• Please review our updated Terms and Rules here

IBM 5150 hates floppies

SquallStrife

Experienced Member
Joined
Aug 24, 2011
Messages
84
Location
Brisbane AU
Hi guys,

I've got an IBM PC 5150, and I'm having floppy disk trouble. I have two Tandon TM100 drives (a -2A and -4) that both display the exact same problem.

The first symptom is that it won't read floppies that are known to be good. You can occasionally get a directory listing, but if you try to copy or otherwise read a file, you get "Sector not found", and it won't go any further.

The second symptom is that even though it will get 100% of the way through a format and display the message "Format complete." without any problems or head-seeking sounds, at the end it says "Unable to write BOOT" and fails. This process can be repeated with any number of known good floppy disks, and every time it's the same. Even if I try "format a: /u /1" to just format one side.

The system contains an Adaptec MFM controller and a Seagate ST-238R, which is what I'm booting from. It came with MS-DOS 5.00 installed. The symptoms are identical for both floppy drives, and I don't have any other 5.25" floppy drives to test with. (Well, unless you can use an Apple II drive! :p)

Any ideas?

Found this thread by LGR/phreakindee, but there doesn't seem to be any resolution at the end? I've got a few things to try now though...
 
Last edited:
TM100-1 / 48 TPI / 1 side / 40 tracks <------------------- Used in IBM 5150 for 160K and 180k floppies
TM100-2 / 48 TPI / 2 side / 80 tracks (40 per side) <----- In IBM 5150, adds 320K and 360K capability

TM100-3 / 96 TPI / 1 side / 80 tracks
TM100-4 / 96 TPI / 2 side / 160 tracks (80 per side)

TM100-3M / 100 TPI / 1 side / 80 tracks
TM100-4M / 100 TPI / 2 side / 160 tracks (80 per side)

The -2A is what one normally sees in an IBM 5150 (read/writes 160k/180k/320k/360k disks).
Are you using double density floppies for that ?

The -4 is equivalent to a 1.2M drive, a high density drive. The IBM supplied floppy controller [photo] does not support high density. The 5150 BIOS has no support for high density. For this drive to work at 1.2M, an upgraded floppy controller (with BIOS) is required.
 
Are you terminating the floppy drives properly? :

One drive on cable: Drive to be at end of (twisted) cable and drive terminated.
Two drives on cable: Drive at end of (twisted) cable is terminated - drive in middle of cable is NOT terminated.
 
The -2A is what one normally sees in an IBM 5150 (read/writes 160k/180k/320k/360k disks).
Are you using double density floppies for that ?

Yes, definitely. And DOS is saying "Formatting 360K" as well, so there isn't any format weirdness.

The -4 is equivalent to a 1.2M drive, a high density drive. The IBM supplied floppy controller [photo] does not support high density. The 5150 BIOS has no support for high density. For this drive to work at 1.2M, an upgraded floppy controller (with BIOS) is required.

My floppy controller looks a bit different to that one.

This is the one I have: http://www.vintage-computer.com/vcforum/showthread.php?20617-PC-5150-Diskette-Adapter

But you're right, and I only expect it to work with 360KB floppies.
 
Last edited:
My floppy controller looks a bit different to that one.
This is the one I have: http://www.vintage-computer.com/vcforum/showthread.php?20617-PC-5150-Diskette-Adapter
Yes, there are two versions. In post #1 of the thread that you refer to, the top controller is the one supplied in early 5150s. The bottom controller is the later version, supplied in later 5150s, and in 5160s (and in 5155s). Both controllers are functionally the same.

It sounds like you may have an early 5150. Is your motherboard the "16KB-64KB" version, or the later "64KB-256KB" version? The motherboard is labelled as such. Details on the early 5150 are at http://www.minuszerodegrees.net/5150/early/5150_early.htm

If you have the "16KB-64KB" motherboard, what is the 7 digit part number on chip U33, the BIOS chip?
 
