No it's not normal. The head is suspended with a very sensitive metal spring/plate thingie. This spring mechanism gives it some flexibility to move up and down. Looks like the head was pulled off in a direction that stressed the spring out of its normal range, so now the head is crooked. Try to push back lightly on it and see if it gets closer back in position. Hopefully the spring parts that connect to the head aren't broken off. Be gentle to avoid breaking it if it's not already.

BTW, this is a symptom you get by using those stupid cleaning disks. It can often clog to the head and the spindle motor will keep pulling the disk and the stuck head along with it. I don't recommend them at all.

I concur, the spring is bent or even broken. If only bent it could be perhaps fixed, but to do this properly the upper head arm needs to come off. And that will require realigment afterwards, though even fixing the spring in some other way might shift the head a bit so I'd do that anyway.

If the spring is broken then you need to transplant the upper head, with the arm, from a donor drive. Or just replace the drive if any of this seems too difficult or time consuming.

An upper head transplant is a bit pointless unless you have an analog alignment disk to get both the radial alignment *and* the azimuth right

Thanks for confirming my suspicions, it seems it's exactly what I was afraid of... So having nothing to loose at this point I released the spring that presses the upper head down. This revealed just enough space to have a good view, the thin metal plate suspending the head was clearly pulled from one side. Using a cotton bud soaked in isopropyl (to be soft and not leave anything behind) it was very easy to push the head from the opposite side and make the metal plate sit completely even again.

Unfortunately this head still doesn't read anything in imagedisk. It's not off, just plain nothing. I assume something was ripped when the head was pulled, but it's not obvious. The head's ribbon cable looks ok and doesn't move, but you can't know of course the condition of the part that goes under the metal plates. I think I'll have just one more look with fresh eyes and consider this gone.

maxtherabbit wrote on 2023-10-17, 02:09:

An upper head transplant is a bit pointless unless you have an analog alignment disk to get both the radial alignment *and* the azimuth right

Difficult, yes, but not pointless. Floppy discs are not that densly written and rather forgiving of minor head shifts. Usually there isn't all that much slack for the upper head to move sideways with 2 screws holding it down, so getting it in the middle, and then comparing with lower head (whether it sits properly and evenly or not) plus some scope measurements will usually get it close enough to work without any issues. It sure beats having to find and possibly clean/fix another drive, including minor cosmetic issues like front bezel color.

Now if the head spring separated completly from the arm itself (it broke, the arm broke, or the glue somehow let go) then that would be difficult to get right. Not impossible, certainly not with some luck and trying to position it using what is left of the original glue (or any other marks) but it would require a lot of effort.

As for the failed repair itself, it's very critical that both heads come together without a gap - magnetic heads will tolerate some lateral shift with only minor degradation to signal strength but any air gap between head and media will pretty much kill the perfomance.

Exactly. The two heads must touch each other when the drive is closed without a disk inserted. If there is a gap, then it means the upper head was actually adjusted inwards, and now must be pulled out again. But also if you pull it outwards too much, it can start scratching your disks. Test against disks of no value.