PET 2001-8 Datasette Problems

[Holmes]

Hello All. I am trying to bring my PET 2001-8 back up to health, and have been working through some memory and stability problems in another thread.

I have discovered load problems in doing that, mainly with the internal datasette. I thought I would start a new thread for this problem.

So basically, when I try to load with LOAD and no arguments, it says SEARCHING but never says FOUND and never gets to LOADING. When I got this PET, this datasettebelt was broken, so I replaced it by pulling one from a donor C2N (rounded one). It seems to work and turns the spindles fine visually. I have cleaned both heads with alcohol.

So I also have another external C2N (rectangular one), and can load and save from this one, plugged into the cassette input plugs 1 and 2 on the PET. I have a reference tape (Time Trek), that works on the external c2n.

I also tried plugging the internal datasette to port 2 to see if that changes anything, and still no difference. I also used an electronic demagnetizer tape, and no difference.

I suspected it might need azimuth adjustment. I tried recording a program and reading it, and it didn't work. The cassette is confirmed working on my external c2n.

So now if I need to adjust the azimuth, how do I do that? I have a scope, but don't really know what I would look for. I tried googling for images of someone doing this, but couldnt find anything i thought was relevant.

I found a c64 program someone wrote to provide a visual signal to help adjust the head. I've also found a link to an old compute! Article where they suggested just listening to a playback and adjusting it to where it sounds best. Don't I also need to use a reference tape also?

Anyone with any experience with any of these methods or other words of advice?

 

[dave_m]

I suspected it might need azimuth adjustment. I tried recording a program and reading it, and it didn't work. The cassette is confirmed working on my external c2n.

So Port 1 cannot READ its own recording? I'd start tracing the Cass#1 READ signal from the Cassette Port (J6-4) to PIA#1 at G8-40. It should be a TTL level signal out of the tape unit.

 

[Holmes]

So Port 1 cannot READ its own recording? I'd start tracing the Cass#1 READ signal from the Cassette Port (J6-4) to PIA#1 at G8-40. It should be a TTL level signal out of the tape unit.

Hi dave_m. Thanks for the reply and suggestion.

So I believe my cassette port circuitry on the motherboard is working okay, because I tested my working external C2N datasette on the Cassette Port 1 (on the left side of the motherboard inside the PET where the internal datasette is normally plugged in) and on Cassette Port 2 (in the back), and that datasette player works fine at either port. This led me to think it's the internal datasette player mechanism that is not working. I also removed the internal datasette player and plugged it into Cassette Port 2 and it still did not work - doesn't pick up any programs on tape and just sits there at the "SEARCHING" until I RUN-STOP.

KC9 corrected my understanding of the heads, so I begin to understand your suggestion more. Sorry.

I will hook up the scope to it at the next opportunity and check the voltages I'm seeing. Thanks!

 

[KC9UDX]

Are there really separate heads? Usually there's one for R/P and one for erase.

Have you tried demagnetising the head?

Alignment could be off but that's rare. The speed could be off, too, but that's rare also. Usually it's Magnetic Fields, a dirty head, or an electronic problem.

You have of course verified that the tape is moving, yes?

 

[Holmes]

Are there really separate heads? Usually there's one for R/P and one for erase.

Have you tried demagnetising the head?

Alignment could be off but that's rare. The speed could be off, too, but that's rare also. Usually it's Magnetic Fields, a dirty head, or an electronic problem.

You have of course verified that the tape is moving, yes?

Hey KC9!

So I see two lumps that come out when you hit the record+play buttons, and I presumed these are the heads. One is white (the head on the left), and the other is silver (the head in the middle). Not sure which one does what, but I cleaned both. I just searched online, and now I see that the middle head is the read/write head, and the white one is the erase - thanks for the correction. Now the comment above by dave_m makes sense to me. Sorry dave_m - it seems I didn't even understand your suggestion. I'll go back and modify my response.

I demagnetized the head with an electronic cassette head demagnetizer. This is a vintage one I purchased on ebay with a circuit board in a cassette case. I changed the battery on it, and verified that the red LED in it turned on for the specified period of time while playing the cassette mechanism, and instructions said to do it for one second, and I did it for about two. I have cleaned the heads very thoroughly with cotton swabs and 99% isopropyl alcohol, switching tips several times.

