I was a bit too fast there, I meant "universal tester"! The one I think I saw on ebay was in a small suitcase, maybe it was able to test hard drives too! Can't remember any brand or so!

Date: Sat, 10 May 2014 10:36:57 -0700 From: cclist at sydex.com To: Subject: Re: Restoring old floppy drives!? On 05/10/2014 10:14 AM, Fred Cisin wrote: What stopped me behind the starting line on this discussion is what was meant by "old". It seems that almost everything I have is considered "old" by someone--even the PC I'm typing this on. So, does the OP mean 8, 5.25, 4.0 or 3.25, 3.0, 2.8 or 2.5 inch drives? 48, 96, 100 or something else tracks per inch? If it's 8", there are some pretty good maintenance manuals out there that should be the definitive references for their drives. If you're looking for an exerciser on a PC, Dave Dunfield's ImageDisk has some very useful functionality. If you're looking for an alignment kit, you can still buy them from Accurite, the last time I checked--they also sell plain old analog alignment disks. Have cash ready. --Chuck

From: ard at p850ug1.demon.co.uk Subject: Re: Restoring old floppy drives!? To: cctalk at classiccmp.org Date: Sat, 10 May 2014 23:43:00 +0100 Waht do you mean by 'old'? What sort of drvies (8", 5.25" ?). I've done quite a few floppy drive repairs, 8", 5.25" nd 3.5", My methods is togive the drive a good clean (includign the heads), then put it o na drive exerciser to see which of the functions (spindle rotation, index sensor, head stepping, track 0 sensor, read/wirte, etc) work or not. Then to sort out each bit at a time. The manufacutrer's service manual is useful if you can get it, Ther are a few picturs in my flickr account (tony_duell) of a couple of 5.25" drives I've restored. It shouldgive you the general idea of what's invovled intaking one apart. -tony

From: ard at p850ug1.demon.co.uk Subject: Re: Restoring old floppy drives!? To: cctalk at classiccmp.org Date: Sun, 11 May 2014 21:46:32 +0100 Actuakly, the basic functions of all 'normal' floppy drives (I am overlooking the very high density ones with servo infromation on the disks) are much the same no matter waht the diameter of the disk. So the repairs follow much the same pattern. As for cleaning, I normally use propan-2-ol (isopropanol, isopropyl alcohol, oh damnit CH_3.CHOH.CH_3) to cloan everything. It same on the heads, but it will also disolve oil from bearings, etc, so it's not a good idea to soak spindle bearings in it unless you intend ot re-lubricate them. A few other comments... If you dismantle the heads, stepper motor, track 0 sensor, index sensor, and someimes other parts too, you will upset the alignment. To re-align the drive you need a special alignment disk which is expensive and for obvious reasons cannot be coppied usign an ormal drive (it has deliberately off-track recordigns). You may notices in some of the hpotos that I measture where thigns are (e,g, use a feeler gauge to measre the dstance ebtween the head carraige and the stop whei it's on cylidner 0). Refittign the parts so this clerance is the same will not, normally, be accurate enough, but it will be accurate enough for you to find the alignment track with no problem,s (you'll stepp to the right cylidner for the catseye pattern and will see it, albeit not equal-sized lobes). If you don't go back to near the original settings, you may not see the alignment track at all. The normal way to use the alignment disk is with a disk eserciuser and a 'scope. The disk exercsier basically lets you control and monitor the drive itnerface signasl, so that you can step to a given cylinder, check that hthe index sensor is working, thigns like that. There were stand-alone ones (mince is in one of the phoots) that are quite useable, or you can use the host system with the right software in a lot ofo cases. You need the 'scope to look at signals in the drive circuitry. to do the head radial alignemtn (that is, to get the heads the right distance form the spindle for a particular cylinder), younormally look at the analogue outputs of the read amplifer and have an alignm,etn trck with off-track signals under the heads. But there was a very nice unit in the 1980s called the Microtest that allowed you to do a full alignment, using the nromal analogue alignment disk, withotu a 'scope. It consisted of a box containing a microcontroller (8035 IRIC) and an ADC. It requird a IBM PC, clone just aobut any classic PC (I think the spec wassomething like 256K RAM, one RS23 2port, an video card, even MDA). You plugged the unit int othe PC serial port can connected the drive-uneder-test to the PC as drive B (there were adapters for 8", PS/2, etc drive,s which I would love to find...).Youtne ran the software suppkeid with the Microtest. This asked you to select the drive you were workign on from a menu (several hundred models IRIC), then it used the IBM line drawign chracters to display the drive PCB and told you where to connect half a dozen clips leads from the unit (grounnd, outputs of the read amplifier, index, track 0, somethign else?) You could then put the alignment disk im it To do a rdail alignemnt, for example, the unit owuld step thhe ehads to the right track and display the catseye pattern (the ADC In the microtest too the place of the 'scope). Even tell you how far off-track you were. I find it works very well, but I have no idea where to find one. Getting back to the stand-alone exercisers, some older service manuals contain the schematics fo the exerciser. I am pretty sure the manual for my Hitachi 3" drive does. And the manual for the original, full height, Sony 600rpm 3.5" drive. It would eb possible to reproduce the exerciser if toy wanted to. An older drive is likely to be a number of separte supsystems -- spindle motr, index sensor, head stepping, track 0 sensor (althoguh this will often inhibit stepping putwards), writing, reading. I therefore work through the funcions, see what works and what doesn't. Start by seeing if the spindle will rotate. On an 8" drive this is liklely to be an AC motor, running lal the time the thigh is powered. On smaller drives the motor is controlled by an interface signal, check it can be got ot run . Then I look at the index signal, Is it pulsign as the disk revolves. Does it puse and the right frequency (so is the disk turning at roughly the right speed)? Then I try to step the head around.Does the track 0 signal do the right htings. If I have a suitable exerciser I will set it to repeadedly step from cylindr 0 to the innermost cylinder and back again. Does that work? A faulty track 0 sensor can casue all srts of problem here. oOn';t asl... Then try writign and readign a test signal (moter exercises probide a suitalge signal) aon a scratch disk. if it can be read back, then te drive is basically working, althoguht it may be out of alignemnnt. But since the alignment settings are 'interchangeabllity settings' -- that is theya mean disks writtign on different drives cna be interchanged, a misaligned drive will work on a scratch disk, reading and writing its own disks. So at this point I typcially try to formate and use a scratch disk usign the host machine. If that works, I know the drive is basically OK _and_ I know it's safe to put an alignment disk in (on that note, a stcuk write gae signal wil lruin an alignment disk (it cuases the drive o overwrite the special information) and also ruin your day...). -tony