Osborne 1 floppy drive troubleshooting

[jltursan]

I'm trying to fix a faulty Osborne 1 drive (a Siemens FDD-100-5) and that's what I've found until now:

Write troubleshooting:

- The faulty drive cannot write at all, when formatting with COPY tool, the Osborne hangs when writing track 0, drive is spinning but nothing more, it stays forever...
- If I try to format a protected floppy, it's detected correctly as protected (or at least seems so) and each track shows E and V. The working drive does the same.
- Just to be sure there's no problem with the WP sensor, the voltages measured are the same in both drives, low (0,1v) when open, high when closed (4,8V).
- As I have another working drive, swapping PCBs shows that the PCB seems to work fine.

Reading procedure:

- The faulty drive reads mostly flawlessly, at least all the Osborne originals.
- When pushed far seems to have some little misalignement compared with the working drive. It usually fails in reading PC written disks (with an HD drive), if the disks is in very good shape it could work sometimes. The working drive always read the disks created with the PC.

If PCB seems to be fine, could it be simply a head problem?, is this a common failure of Siemens drives?

 

[Chuck(G)]

Check to see if you've got an INDEX signal active coming out of the drive. Hanging while writing is sometimes a symptom of the sensor being kaput.

 

[jltursan]

Thanks!, yep, INDEX seems to pulse (checked with a logic probe) correctly, I can clearly see while reading the disk. I suposse that it means the sensor works fine.

If I try to format, it "pulses" one time and that's all...:-(

 

[Chuck(G)]

That doesn't make much sense, unless you're losing the drive select or motor enable. Formatting is performed index-to-index.

There's something strange going on. Have you checked the spindle speed? Is it possible that you might be having a slipping drive belt?

 

[jltursan]

It must be something related to the writing procedure as reading seems to work fine, really weird. The motor is constantly spinning so the signal must be present, I can check if it's loosing the select signal...
About the RPMs, I don't have a stroboscopic light, is there any tool available for the OS1 capable of measure the speed?. Anyway, I think it's mostly correct, it can read flawlessly a floppy written with the working drive.

I'm going to double-check PCBs and swapping them again, just to be sure.

 

[Chuck(G)]

Well, an LED night light might make a good stroboscopic light, as well as one with a neon lamp.

If the drive is selected, it doesn't make any difference if you're reading or writing or doing nothing--the INDEX signal should be active regardless as long as the disk is spinning. I can think of only a few ways that a format or write would fail.

1. Bad media--but I imagine that you've eliminated that.
2. Write-protected media (or a bad sensor) you can check pin 28 on the drive interface for that
3. Wrong disk speed--a format involves writing a track from index to index--if the drive is spinning too slowly, the software runs out of data before the second index pulse and throws an error. Note that reading a disk isn't proof of correct speed--the data separator can have a range of +/-20 percent and still work when reading.
4. For some reason, the system isn't asserting WRITE GATE (pin 24) and not sending data (pin 22) while formatting.

No. 4 is a catch-all--it could point to a bad driver chip in the system, a broken cable, or a fault in the drive. Do you have the double-density option installed? Is it still secured firmly to the main PCB?

Just thinking aloud here--diskette drives aren't rocket science. I'm assuming that you have two drives and that if you swap drives, the fault follows the drive and not the position in the drive cable.

 

[jltursan]

Thanks for the diagnosing!, indeed, I've swapped drive positions and no change at all, I've also swapped again PCBs and definitely both are OK, the failure remains in all cases with the mechanism. That leaves only the sensors (WP & INDEX), magnetic head, rotational speed and cabling, right?
And yep, I've a double-density board installed (but I'm not sure if the Siemens drives support this double density) and IIRC it was firmly secured indeed.

From the above points, I've checked so far:

1) No bad media, I've used several floppies and after formatting in the working drive, the same floppy is not formatted in the bad one. Note that there's no error at all, the operation hangs with the drive spinning and the red led permanently on.
2) No (manual) write protection, the sensor was my first testing point; it seems to react when the notch is closed or not and the voltage values are the same as in the working drive. I haven't checked the cables tho...maybe the signal is lost in its path to the PCB...
3) If writing operation depends so badly of the speed, definitely I need to check this.
4) Yep, I still need to check every signal involved in the writing procedure. INDEX is the only one checked with a logic probe right in the PCB so the signal reaches the circuit.

I'll start checking every cable in the mechanism...