• Please review our updated Terms and Rules here

Tandon TM-100-1 problem

3

3pcedev

Veteran Member
Joined
Jun 8, 2014
Messages
735
Location
Australia
Hi all,

Just as I got my newly acquired TRS-80 model 1 clone going one of the Tandon TM-100's has decided to quit working.

Basically it was set as drive 0 and was working perfectly. NEWDOS booted up fine from this drive; read and write to the device seemed fine.

I then rebooted the system after it was running fine for 10+hrs and it started having problems reading disks and throwing out seek errors to NEWDOS. Tried reformatting the boot disk, tried other disks, same result. Finally I swapped drive 0 for drive 1 (swapped the terminator over and changed the dip switches which set the drive number) and tested it again. Drive 1 booted up fine using the same NEWDOS boot disks. This told me my disks are fine; plus the controller seems ok.

I then pulled the circuit board off the top of the drive and cleaned / lubricated the rails which carry the head. They were actually quite clean; but I decided to do the job properly. I then checked head moves smoothly back and forth and also checked the little bit of spring steel that drives it is secure on the stepper motor and the head itself. I also cleaned the head (again, not really dirty) just to cover my bases. I also checked the track 0 end stop microswitch wasn't sticking, but it was also ok.

After all that it still doesn't really work. If I try to do a DIR on a disk in the drive it seeks back and forth about 5 times (you can hear the stepper going fully forward and then the 'click' as it hits the endstop microswitch). It SOMETIMES then completes the DIR, but usually fails with a seek error. I also tried a disk diagnostic, but it failed everything from motor speed to read/write as it always failed with a 'seek error' or track 0 error. When the drive seeks it sounds (and the physical head movement looks) harsh. I tried looking for things like failed caps on the logic board however the only electrolytics seem to be for the spindle motor controller which works fine.

Finally I tried to format a disk in the drive; the format completed but the verification failed immediately at track 0.

I have been through a plethora of references on the internet; but I have found most people have problems with the alignment. I don't think this is the case here because it cannot format / read / write a disk. From what I have read if you format a disk in a mal-aligned drive it should work in said drive (but no others). This isn't the case here.

Any ideas?
 
I'd swap the drive controller boards, assuming you have another Tandon as your 2nd drive. Try booting with both drives again. This will help you determine if it's the controller or the drive mechanics. Keep swapping parts until you find the bad one.

Could be you have a burned out cap on the controller board. You can also compare voltages between the working and non working.

In short, use the working drive as a comparison against the non working.

b
 
Tried swapping the controller boards over - no change.

I did some testing with ImageDisk today. The results are.....

When running the alignment check tool the drive gets 'out of step' with the current track. For example with my working drive if I select track 30 the output will show 30..... If I then select 0 it will output 0 etc.... With the Tandon if I select track 30 it may show 29, 28, 25 etc. When it first powers up it also shows track 0 as track 1 which leads me to believe there may be an end stop problem. There is definitely no problem with the head assembly sticking in its runners, so failing that it probably is the stepper motor itself.
 
I've seen the same thing on relatively new (90s) 3.5" drives that were brand-new. In my case it was the white lithium grease on the positioner leadscrew--cleaning it off with some electrical contact cleaner made the drive seek correctly. Since the TM-100 uses a taut-band/capstan setup for positioning, you really need to look ever little part over for hardened grease or dirt.
 
Chuck(G) said:
I've seen the same thing on relatively new (90s) 3.5" drives that were brand-new. In my case it was the white lithium grease on the positioner leadscrew--cleaning it off with some electrical contact cleaner made the drive seek correctly. Since the TM-100 uses a taut-band/capstan setup for positioning, you really need to look ever little part over for hardened grease or dirt.
Click to expand...

You'd never guess, even though the collar was 'tight' on the capstan it was still slipping on the stepper motor shaft. After much swearing due to releasing the 'taut' band arrangement I managed to fix the collar (it was over compressed and the compression 'gap' needed to be filed out a bit).

Now it seems it's a bit out of alignment which isn't surprising. Can I get your advice regarding which adjustment is required:

- In imagedisk the drive now reads and seeks consistently (using a microsoft windows floppy as a reference); but everything is 2 tracks out. For example if I seek to track 20 I get track 18, track 40 is 38 etc. Track 41 is in fact track 39 and track 42 has no data as you would expect.
- Track 0 is kind of reading as track zero, but with some errors. Track 1 and 2 don't work (just outputs ???) and then track 3 onwards are perfect.

Do you think this is a track zero endstop adjustment problem or a radial alignment problem? Thanks!
 
Assuming that you didn't move the track zero sensor, I'd start by rotating the stepper motor body first--it's a very touchy operation. I'm sure there's a maintenance manual on bitsavers to give definitive instructions--I also seem to have seen a Sams Photofact floating around for the TM-100.
 