I have verified the cassette is turning, and it looks to be normal speed to me, but I'm far from an expert. Can't think of how I would test this, other than putting in some cassette and playing the output and listening to it and seeing if it sounded slow. Not very knowledgeable about if hooking up some headphones to the output line would destroy anything due to a mismatch of the impedance or something like that, especially since this is not for its intended use.

I did switch out the drive belt on this datasette for one from a rounded C2N datasette, and it seemed to be in good condition and it seems to fit and work well. I thought I found online that these internal belt drives were interchangeable between the commodore datasette models.

 

[KC9UDX]

Here's what I'd do: save a program to tape, same program but two different tapes, one in each drive. Play the tapes back in an audiocassette player, and see how they compare. Report what you hear; with a good ear this can make the repair obvious.

Also, one sticky wicket with cassette recorders is the multi-pole switch usually used to convert the unit from being able to play a tape to bring able to record a tape. If that switch is dirty, operation in either mode can be unpredictable. Usually they can be fixed by carefully spraying tuner cleaner into the switch and working it rapidly for a while by hand.

 

[dave_m]

KC9 corrected my understanding of the heads, so I begin to understand your suggestion more.

No, you are right the problem seems to be in the internal cassette player. Follow KC suggestions and see it it's repairable. If not, I wonder if the external C2N can fit in the PET? They are still plentiful.

 

[Holmes]

Here's what I'd do: save a program to tape, same program but two different tapes, one in each drive. Play the tapes back in an audiocassette player, and see how they compare. Report what you hear; with a good ear this can make the repair obvious.

Also, one sticky wicket with cassette recorders is the multi-pole switch usually used to convert the unit from being able to play a tape to bring able to record a tape. If that switch is dirty, operation in either mode can be unpredictable. Usually they can be fixed by carefully spraying tuner cleaner into the switch and working it rapidly for a while by hand.

OK, will do. Thanks!

It may not be until this weekend when I can so any more testing. Hmm, it sounds crazy, but I don't have a regular cassette player anymore! I think I've got an old Walkman around here that hasn't been used for about 20 years! I've also got an old Radioshack computer cassette player which I got for my IBM 5150.

 

[Holmes]

No, you are right the problem seems to be in the internal cassette player. Follow KC suggestions and see it it's repairable. If not, I wonder if the external C2N can fit in the PET? They are still plentiful.

OK, thanks dave_m! They seem to be about the same shape and size (at least the rectangular ones). No way with the rounded ones. I'd really love to keep the PET as original as possible, though, at least externally.

 

[KC9UDX]

Are the gubbins interchangeable between the round and rectangular ones?

Anyway, a Walkman will certainly work. That Radio Schack one very likely has a headphone jack, so you could use that, too.

Barring that, you can get one of those cassette to 8-track adapters. Funny that cassettes were once so ubiquitous that there were everything you can imagine-to-cassette adapters, but cassettes only can be adapted to 8-track players.

In any event, cassette players are fairly easy to diagnose and repair. Much easier than a PET, or a 2040.

 

[Holmes]

Are the gubbins interchangeable between the round and rectangular ones?

Anyway, a Walkman will certainly work. That Radio Schack one very likely has a headphone jack, so you could use that, too.

Barring that, you can get one of those cassette to 8-track adapters. Funny that cassettes were once so ubiquitous that there were everything you can imagine-to-cassette adapters, but cassettes only can be adapted to 8-track players.

In any event, cassette players are fairly easy to diagnose and repair. Much easier than a PET, or a 2040.

I thought so. The drive belt seems to fit and the spindles seem to turn okay. I seem to recall that it didn't seem extremely tight when I got it in there.

I need to dig out the RadioShack cassette player. Last I checked, it didn't quite work with the IBM 5150, so I'm not sure if I can trust it.

 

[KC9UDX]

You don't have to trust it. There are only two outcomes:

You hear nothing, in which case you know the RS deck is totally dead (unlikely).

Or, you hear something from one tape, and something else from the other.

 

[Holmes]

You don't have to trust it. There are only two outcomes:

You hear nothing, in which case you know the RS deck is totally dead (unlikely).

Or, you hear something from one tape, and something else from the other.

Ok, I'll give it a go and report back this weekend. Thanks!

 

[Holmes]

Hey All!

So I'm back and did some more testing this morning.

I only have one working cassette that works reliably on these datasette players, so I recorded a sample program save from the working external C2N player, and then recorded a similar program save immediately after it on the INTERNAL datasette. Both were plugged into the external datasette port on my full functional PET 2001-32N.

