Clean the head one time with window cleaner, afterwards with isopropanol to remove the remains of window cleaner and retry it.
Did you generate the reference disk with ibm tools ?

No I did not create them with IBM tools. I downloaded an img file and 'dd'-d the .img file to the diskette drive device file (as in dd if=~/Downloads/recuedisk.img /dev/fdd . Doing so it did it bit by bit. I'm not 100% sure how diskettes work to boot from but if there would have been a MBR, it should have 'dd'-ed that over as well.

I have cleaned the heads with isopropanol alcohol. Still the same.

Could it be that the problem is still deeper? I guess the diskette drive should at least attempt to spin the diskette. It's not doing that.

Maybe a bit more diagnostic information. On the bottom of the diskette drive, at the middle front, there's a brown flex cable with 3 conductors that goes to a device (magnetic head?) that sits very closely to the circle that spins. When I boot the computer, I measure 2V on the 2 outer conductors. It could be possible that those 2V should spin the circle? The circle itself can move freely. It's not stuck in any way.

matze79 wrote on 2022-12-07, 12:48:

Did you generate the reference disk with ibm tools ?

Maybe I'm wrong, but I don't think a diskette drive can know if there is a bootable reference disk in it if it did not spin at all.

I also checked the caps I replaced, they all make proper contact to the components ~next to them (electrically speaking)

I used this reference disk: https://www.ardent-tool.com/disks/rf5565a.zip . There's an img file inside, and as said before, I put the contents of the .img file byte by byte on the floppy disk (with dd)

OK this is funny 😀.

Normally the diskette drive's LED stops blinking after a few seconds, then I get the 161/162 error with the NOT OK symbol we probably all know all too well. But ... If I spin the diskette drive manually, it keeps trying to read the diskette much much longer, I can even feel the head moving sometimes. It's actually trying to read it now. Then after some time out, I get an animation where it requests me to put in a floppy disk, then hit F1. Obviously my "floppy disk spinning skills" are not fine tuned enough to actually boot this way 😀.

So that leads me to believe it's actually the motor that spins the floppy disk that is not working for whatever reason. Is there someone who can help out? 😀

I had the same issue with a drive of an IBM L40SX. After recapping the main board the issue still persisted. It turned out there were some traces around the old caps near the motor that were corroded. Managed to sort that out with a multimeter and a botch wire. After that it happily spun away.

As additional advice; try to disassemble us little as possible. If you get the drive motor out you can mess up the relative head alignment, which is super tricky to fix.

bucovaina89 wrote on 2022-12-07, 13:17:

No I did not create them with IBM tools. I downloaded an img file and 'dd'-d the .img file to the diskette drive device file (as in dd if=~/Downloads/recuedisk.img /dev/fdd . Doing so it did it bit by bit. I'm not 100% sure how diskettes work to boot from but if there would have been a MBR, it should have 'dd'-ed that over as well.

I have cleaned the heads with isopropanol alcohol. Still the same.

Could it be that the problem is still deeper? I guess the diskette drive should at least attempt to spin the diskette. It's not doing that.

Did you use USB Floppy ?

matze79 wrote on 2022-12-07, 20:41:

Did you use USB Floppy ?

Yes I did.

You could try a normal 3,5" Drive.

Or make a Adapter Cable:
https://ps1stuff.wordpress.com/how-tos/ibm-ps … py-replacement/

I was already thinking in that direction. Combined with a USB-stick to floppy adapter piece to at least get the machine running again. I do believe it's rather OK apart from the diskette drive and likely the HDD but there's no way now for me to make use of it.

i guess you got a ALPS drive in your 55sx ?

Blavius wrote on 2022-12-07, 16:56:

It turned out there were some traces around the old caps near the motor that were corroded. Managed to sort that out with a multimeter and a botch wire. After that it happily spun away.

Blavius said it.
on ALPS drives, sometimes it's even necessary to desolder the IC which control the spindle motor as traces can be cutted under it... or sometimes tin is so corroded than pin of the IC are isolated from the pads 😁
ALPS drives can require a lot of patience, i saved around 10 of them (except one with a defective head coil, but kept as donor to save others).
they are fixable most of the time, but can be very tricky...

Thanks Waka for your reply. Can you say which IC it is that controls the spindle motor?

you can't really miss it, it's near the flywheel.
and if we talk about the same drive (DFP723Dxxx), you replaced 2 capacitors near it.

OK I agree, I did replace 2 caps near it. and I guess each cap directly connected to a pin on the IC.

So would logical next steps be?
1) desolder IC
2) check traces under it
3) solder IC back on.