USB 3.0 PCIe card sometimes not detected in BIOS on one PC but no issue on another identical PC


This problem is solved. See my post 2nd post below

About a month ago I bought two Dell Optiplex 760 MTs. Both systems are exactly the same in every way and work very well.

Last week, I bought a Dodocool 2-port USB 3.0 PCIe x1 expansion card from ebay and it works on both Optiplex 760 PCs when placed in either the PCIe x1 or x16 slots.

However, on only one of the systems, it's a hit or miss as to whether the BIOS detects the card from a cold start up (the BIOS detects all the PCIe slots, just sometimes not this particular card – it will just show "empty", when it should say "USB").

As a result, if the BIOS doesn't detect the card, then it's not available in Windows when Windows loads in. Therefore, when I start the PC, I always have to enter the BIOS to see if the card is detected before windows boots.

If BIOS shows the card, fine, I just exit the BIOS and let the PC boot into Windows and all is fine. But if the BIOS doesn't detect the card, I have to exit the BIOS, push the power button on the tower to shut the system off, then power on again and usually the card is recognized (sometimes might take 2 or 3 tries but many times only one).

It never happens if I simply do a restart from the OS. On a restart, the BIOS always remembers/sees/detects the card. It only happens if I do a shutdown and have to power-up the system using the power button on the tower.

This problem happens only with this one system and only with this particular PCIe card and not any others and it doesn't matter if it is installed in the PCI x1 or PCI x16 slot – the problem is the same. No problems on either PCIe slot with any other PCIe cards.

I have tried many, many, many times to replicate this problem on the other identical PC but I cannot replicate it.

At first I thought it was a defective mobo or defective PCIe x1 slot but when I saw that it also happens on the PCIe x16 slot and all other PCIe cards I've tried have no problem, I don't think I can conclude it is a bad mobo or slot(s).

I also don't see how it is a defective PCIe card since it works flawlessly on one system. In fact, it works on both systems, just that in one the BIOS might not detect it.

I'm hoping someone here can shed some light.

Here is a list of everything I have tried…

  • Flashed the BIOS 3 times now.

  • Tested voltages both PSUs under load. Both systems' voltages tested out fine.

  • Ran the command "bcdedit /set CONFIGACCESSPOLICY DISALLOWMMCONFIG" in Safe Mode with command prompt. I did this because I read an old thread from another forum for a similar problem and this fixed that person's problem. But it did not fix my problem.

  • Reset the CMOS

  • Changed CMOS battery with fresh one, even though the old one tested at over 3v

Both systems:

  • Dell Optiplex 760 MT
  • Nanya 6GB DDR2 PC2-6400 (800MHz) RAM
  • Intel Core 2 Duo 3.0GHz E8400 Wolfdale CPU
  • Dell Inc. 0M858N mobo
  • Dell BIOS vA16
  • Windows 7 Pro 64-bit SP1
    (using on-board GPU)

Best Answer

The problem is solved!

I tested the output voltages on each PC's PSU and since the voltages were correct, I assumed each PSU was good and that the PSU could not be the problem, therefore switching the PSUs would be a waste of time.

I was wrong to just test the output voltages on each PSU and not switch them out. The problem is in fact the PSU.

This morning I decided to take a look at the spec sticker on each PSU and, lo and behold, they are not identical. They look identical but each has a different part #.

  • The part # on the problem PC's PSU is P192M, model # L305P-03.
  • The part # on the "good" PC's PSU is 0MK9GY. model # H305P-02.

All the specs on the stickers are identical expect for the input specs.

  • On the problem PC (PSU P192M), the input says 100-240V~/5.6A (50-60Hz) (which is the same as many of the "Bronze" aftermarket PSUs I have been looking at recently online)
  • On the "good" PC (PSU 0MK9GY), the input says 100-240V~/4.7A (47-63Hz).

I'm not sure if these specs are what make the difference, I'm just pointing out the differences in the specs on the stickers. It very well could be that the input specs mean nothing and that the P192M is starting to fail, but I have had no other problem than the intermittent detection of this USB card using the P192M PSU. The PC with the P192M part # was manufactured about 8 months before the PSU with the 0MK9GY part #, so it could be that Dell replaced it due to this defect I am seeing.

So now, the P192M PSU is in the other PC (the "good" one) and now that PC is suffering the same problem - intermittent detection of this one PCIe card in either PCIe x1 or x16 slots.

And now, the problem PSU has the 0MK9GY PSU in it and it is no longer the "problem" PC, it is now the "good' PC. Detection of the PCIe card in either x1 or x16 slot is always detected...ALWAYS!

So the formerly "good" PC is now the problem PC and the former problem PC is now the "good" PC, all because of switching the PSUs!

Just goes to show you that testing output voltages, even under load, is not a 100% indicator of a faulty or "problem" PSU!