But you're right, and I only expect it to work with 360KB floppies.
I expect that DOS itself won't be able to read a 360K floppy in the TM100-4 drive. That's because one of the things that needs to happen in order to read a 360K floppy in a 1.2M drive is for double-stepping to occur (i.e. compensate for a floppy with tracks laid at 48 TPI within a 96 TPI drive), and the 5150 BIOS (used by DOS) won't have double-stepping code.
 
It's fitted with 256KB of RAM, and the motherboard is marked as such. It is a two-screw case, but it's been retrofitted with a 146W power supply.
 
The -4 is equivalent to a 1.2M drive, a high density drive. The IBM supplied floppy controller [photo] does not support high density. The 5150 BIOS has no support for high density. For this drive to work at 1.2M, an upgraded floppy controller (with BIOS) is required.

No, the TM100-4 is equivalent to a 720K drive. That is, the spindle rotates at 300 RPM and there is no high-density mode. So the 5150 controller will work on the drive, but the result won't be readable on a TM-100-2 48 tpi drive. If you want to double-step the drive, you'll have to use a special driver. Or just use it to handle 720K 96tpi disks. I did that for a couple of years using a Micropolis 1115-VI 5.25" 96tpi drive for my B: drive before I got a hard drive.

FWIW, a Tandon TM-100-4M is yet another beast--it's 100 tpi, 77 track drive. Even more bizarre.
 
Last edited:
It's fitted with 256KB of RAM, and the motherboard is marked as such.
"64KB-256KB" or just "256KB". It's hard to tell from your answer as to whether you have an IBM 64KB-256KB board, or a clone board marked "256KB".
It's important that we know what we are dealing with (thus possible limitations/problems).

No, the TM100-4 is equivalent to a 720K drive. That is, the spindle rotates at 300 RPM ...
Thanks for the correction Chuck.
 
"64KB-256KB" or just "256KB". It's hard to tell from your answer as to whether you have an IBM 64KB-256KB board, or a clone board marked "256KB".
It's important that we know what we are dealing with (thus possible limitations/problems).

Ah OK. Sorry.

Yes, the marking is "64KB-256KB CPU".

Well, the TM100-2 just let out the magic smoke. One of the tantalum caps has a big black hole in it now. Lame. Going to have to wait til tomorrow now.

Thanks for your help thusfar! :)
 
Yes, the marking is "64KB-256KB CPU".
Okay. That board will have the third and final BIOS revision. No issues with 360K drives/floppies at all.

Regarding the TM100-2A.
Presumably you have already checked termination per post #3.

A common problem seen with 5.25" floppy drives that have not been used in years is deteriated lubrication on the head assembly rails. Have you tried re-lubrication of the rails?
 
Okay. That board will have the third and final BIOS revision. No issues with 360K drives/floppies at all.

Regarding the TM100-2A.
Presumably you have already checked termination per post #3.

A common problem seen with 5.25" floppy drives that have not been used in years is deteriated lubrication on the head assembly rails. Have you tried re-lubrication of the rails?

Yes, cleaned them with a cotton swab in isopropyl, then applied some silicon spray the same way.

Regarding termination, there is a metal bar (looks a bit like a staple) in the third position in socket 1E, and a blue IC in socket 2F.

I just replaced the popped tantalum cap, and the -2A drive is alive, kinda.

When I try to format a disk now, it says "Invalid media or Track 0 bad - disk unusable"

I really hope that when the capacitor went, it didn't take anything else with it!!
 
, there is a metal bar (looks a bit like a staple) in the third position in socket 1E,
1E is the options socket. That 'metal bar' (in third position) is setting the drive at the second Drive Select. That's what we expect.

and a blue IC in socket 2F.
2F is the terminator socket.

When I try to format a disk now, it says "Invalid media or Track 0 bad - disk unusable"
If you are using floppies that you used earlier in the TM100-4 (i.e. formatted earlier to 96 TPI), then that could be a cause.
Try 'bulk erasing' them if you have the facility.
Otherwise, do you have some 'virgin' double density floppies to try?
 
Floppies formatted at 96 tpi low-density (720K) shouldn't bother a 48 tpi drive any more than other 48 tpi low-density floppies. The write currents used are identical; the track width is just half.
 
