the problem that was solved by moving the RAM. Try removing the heatsink, cleaning it and the CPU thoroughly until there is no trace of thermal gunk remaining, then applying a paper thin, even layer of thermal paste to the CPU core (blue square). If you don’t have any thermal paste, you can obtain a packet for $1 from any computer shop or Radio Shack. Reinstall the heatsink. Then, open the case and point a room fan into it. This should quite thoroughly eliminate cooling as a factor. If this doesn’t affect stability (if it becomes more stable, but doesn’t work completely, then you need to buy a better cooler), and the voltage readouts in the BIOS are fine, then the problem is either the RAM or the motherboard.

On preview, it also occurs to me that I have heard that Asus is having problems with the A7V333, to the extent of changing the warranty procedure to make it easier to replace the board. Something to check out if it turns out to be a faulty mobo.

4

I agree with FDISK. Sounds like the CPU is over-heating. Had a similar problem and re-installing the heat sink with the proper thermal goop ( technical term ) fixed it.

5

My money is on a faulty mobo.