Computer lights up but doesn't post, hard drive doesn't work but works fine. [ Edited ]

[BubutBoy]

I have a 3-year-old s3100n with a very interesting power problem.  I've searched but haven't found anything that deals specifically with this issue...

When the computer is plugged in, there's a constant green light on the power supply.  The computer will power on - the blue power light is on, CPU fan spins, optical drive initializes, but there's no power in the USB ports, and the original hard drive doesn't spin up.  If the drive is plugged into any other computer or external enclosure, the drive works fine.  If a different SATA drive is plugged into the computer, the drive spins up when turning on the computer.  Regardless, however, the computer doesn't post., there's no image on the monitor, nothing more.

In the event it makes a difference, the computer was in use when the monitor went blank, but the computer's power light remained on.

I've removed all external components from the motherboard (optical and hard drives, and the memory) with no change.  I removed the BIOS battery with no change.  I even reseated the processor (yes, I was grounded, and used a very good thermal compound when replacing the heatsink - I've done this many times) and again, nothing changed.

Is this a PSU issue, or is this a motherboard issue?  Is the drive bad, or is this a PSU issue.  What's up with the hard drive problem?  I don't have the ability to test the power supply, or the motherboard, so advise would be appreciated.

Thanks for you help!

[8lives]

Are you using a power bar?

Get a PSU tester.

[BubutBoy]

I don't use a power bar, and am unable to find a PSU tester that will support the mini-24 pin connector, nor have I been able to find an adapter for this application.  I do have a meter, and have acquired a pin diagram so I will check the power levels individually.  I'll post the results.

[BubutBoy]

I checked the pins of the power connecter, and all powered pins have pins, and all grounds are grounded.  

Voltages are as follows:

3.3v pins measured 4.67v

5v pins measured 7.13v

+12v pins measured 16.67v

-12v pin measured -16.67v

I checked the power to the SATA connectors, and both measured the same measured values for their intended values.

These seem a bit high to me, and I'll need to know for sure.  This still doesn't explain the mystery of the hard drive, but one mystery at a time.

[8lives]

Hi, the PSU needs to have a load on it.

Get yourself a Antec PSU tester if you can. (20$)

[BubutBoy]

A PSU tester would be nice, but the power connector on the power supply is a smaller version of the standard ATX 24 pin connector, so it won't work with standard testers.  This is an HP proprietary connector, to the best of my knowledge, so it's not found in other computers.  I do have a tester ordered, but will only be able to test the SATA power connectors.  Hopefully this will be sufficient to check the voltage levels.

The power supply was tested while plugged into the computer, and the computer was in a powered on state.

[8lives]

There are 20 pin to small ATX 24 pin adapters available if you have a spare PSU.

[BubutBoy]

I thought of the spare PSU idea and I wish I had a spare, but I don't.  I installed two spare hard drives (500GB & 1TB) on the power supply, so as to add more load to the power supply and turned on the computer again.  This computer only had one hard drive installed as it's only designed for one.  The other is an optical drive.  With both hard drives plugged in, I measured the three positive voltages, and they were still beyond the rated limits.

3.3v pins measured 4.53v
5v pins measured 6.91v
+12v pins measured 16.15v

The voltages gradually climbed while the system was turned on.  The readings above reflect about 20 seconds of power.  I didn't leave it on long enough to note whether they would have reached the high readings I recorded earlier, as I didn't feel it was necessary, and I didn't want to put my other drives at risk.  The rate of climb varied for each and was about .01v every 2 to 5 seconds.  The lowest recorded voltage for each was right at power-on; the 3.3v line read about 4.07, the 5v line read 6.30v, and the 12 volt line read 15.86.

I don't know if such a gradual power increase is normal.  I don't know if the original drive, a Seagate drive has built-in protection from being over powered, and I don't know if replacing the PSU will in fact make the drive and the computer usable.  I suppose I will have to replace it to find out.  Currently the original drive is in a USB enclosure and is working fine.

I'm expecting a new power supply on Friday, 3-19.  I'll post the results.

[BubutBoy]

I learned the computer had a history of having the motherboard replaced, and with it once the processor was replaced.  The BIOS tester suggests the CPU is bad, which means either the motherboard, processor, or both.  The power supply is indeed bad, because it damaged three motherboards and one processor.  The computer is not cost effective to repair, so it will be replaced.  The salvageable parts will be used to build another computer.

It's still a mystery why the Seagate drive is not working on the original or the new power supply, but works everywhere else.  I'll chalk it up to HP proprietary technology.  I'd really love someone to explain this mystery.

8lives, thanks for the input.  I'm glad someone was there to lend advise.