• Please review our updated Terms and Rules here

Tape issue on IBM 5110 Type 1

V

voidstar78

Veteran Member
Joined
May 25, 2021
Messages
690
Location
Texas
I've gotten some 3M DC 6150 tapes, to try on an IBM 5110 with internal tape deck.

Results are inconsistent, so I'm looking for anyone experienced on how the workflow is "supposed" to work.

I can issue a REWIND, and it sometimes works (the tape physically spun). Other times I get ERROR 003 E80. On a "brand new" tape, is REWIND necessary? I know there are "BOT" and "EOT" markers that are supposed to be on the tape media itself.

UTIL doesn't work for me, since I assume it identifies there is no content on the tape.

If I remove the tape and try certain commands, I get ERROR 005 E80 (i.e. it is identifying the presence and non-presence of the tape)

I just insert the tape until I hear a couple light "magnetic clicks".

MARK 3,2,1 command is also giving me ERROR 003 E80.

Page 4-96 (page 218) of the MIM located here: <<http://www.bitsavers.org/pdf/ibm/51...er_Maintenance_Information_Manual_Feb1979.pdf>> states this...



1652979971005.png


Could it be the 3M DC 6150 tape is incompatible or just bad? (it's a "new stock" 150MByte 620ft version)

If the tape is all good, then should these steps work:

- insert the tape
- issue REWIND
- issue MARK 5,5,1 (5K per file, 5 files, starting with index 1) (this also format/overwrite anything that might already be there?)
- issue SAVE 4,'TEST' (probably will error if there is "0 bytes to save" - so make up some simple program first)
- issue REWIND
- issue LOAD 4
 
Last edited:
I don't recognise the error, but on my 5100, the little rollers that press against the cartridge and retain it in the drive (they have a real name but I haven't looked it up) have degraded. I have to press the tape into the drive with my thumb for things to work. Maybe give that a try?
 
I've had no trouble using DC6150 tapes in my 5100; they seem to hold the data despite the difference in the media's coercivity.
 
No impact on holding the tape (lightly pressed in). But later I'll try to inspect further inside on the internal condition of the unit.

I've found an IBM version of the DC6150 - may take a week to get here, but we'll see if it makes a difference. There are some notes in the 3M instructions about giving 8 hours for the tape to acclimate to the room conditions.

I think the Tape Status byte is a couple columns after the "5566" entries on row 5 and row 6 of the registers display.

I'm not 100% sure on which way the bits go, but I think "86" is Cartridge in place, and LED OK, and File Protected?

I don't see any external LED on the tape unit itself - I think I recall LEDs listed in the logic diagram of the tape cassette. Are these internal visible LEDs? Does it need to illuminate the tape somehow to read/write? Maybe one of those LEDs it out.


I'm not sure which way the "SAFE" dial goes (it's a small cylinder that is either "closed" or "open" in the portion exposed to the outside of the catridge).



1652993784994.png
 
The LEDs are what the tape drive uses to detect the tiny holes that mark the beginning and end of the tape. Look at your tape cartridge and notice how it has a little angled mirror on it next to the write-protect control. The light from the LEDs is supposed to come through holes in the tape, bounce off of the mirror, and shine into detectors located above the mirror.

On the 5100 (and the 5110 too I bet) the LEDs are infrared devices. You can't see infrared light of course (unless you're really talented), so use a camera or phone and peer into the tape drive. To the left of the heads, your phone ought to reveal two dim violet lights --- you may need to move the camera/phone a bit until you get it in the right place. Those are the LEDs.

My 6150 tapes are 3M tapes; they work fine.

Too bad that the simple trick of pressing the cartridge in doesn't work. I should have explained that the purpose of this is so the spindle makes good contact with the capstan in the tape. If the contact is weak and the spindle rotates without engaging the capstan, the tape drive complains. I think this is because the drive is looking for a track of regular 1s and 0s that should generate pulses as the tape goes by the heads, and if it doesn't see the pulses, the drive assumes something is wrong.
 
Ok, so peeking inside, everything looks clean and intact. I see the two infrared LEDs.

1653008124783.png


But then I noticed something out of place... It's hard to see in the prior image, but the black wire on the right isn't secured. I used a wooden chopstick to move it out a little, to get a better image. I assume this is a ground wire, which I could secure to just about any screw in the area. I'll probe around a little bit more - with the case cover off, I can get to this black wire from the outside. But it's short, not much room to work with. Anyone happen to have a photo that might show the proper location for it?

1653008347179.png
 
2-14 of the 5110 MIM (page 32 of the PDF) has this. And looks like the Tape Control Card comes out (removing it is listed as part of the service check steps). Looks like i'll be going in!

1653010318347.png
 
Well, the Tape Control Card was easy to remove. And with the A1 flipped over, you remove a couple screws to help get the wiring harness out of the way.

I think I'm going to go with this screw to secure that black wire. Any opposition? :D When I loosen that, I'm not entirely sure what will come apart - hopefully nothing drops down.



