the beginning of the end for floppies

The basic procedure is the same for all drives. You conenc the drive to some device that will let you move the heads to a given cylinder (either the machine the drive works with running a suisable program that talks to the disk cotnroller, or a special floppy drive exerciser). You monitor the output of the read amplifier (notmally using a 'scope). And you put in a special 'catseye' alignment disk. Then you step to the right cylinder and look at the read signal. It should have 2 lobes Yiu loosen the head positioner moutings nad moce the positioner until the 2 lobes have the smae amplitude. Clamp up the positioner moutings and you're done Now for the problems. The first is getting the alignment disk. They are not commn (particularly the 3" one), they are not cheap, and they can't be copied (for obvious reasons). The second is that uou normally need a 'scope. And you need to find where to connect this 'scope to the drive electronics (the signals you need are the output of the index sensor as a trigger and the differential outputs of the read amplidier). FInally moving the positioner may be non-trivial. It's a delicate adjustment, some manufacturers made speical tookls to make it easier -- but it's possible to manange without them. There was a produict called the 'microtest' that did away with the need for a 'scope. It was a microcotnroller + ADC that linked ot a PC [1] serial port. You connected the drive as drive B: o nthe PC (with adapters for 8" drives, etc), and ran the suppleid software. You selected the drive type ADC box, and then you put an alignment disk in the drive. You could see how far off-alignment it was on the PC screen and adjust appropriately. [[1] The minimum spec was 256K RAM and a serial port. And any display adapter, even MDA. The problem with the microtest is that (a) it only works for drives that it knows about (which means not many modern ones..) and (b) the instrument is not common. I was given one, and I use it for my older drives. but for the modern drives I have to us a 'scope. Actually the bigest problem is finding the testpoints. On old drives, with simple chips (733s, 592s, etc) in the read chain it's pretty easy to spot them. If you have the service manual (again more likely for older drives) they will be shown. Most modern drives do have said testpoints, if only as solder blobs on the PCB, but it can be a pian to work out which poits to use. They are rarely marked. I would certianly use an anti-static bag! I don't worry about doing it if I have the right size alignment disk and have some idea of the testpoints. For example when I was cleaning up an HP9836CU I didn't worry about totally stripping the floppy drives (losing the alignment)[2] because realigning them was not going to be a problem. As I mentioned, when I bought my last floppy drive 15 years ago, I also bought the service manual, wghich of course shows the testpoints. [2] Althought I did scratch amarks on various parts of the drive, I also seeked the heads to drive 0 using an exerciser and measured the clearance between the head carriage and the end stop with a feeler gauge. And I put things back ion the same positiones. This is not accurate enough to do away with the need for a proper alignment but it meant that I could at least find the right cylinder on the alignment disk :-) I've come across some more modern drives that would not reliably read disks formatted on my PC. I found what I think were the read amplifier outputs and looked at hte catseye pattern with a 'scope. It was wildly out (one lobe much larger than the other). Of course I might bave been using the wrong TPs (say with a filter circuit in one side only), but I moved the positioner to eqalise the lobes and the drive worked reliably after that. I think it ws just badly aligned. Contrary to what it says in most manuals, I have never had any problems cleanin the heads on any type of floppy drive uisng a cotton bud dipped in propan-2-ol. You have to be careful not to distort the gimbal spring for the upper head, but it's quite easy to avoid doing this. Do you have any particular makes nad models of drives in mind? If we can track doewn service manuals it would help... -tony