|
|
Post by Cube Inc. on Jan 25, 2023 8:45:27 GMT -7
I purchased a SFD-1001 from an European seller, and although I know it was 100% working when he shipped it, when it arrived I noticed a bit of an odd problem. At first everything looked good, I was able to read disks I'd created with my old SFD-1001 and 8250 drives, but when I started copying files off of them, some files read perfectly while others would not read at all, resulting in a "FILE NOT FOUND ERROR" message when I attempted to load them. After some further testing I decided to try formatting a blank disk, and although the drive would attempt to start the format, it sat there after banging the heads for about 2 minutes before erroring out with a "21, READ ERROR, 78, 0" error.
I checked the usual stuff: I cleaned the heads, I tried another couple of disks, I even tried turning it off and then on again (in an English accent). Same result every time. Then I started trying oddball stuff: I disconnected the read/write heads from the main PCB and tried a format. The format started with the same head bump, but then immediately errored out with a "75 FORMAT SPEED ERROR, 1, 0" error (Track 1, sector 0.) If memory serves me correctly, If I leave the heads connected but physically "lift" the top head off the disk, the format starts the same way but errors out with a "75 FORMAT SPEED ERROR, 78, 0" (Track 78, sector 0). Track 78 is the first track on the top side of the disk, so it seems that the head is reading [something] from the disk.
After examining the schematic for the SFD, I put a scope on the center tap pin of the bottom head and can see regular pulses of about 9VDC on that pin, even when the head is idle. However, when I scope the center tap pin of the top head, there are no pulses at any time. This makes me wonder if the LS7445 BCD decoder that controls the center tap pins for each head to determine which head is selected for reading or writing is either stuck sending power to only the bottom head, or it is being told to do so from further up the digital chain.
I'm going to continue investigating further, but I thought I'd put the question out there since I know there are at least a few members here who still have SFD-1001 drives: Have you ever encountered this situation with your own SFD's and if so, were you able to fix the problem?
|
|
|
|
Post by Cube Inc. on Jan 25, 2023 8:58:54 GMT -7
I should also point out that I am able to save files to disks with the drive, so long as the files are being stored on the bottom side of the floppy. It is as if the bottom head works perfectly, while the top head does not work. I don't think the top head has failed entirely, since lifting it off the disk results in an immediate error whereas it tries for some time before giving up otherwise. Whatever is causing the problem, the drive just cannot seem to properly get data from the top of disks, like there is too little signal or too much noise that it just cannot get a lock on what it is searching for. I almost wonder if the top-to-bottom head alignment or azimuth could be out.
I have also adjusted the disk rotation speed so that it is spot on 300 RPM, and the coil resistances between each of the coils in both heads seems reasonable. I'll measure it again and record it here for posterity, but I didn't notice a significant discrepancy from one head to the other - certainly none of the coils was open or shorted.
The 12V and 5V supply rails are also both nice and stable. I even tried running it off my bench power supply but it made zero difference. So far the biggest lead I've come across seems to be the difference in the voltages I see going to each head. It seems very odd to me that one head is getting regular pulses (the one that works) while the other head (the one that cannot seem to read any data) gets nothing, ever. I will have to spend some more time investigating that anomaly in more detail. The technical & service manuals give the schematic, parts lists and the proper value for certain measurements, but I couldn't seem to find a theory of operation or sequence of operation description.
I have ordered a second SFD-1001 so that I can compare one to the other, which should help with the troubleshooting. It should be here in a week or so.
|
|
|
|
Post by Cube Inc. on Feb 5, 2023 12:25:51 GMT -7
A quick update: I bought another SFD-1001 and swapped the main PCBs between the two... The problem follows the drive mechanism and both PCBs work perfectly well in the (other) SFD-1001 with its drive mechanism, so this is definitely a problem with the mechanical drive and/or the R/W head. At least now I know where to focus my troubleshooting, and I have a good working unit to compare against now.
I'll keep you posted...
 |
|