IMG_1933A.jpg
 
Ground wire secured, and so REWIND is working more reliably. However, MARK command is still not working.


The system was "unrolling" the 3M tapes completely off the reel (before the fix). So, I used one of them to examine these BOT (beginning of tape) holes. I also have a non-functional IBM tape, so I took it apart as well to compare. Here is what I found between them (at least for the beginning parts):

In regards to distances from the beginning of the tape to the double and single marker holes... (not to scale)

1653022453872.png
NOTE: One other minor point - on the IBM tape I could see "seams" consistently about every 5.2 inches. You see one of these in the single-hole comparison image right above (the one with the smaller hole on the bottom is on the IBM tape -- the seam is slightly to the right of that). On the 3M tape, I couldn't notice any seams.


The IBM distances I measured are consistent with the MIM:
1653023217462.png


When I issue the "UTIL" command, I now get ERROR 005 E80 - where 005 indicates "no content" or "empty" media (which is correct since I haven't yet saved anything).

If I set the tape to SAFE, then issue a MARK command, I now get ERROR 006 E80 - where 06 indicates that the tape is write protected (which is good, that's what SAFE mode does).

And as mentioned, REWIND is consistently returning to the first single hole on the tape. So that's good also.

But the MARK command is still giving me ERROR 003 E80.


And, both LEDs seem to be working...

1653023474977.png



Maybe it's just the extra distance the 3M tape has (60" vs 36") is throwing it off (to perform the MARK)?
 
Last edited:
I think it's unlikely to be the extra distance; again, 3M 6150 tapes work fine in my machine despite all their differences. I don't think the 5110 is too likely to be different to the 5100 in this regard.

Since the spindle seems to work, the microswitches seem to work, and the LEDs seem to work, to me that would seem to leave the tape head and the electronics behind it.

The 5100 MIM has a lot of information about maintaining the tape drive; I bet the 5110 MIM does too. Maybe there's some troubleshooting inspiration in there?

One last, silly thing: have you cleaned the tape head with cotton buds and IPA?
 
In case it helps, here's a fairly unglamorous look into my 5100's tape drive

photos.app.goo.gl

New item by Tom Stepleton

photos.app.goo.gl photos.app.goo.gl

It looks quite similar, although IBM appear to have added a metal shroud around the tape head for the 5110, maybe for EMF shielding or some such.

ETA: Looking at both pictures, I can't help but think that your tape head might be angled downward slightly? I think that the face of the tape head wants to be at a 90 degree angle to the bottom of the tape drive slot. Is it an illusion (caused by e.g. the positioning of the metal shroud) or is it real?
 
Last edited:
Well, actually - no I hadn't cleaned the head. We're talking the silver square piece near the yellow wires here? (I don't think they're tilted - maybe, and the enclosure is odd shaped - but I think the head itself is flush; it's two heads, right? a read and write one? upper/lower?)

1653093601808.png


BTW, yes I am now having to hold the tape lightly during REWIND. I don't really blame the FWD/RWD wheel (at the center above). It is making contact, but I think the real issue is there are two little plates on the left and right -- hard to see in the photo. The MIM manual stresses aligning those, or tightening them, I think to better "grip" the metal plate side of the tape - which in turn helps get more friction on that wheel. I thought about adding something like a Velcro strip to the wheel, to help get more friction - but don't think that will help. With the two roller wheels near the entrance (kind of orangish color), there isn't anything to help hold the tape in (aside from those two grip pieces on the left/right). Like, you can add/enhance the friction of the wheel, but it may just end up pushing the tape back out slightly. So yep, it needs something (like your hand) to tighten things up.

But either the wheel or those grips will wear eventually, so kind of a poor design overall :\


I wonder if just having an adjustable leather or nylon strap at the front would have helped -- set by two push-button type clasp. Sort of like this, but with a tension adjustment at the center or one end. Not sure how to secure a metal button into the plastic bezel though (and not have the plastic chip away over time). Maybe going all the way up to the top of the case would be even better. Every vintage computer ought to have some leather ;)


1653094754565.png


I'll find some cleaner and give it a shot. If the magnets are just worn, I'm not sure how to charge them back up - I have a mega-super magnet that I use to charge up screwdrivers.
 
Re cartridge retention: I think one option is just to live with it. In my case, even though all of my little programs and experiments are saved to tapes, there's not very much hardship in holding the tape inside the drive while it reads, writes, and seeks. It takes less than a minute to do its work for any operation I've needed to perform, and the computer is completely frozen in the meantime anyway. I can use my one free hand for fiddling with my phone if I'm bored.

If it did become too much of a hassle: 8-track tape players, which often have a similar cartridge retention mechanism, were occasionally subject to analogous problems. I wasn't there, but one remedy apparently was to take a paper matchbook (of the kind you used to be able to collect from restaurants) and wedge it underneath the 8-track cartridge to hold it against the heads and the spindle (what you've called the FWD/RWD wheel). I suspect a similar shimming technique would work here.


