Looking for a particular 5.25" floppy drive - Epson SD521

shadoooo shadoooo at gmail.com
Wed Jul 27 17:31:58 CDT 2016

Hello Adrian,
if the head positioning motor works, but the drive simply doesn't read 
the data,
maybe the problem is in some track misalignment or bad head signal 
amplification / A/D conversion.
Don't know about these drives, but if the interface is pretty standard, 
you could try the correct operation
of the positioning motor simply powering the drive setting the signal 
DIRECTION high or low, and giving pulses to the
STEP signal. You can do it connecting simple wires between 5V and GND 
and the pins on the connector on the back.
The positioning motor should move back and forth in steps.

For the track-0 sensor simply use a multimeter / oscilloscope and check 
if the level changes when the head is in track-0 position.

For the READ signal the thing is more difficult, because you need an 
oscilloscope to check the fast changing signals.
You can check for good looking output signal, without spurious pulses or 
bad edges.
Then you possibly have to check for correct timings, and this is even 
more difficult, but doable with the right equipment (powerful 
oscilloscope). Some drives have a small trimmer to slightly correct the 
spindle motor speed, whence the timing.

Alignment can be corrected loosening slightly the blocking screws 
between the head carriage and the stepper motor moving it,
or in some drive the screws holding the stepper to the chassis frame, 
and moving head carriage / stepper motor very slowly in both directions 
until the amplitude of the signal of the read head is at the maximum. 
Then lock again the screws.

Years ago I had to repair four floppy drives for two DG One machines, as 
they were all failing, not reading nor formatting.
Spindle rotated well, head moved correctly, but the drives couldn't read 
disks, and not even re-read its own data (formatting),
hence something bad on read / write circuits.
After various tests, the problem was on bad electrolytic capacitors on 
the spindle board. As the spindle motor started,
it emitted high current pulses that caused ripple on the supply and also 
induced spurious pulses on the analog read signal.
Result, the read data was badly interfered, and of course not correctly 
interpreted by the controller.
I changed the capacitors, 3 over 4 drives worked. The last one required 
head alignment, but I didn't have any alignment disk.
So I wrote three test disks with the 3 working drives, and tried to 
align the 4th to the best match.
Then finally even the 4th worked!

If you are in willing to try something, let me know.


More information about the cctalk mailing list