No temperature or voltage readings from 990FX Pro

Hi,

I have had my AsRock 990FX Fatal1ty Professional now for 4 years without any problems, but now I cannot get temperature readings from CPU package or ambient sensors or any voltage readings. In BIOS (latest) at HW monitoring it shows only "N/A" on everything else than fan speeds and in Windows AIDA64 can see CPU core, VRM1 and VRM2 temperatures CPU core, CPU VRM and +12V VRM voltages and SpeedFan shows system, CPU, AUX and SMIOVT4 - 6 temperatures as "-128C" and all voltages as "0,00V".

The problem started when I was troubleshooting my girlfriends PCs GPU (MSI GTX 550Ti) on my own PC and after I switched back to my own GPU (Palit GTX 770) and then the fan speed automation went haywire and I noticed I can't get temperature readings. I tried to switch back to the GFs GPU, but that didn't work. I have also tried to reflash BIOS without any effect.

On motherboard there doesn't seem to be any physical damages and I don't get the "burned electronics" smell from it. Also there seems to be no other problems than just that I cannot get the sensor readings and the AsRocks fan automation doesn't work (because it doesn't get the temperatures).

Might the problem be in the hardware or could it be fixed with some BIOS flashing or other means?

My PC is:
AMD FX-8350
AsRock 990FX Fatal1ty Professional
Noctua NH-D15
4x4GB Kingston HyperX 1600MHz DDR3
Palit GeForce GTX770 2GB Jetstream
OCZ Fatal1ty Chapion Series 750W PSU
Creative Sound Blaster Z
3x Samsung 1TB HDD
Windows 10 Pro 64-bit.

Try the three things parsec suggested yet it sounds as if the sensor chip or surrounding area took a hit or volt.

IcedForce wrote: parsec wrote: -snip- Thanks. The GTX 550Ti works perfectly, problem with GFs computer was fans and mb just being too old. Problem with AIDA64 is that it doesn't even see those sensors so I can't get graphs from them. What I find odd is that AIDA64 can see and monitor the VRM temps which I would think to be connected to the Nuvoton chip like the sensors not working. CPU core temps, GPU temp and the +12V rail sensor data comes from elsewhere. The PSU seems to be ok. Tried with my older PSU which does have some problems with PCIe cables and while on it probed the PSUs rails which were within acceptable margins (I tried with some older HW that has been left over from our PCs). Forgot to check the battery, so that's still a possibility. Do you think it's ok to keep using the mb even if the nuvoton chip is gone? At least at this moment there's no other problems and I control the fans with SpeedFan through CPU core temps and GPU temp. Also no other symptoms arise even with stress tests, so the loss of sensor data doesn't seem to affect the performance.

Don't forget we are doing a diagnosis simply from the information you gave us. We have in no way tested the Nuvoton chip, if that is even possible, and if we knew what to do if we had the appropriate test equipment. All you could do is describe the situation in your post, which is all I could do myself too.

For all we know, the data connections between the various sources and the Nuvoton chip were damaged, or a component like a capacitor, resistor, or some kind of buffering chip between the data sources and the sensor chip failed naturally, or was damaged. It seems like that could be what happened, since you still get some sensor readings, as you described.

My diagnosis was a pure guess, based upon what you told us. To be more accurate, I should have said something failed in the sub-system of your board that provides the sensor data that you listed as missing, to hardware monitoring programs. Blaming the Nuvoton sensor chip is easy and understandable, but I truly have no idea what actually happened.

All it takes is to drop a large screwdriver, whose tip impacts the board at just the wrong place, and one of those small parts is damaged. What function that part has will be random, and we might not even notice, for example, that our 1394 port on the IO panel no longer works, because we never use it. Or the part is for the control circuit of a PCIe x16 slot, which then no longer works, which we notice immediately because our video card no longer seems to work.

It's a miracle those tiny surface mount components work as well as they do, and stay in place on a mother board. There must be hundreds of them spread across the surface of a mother board.

My point is it is difficult to give you an answer to the question, is it safe to use the board, that is accurate. I could write up scenarios I imagine in my mind about what might happen, but if nothing else has happened so far, you may be safe to use the board as is.

But can anyone guarantee that is the case only based upon a description like you gave us? Not that you could have told us more, unless you removed the board from the PC case, and spent hours going over it with a magnifying glass and actually found damaged parts. Plus a part can fail and show no physical signs of damage. Even if we knew a part was damaged, could we know if the failure of one circuit would not affect another? Electronics are not simple, at the level of a mother board.

I wish I could answer your question with a correct response, but honestly I can't, sorry to say.