I don't think there are any permanent magnets to recharge to solve your problem. As before, it seems to me that you've isolated the issue to the tape heads or the electronics behind them, and all magnetic parts here that I'm aware of are electromagnets. Some cheaper audio tape recorders would use a permanent magnet to erase tapes while recording on them, but the drive here uses electromagnetic erase heads. Notice that the tape head assembly (which yes, is the metal thing you've identified) has two ovals on its left-hand side and two vertical features on its right-hand side: those are the read/write heads and the erase heads respectively. The reason you see two of everything is because 5100/5110 tapes have two tracks on them, an upper track and a lower track.


MARKing a tape is a complicated operation: it requires both reading and writing to work. It would be nice to know if reading works even if writing does not, or if you have a different sort of problem. Is there anyone with a working 5110 with a tape drive who could share a marked tape with you? I am not sure a 5100 tape works in a 5110 since the character sets are different, and anyway shipping to and from London could be a challenge!


If I had to guess, I'd suspect you have an electronics problem. Signals that come off a tape are very small ones: they need sensitive electronics to convert them into the logic levels that the computer uses. It may not be surprising if these start to fail or go out of tolerance sooner than other parts of the computer.
 
Cleaning those heads (with a 91% alcohol cleaner) had no apparent effect (still same issues as before).

The original IBM tape I had - I bet it had some "IMF" programs to load from the DCP. What happened on that tape is there is a thinner sort of tape used to actuate turning the spindles, it may have been made out of rubber? Anyhow, that part deteriorated and separated from the data-tape. Hmm, maybe I can feed the IBM tape onto a 3M cartridge? The data might still be ok

Well, I wasn't sure about the data-speeds. Like on the Commodore PET, a full 32K program was something like 7-9 minutes to store and reload. True, most content I'd produce for the 5110 probably won't be much over 1K. But I worry about applying inconsistent pressure during either write or reads.

Still waiting for the proper IBM tape replacement - I agree it's likely to have the same issues as the 3M tape. If that's the case, then I'll start more through the MIM procedures on checking out the components.
 
Yes, that problem is what happens to all DC300 tapes: the drive band (it goes by various names: tension band, drive belt, etc.) degrades and breaks. You have to find a replacement. The most successful method I know is to cannibalise the band from newer tapes; I've had DC6150 bands work well in my old DC300s. Feeding a tape onto a different cartridge with a working band is basically the same thing and should work fine. Respooling the IBM tape onto a new cartridge could be a good way to see if your tape drive can read a properly-formatted tape, assuming the IBM tape has retained its data and is in good condition (more on that below).

Another replacement method known to the community is to use polyurethane (not latex) bands, sometimes called "plastibands". Some find these work fine as replacements, others find that they only work for a while before they lose tension.

This issue has been discussed several times on vcfed and elsewhere. I'd recommend Googling "dc300 cartridge drive belt replacement" or similar and seeing what you get.

Looking ahead: keep in mind that you are dealing with media that is potentially over 40 years old. Some of it has started to fall apart, so another condition that you might look out for is "sticky-shed syndrome", which involves little sticky bits of the media starting to come off the tape and adhere to heads, rollers, etc. This too is a useful term to Google.


Data speeds are much better than on the Commodore. The tape cartridges and media were designed to store digital data, not audio recordings like compact cassettes were, and I assume this allows for quicker reading and writing. Either way, on the 5100 series, the tapes can only go one speed --- there is no fast-forward or faster rewind. My recollection is that spooling from one end to the other takes less than a minute.

Pressure inconsistency is not unreasonable to worry about, and holding the tape in place may be easier or harder for people for all kinds of reasons. That said, as long as the spindle is engaging the cartridge capstan without slipping, I think you will be fine.
 
By the way, what makes you think that the IBM cartridge contains original IBM data?

If it does, that could be interesting. We don't have a lot of original IBM tapes out there, so far as I know, and I haven't heard of anyone backing the tapes up to modern media.
 
In the black binder of manuals that comes with the IBM 5110, it has two plastic inserts in the back sleeve intended to hold two data tapes. The "IBM" tape I have came from there (just one, the other insert was empty). The cartridge actually doesn't have any explicit markings on it. On the back is stenciled the numbers "1229 2008", and I recall the manual referring to loading certain utilities from within the DCP (diagnostic programs, which I think it called as IMFs?). So I figured it had content like that. Or, it could have data from the original owner (a Stone company in Chicago).
 
Another possibility is that this is one of the standard tapes available with the computer's documentation, and the label has come off. An attempt to read them with the computer may tell the tale.

I have a couple of original IBM tapes, but they have a fairly advanced case of "sticky-shed syndrome". I'm still working out the strategy I'll use to read them; I think I'll only get one chance. My ideas include trying to tap the various lines coming off of the head (or any downstream amplifiers) and recording the analogue signals at a high frequency, Kryoflux-style, for later decoding.

What may also have happened to the tapes is that a user decided not to spend money on blank media and just used the IBM software tapes for their own programs and data. If the write-protect mechanism isn't set to "SAFE", I'd consider this possibility to be more likely.
 
You must log in or register to reply here.
Back
Top