If you are using floppies that you used earlier in the TM100-4 (i.e. formatted earlier to 96 TPI), then that could be a cause.
Try 'bulk erasing' them if you have the facility.
Otherwise, do you have some 'virgin' double density floppies to try?

Yeah, it doesn't matter what disk I put in, formatted or otherwise, the message is the same.

Crazy thought, would there be any point to transplanting the PCB from the -4 drive on to the -2 drive? I'm pretty convinced at this stage that when the capacitor popped, it took some other stuff with it. If I did that, would I need to swap over any of the ICs? I'm reasonably sure the -2's mechanism is OK, it does seek back and forth when the system powers up.

I guess a way forward is to connect up a 3.5" floppy drive and use 720K floppies...

Edit: According to this thread, I can use a 1.44MB drive with 720KB disks in my 5150 and it should work. That's a relief. I even have a black 3.5" drive I can use to keep the aesthetics...
 
Last edited:
Crazy thought, would there be any point to transplanting the PCB from the -4 drive on to the -2 drive?
The PCBs may be different, e.g. to possibly to cater for a different model stepper motor, one that perhaps requires extra current.

I'm pretty convinced at this stage that when the capacitor popped, it took some other stuff with it. If I did that, would I need to swap over any of the ICs?
If other damage occured when the capacitor blew, we are not in a position to tell you which components are damaged. It's not predictable. Such damage would vary from drive to drive, requiring individual diagnostic work.

I guess a way forward is to connect up a 3.5" floppy drive and use 720K floppies...
Edit: According to this thread, I can use a 1.44MB drive with 720KB disks in my 5150 and it should work.
I've done it myself.

It would be a useful thing for you to do anyway, because you have yet to rule out the IBM floppy controller as being a contributor to the problem.
 
If other damage occured when the capacitor blew, we are not in a position to tell you which components are damaged. It's not predictable. Such damage would vary from drive to drive, requiring individual diagnostic work.

That's fine, I appreciate that. I guess I just need to go over it with a multimeter and see if there are any shorts.

It would be a useful thing for you to do anyway, because you have yet to rule out the IBM floppy controller as being a contributor to the problem.
[/quote]

That's what I thought too. I connected a 720KB floppy drive to the IBM controller and got nearly the same results as I began with. Occasionally able to get a directory listing, unable to read files.

So I inserted a 16-bit ISA multi-I/O card into the system, with the HDD, Serial, and Parallel portions disabled, and it worked right away. I was able to copy the contents of a 720KB floppy to the hard drive first go, flawlessly.

Obviously the PC only has 8-bit expansion slots, but the controller seems to work just fine (or at least the floppy portion of it).

I'm now going to connect the 360K drive and see what happens.

Edit: No change with the Tandon drive. 720K drive works an absolute dream though, formats, reads, writes, all that fancy stuff. Just had to add DRIVPARM= to config.sys to make it format properly. I wonder if somebody would trade one working drive for two "possibly-easy-to-fix-for-someone-with-the-right-diagnostic-equipment" drives?
 
Last edited:
So I inserted a 16-bit ISA multi-I/O card into the system, with the HDD, Serial, and Parallel portions disabled, and it worked right away. I was able to copy the contents of a 720KB floppy to the hard drive first go, flawlessly.
Obviously the PC only has 8-bit expansion slots, but the controller seems to work just fine (or at least the floppy portion of it).
Add the 2M-XBIOS driver and you should end up with 1.44M capability.
See http://www.minuszerodegrees.net/transfer/35_inch/2m-xbios.htm
 
Obviously, your experimentation now suggests that your IBM floppy controller is faulty (as well as the TM100-2).
Some things worth trying are:

1. A decent clean of the controller's edge connector.

2. Your first generation 8-bit IBM floppy controller uses -5V (the later generation does not). Fitted to a 64KB-256KB motherboard, it is probably the only thing in your 5150 using -5V. If you have the ability to measure DC voltages, check the -5V. Also, it's a possibly that there is a poor -5V connection in the P9 connector from the power supply. It is worth reseating P9 a few times.

p8_p9_pinout.jpg
 
Back
Top