What is pin 34 connected to when the RY is open? Do you have ohm meter? There should be a resistance of about 150 ohms (older drives) or about 1k (newer drives) to +5V rail on that pin.

Older PC DD drives did not support DC (disc change(d)) signal on pin 34, and would instead output nothing or RY (ready) on that pin. RY is not compatible with DC. If RY is asserted or if there is no pull-up that I mentioned above it can be interpreted as DC and that would cause the BIOS/OS to return error code every time you try any read or write operation.

It it possible that your mobo is so new the BIOS just assumes the drive will always have a working DC signal - it should detect or just completely ignore it if the floppy is set to 360k type. It could also be a BIOS bug.

Hi Deunan,

thanks for the feedback. There's nothing connected to pin 34, it's open circuit to both GND and 5V.
I am testing on relatively new MB (Socket 7). I guess I should test on the original 5160 too, or alternatively pull pin 34 up for testing on new hardware, right?

Thanks,
Jakub

The jumpers look fine. Try the drive in its original machine to ensure that it really can read anything. The terminator should be set on this drive, but I don't think it matters for newer motherboards.

DS are the drive number (drive select, standard for the twist cable would be DS1 and the drive after the twist gets that swapped to be drive A)
HS = Head to Select, HM = Head to Motor - not sure it makes much difference which you use
TM would be terminator, single or last drive on cable set ON

Kubik wrote on 2025-03-08, 11:50:

I am testing on relatively new MB (Socket 7). I guess I should test on the original 5160 too, or alternatively pull pin 34 up for testing on new hardware, right?

S7 should support 360k drives properly. If you can, add a 1k pull-up to 5V on the 34 pin. Anything up to 10k will work as well since the signal is disconnected. You can even do it with some sticky tape first, rather then soldering, just to see if works. Pushing one lead of the resistor into the connector might even be better but be careful not to damage the contact springs inside. On older floppy controllers this isn't even required as the inputs are TTL-like and will default to H state with their internal pull-ups. But this mobo is most likely pure CMOS already and the lact of this resistor might be causing issues.

Deunan wrote on 2025-03-08, 22:35:

On older floppy controllers this isn't even required as the inputs are TTL-like and will default to H state with their internal pull-ups. But this mobo is most likely pure CMOS already and the lact of this resistor might be causing issues.

Actually, the floppy Interface is specified that all signals (except for the data line, possibly) are driven by open-collector outputs, and pulled high by the termination resistors, which are located at the "far end" (i.e. the second drive for signals from the PC to the drives and at the controller for signals that go from the drives to the controller). This means even if the controller is not built from TTL logic, it is supposed to already have a pull-up on pin 34.

You're right. Outputs on floppy drive interface are terminated on the input side. I've been working on SCSI lately and actually fixed an issue with partial lack of termination - that probably contributed to my brainfart. But the resistor test could still be useful if the termination on the controller side is somehow faulty. 1-10 kohm is not going to upset anything, worth a try.

BTW an issue with track 0 sensor or signal can also cause this problem where any attempt to access floppy fails. But that should be detected by BIOS if floppy seek option is enabled. It should display a floppy drive error message of some sort.