I obtained a GTX 970 that the previous owner said wouldn't install drivers after they shorted the fan header pins.

I have since re-flashed the latest bios for this card and the card now works, except there is a fault with the tachometer (fan speed sensor). The fans run at 100% speed on loading windows.

GPU-Z doesn't register the fan speed, I'm thinking it's either a fault on the fan itself or the sensing circuit on the graphics card. But because the bios got corrupted I think the bios chip is probably faulty.

There appears to 3.3 v on the PWM line from the fan on loading windows, when the fans kick in at 100% (it starts at around 2v when the fans run quiet before loading windows). I can see 1.6 v from the fan tachometer line to the fan header. 12 v are also present on the GPU fan header and is not shorted to ground.

How can I diagnose where the fault is coming from exactly? Will the fan speed circuitry be controlled in the bios chip?

The fan model is PLD10010S12HH, and the bios chip is Pm25LD020 CE1437.

Thanks.

It's not very clear what the BIOS chip has to do with it. The speed of the fans is controlled by the GPU as far as I know.

Yeah that's what I was thinking, I just assumed there would be some circuitry that I could test for faults or a blown component to get the tachometer working correctly.

Thanks.

Теперь можно только приколхозить отдельную схему регулировки оборотов с термопарой под ВРМ.

Any examples of how this can be done?

I was thinking about adding one or two diodes to reduce the voltage on the PWM signal if I can't figure out how to control with an external circuit.