Chuck(G) said:
Assuming that you didn't move the track zero sensor, I'd start by rotating the stepper motor body first--it's a very touchy operation. I'm sure there's a maintenance manual on bitsavers to give definitive instructions--I also seem to have seen a Sams Photofact floating around for the TM-100.
Click to expand...

Unfortunately I have moved just about everything now. The Tandon technical manual isn't very clear about aligning the zero stop switch so I sort of 'went it alone'. I was able to get the drive to find track zero intermittently by adjusting the physical stop, but it bounced on the stop sensor. I then adjusted the sensor and it then stopped bouncing, but then started stopping on track 2.

The radial alignment screw is a real pain. If I set it so the drive always finds track zero then every other track is 2 tracks out. If I set it on the other tracks (in combination with the zero stop) then track 0 can't be found reliably. It's like adjusting 3 variables where none of them are fixed (radial, track 0 physical end stop, track 0 stop switch).

I finally was able to get it to line up perfectly on all tracks, but again it always wanted to bounce on the zero stop when trying to find track zero. This caused it to only find track zero correctly 50% of the time and found track 2 the other 50% (but when it found track 0 I could seek to every track perfectly).

I just googled SAMS Tandon TM100 and found a photofacts. It goes about things almost exactly like I tried, but with a bit more precision (i.e. back out this screw this much when X happens). I'll give it a shot sometime in the next day or two. Failing that I will have to obtain an alignment disk (SAMS says it's mandatory) and use a CRO. The CRO I have.... the alignment disks seem like unobtainium these days though.... My feeling is imagedisk is a very very crude tool at this kind of work.
 
Last edited:
You can get a bit more precision by recording a 360K (48 tpi) image using 1.2MB drive. However, there's no subsitute for a real alignment disk.
 
I have to admit I'm far away from being an expert in floppy disk drive alignment. Nevertheless the topic is very interesting and I have done a little online search myself.

Do you have seen Terry Stewart's article "Adjusting the radial alignment a Tandon TM100 -2A drive without an alignment disk"?

Alignment disks are available from Accurite Technologies Inc.
Accurite's "Floppy Drive Alignment Test Parameter Summary".

Alignment disks are an expensive tool if needed only one time. Regarding the economical aspect it may be an alternative to buy an other Tandon drive if the alignment attempts without an alignment disk are failing. A Tandon TM100-2A (double sided, 40 tracks) can be relatively easy acquired on internet auctions.

-Rainer
 
Well that was an epic experience; I now have gotten the drive aligned and working well.

Thanks to everyone here who contributed; I managed to piece together what to do by taking a bit of information from every source. For reference the sources were:
Micro 80 Magazine Sep 84'
Terry Stewarts Tandon alignment article
SAMS Tandon TM100 alignment fact sheet
Tandon TM100 Technical Manual

In short the capstan came loose from the stepper motor drive. It then slipped when stepping between tracks which resulted in the drive constantly going out of alignment. I tightened the capstan up however the position in which I tightened it up was in a non-integer step position relative to the tracks on the disk. As such trying to align the drive using the alignment screw did nothing useful.

To summarise here is how I realigned the drive:
  • Loosen the capstan collar lock screw and ensure the capstan can 'slip' on the stepper motor shaft
  • Carefully drive the head to track zero by pushing the head as far rearward as it will go
  • Using imagedisk (or a CRO) start the drive up and set it to track zero using a 'good' formatted disk (I.E. Something from a software supplier like microsoft etc).
  • Make sure the head is still touching the track zero stop. At this point the CRO or Imagedisk should indicate the drive is reading track 0
  • If the drive is not reading track zero adjust the track zero stop until it (roughly) does
  • Lock down the capstan on the stepper motor
  • Start the alignment procedure again - i.e. quit imagedisk and start again. This is imporatant as the drive needs to find track 0 by itself now.
  • Align the drive as per SAMS instructions - start with the track 0 switch, then the track 0 stop, then the radial alignment. All 3 were out on mine.
  • NOTE: I used a CRO and Imagedisk simultaneously to do the last step. Imagedisk tells you what track you are on, the peak to peak value on the CRO assists with the alignment as per Terry Stewarts article.

The drive now works perfectly and there is no hammering the track 0 switch/endstop when initially accessing the disk. Good times all round!
 
Chuck(G) said:
Happy to hear that. Personally, I detest TM100s--compared to other drives, they weren't made particularly well--but they were cheap.
Click to expand...

Agreed. I especially think the track zero switch arrangement is the dumbest thing I have ever seen. It's mounted to the drive body, not the head assembly. As such you move the head assembly to adjust the alignment which puts your track 0 switch in the wrong spot. Of course the lock screw for it is under the disk so you need to remove the alignment disk, loosen it, put the disk back in, adjust it, take the disk out, tighten the lock screw and then start all over again because it drifted....
 
You must log in or register to reply here.
Back
Top