So, after recording the program samples for both datasettes, I tried loading them.

Here are the results:

For all testing, I attempted to load both saves via both players. Each time was to the PET 2001-32N, and each time was through the same cassette port.

C2N recording: attempted loading via C2N player. It WORKED. This was expected, since the C2N player works fine.
C2N recording: attempted loading via INTERNAL datasette. It did NOT work. All it did was sit at "SEARCHING".

INTERNAL datasette recording: attempted loading via C2N player. It did NOT work.
INTERNAL datasette recording: attempted loading via INTERNAL player. It did NOT work.

So it looks like the INTERNAL datasette does not load its own recordings, NOR does it load other working recordings.

So as recommended, I used the "ear" test. I played the tape recordings back on my Walkman. The C2N recording sounded like a normal computer recording. The INTERNAL datasette recording, by contrast, sounded much quieter, and more "tinny". This would definitely explain why the INTERNAL datasette recordings didn't work while attempting to be loaded via the C2N player.

So additionally, I connected the Walkman to my laptop, and recorded the cassette recordings.

I loaded them into Audacity, and looked at the levels, and as expected, the INTERNAL datasette recordings are MUCH lower than the working C2N recordings.

I cannot attach the .wav files, but did upload the screenshots from Audacity. I can provide the .WAV files if anyone is interested.

This shows both waveforms zoomed out. The top two are the C2N recording, the lower two are the INTERNAL datasette recording:


This shows the range I was going to zoom in on (it's a narrow gray section in the middle, showing I am zooming on the loudest part of each waveform):


And this is the zoomed range, showing the C2N levels much higher than the INTERNAL datasette recordings.



So it looks like my INTERNAL datasette levels are much lower when recording, and perhaps something similar is going on when it attempts to playback a working recording.

I have tried demagnetizing it, and cleaned the heads with 99% isopropyl alcohol. Any ideas on how I can get this INTERNAL datasette player working again?

 

[KC9UDX]

It could be grossly misaligned, or you could have a bad head. I'd rule out electronic issues for now, only because it can neither read nor write.

Take a look at the tape head in each one, with, and without a tape installed with the head out. You may be able to see if the head in the nonworking one is out of whack. If it is, you can probably find alignment with a logic probe or oscilloscope.

Otherwise, I'd be inclined to swap heads and see what happens. But that may not be feasible.

 

[MikeS]

It could be grossly misaligned, or you could have a bad head. I'd rule out electronic issues for now, only because it can neither read nor write.

Take a look at the tape head in each one, with, and without a tape installed with the head out. You may be able to see if the head in the nonworking one is out of whack. If it is, you can probably find alignment with a logic probe or oscilloscope.

Otherwise, I'd be inclined to swap heads and see what happens. But that may not be feasible.

Lots of schematics and other docs on Zimmers:
http://www.zimmers.net/anonftp/pub/cbm/schematics/datassette/index.html

Looks like all versions have separate read and write electronics (as opposed to switching the same electronics back & forth) so yeah, sounds like it might be a problem with the head, either electronic or mechanical (not moving far enough, azimuth way off, etc.) or, less likely, power.

Maybe work through the troubleshooting section of the manual and, if your two drives use similar electronics, compare the readings at the head or first amp between both drives playing the same tape.

Good luck!

And Happy New Year to all the Commie fans!

:bigparty:

 

[Holmes]

Thanks for the input guys.

So I opened up the internal datasette, and here are some pictures. Nothing looks odd to me. I checked the mechanism operating some of the keys. Pressing record with play presses on a long, funky slide switch, which seems to be pressed in fully by a metal tab. There are 2 electrolytic caps on the PCB, and they looked fine under magnification (no bulging, no residue pooled at the base, etc). There are 2 ceramic caps which don't look too good to me (white residue on it and the insulation around one of the legs looks "pulled up"), but I'm not sure what a bad ceramic cap looks like. My impression was if they failed it was because they cracked, and that you couldn't see this as it's under the coating.













The spring under the azimuth screw looks good still, but I didn't turn anything. Nothings stands out on the head itself.

I looked at the troubleshooting guide in the manual, and it didn't look too helpful to me. Basically tells you to check the leads, and something like "is the head closed" which I don't understand. The head mechanism seems okay when you push the buttons.

I checked out the manual on zimmers that MikeS linked to, and on page 10 (http://www.zimmers.net/anonftp/pub/cbm/schematics/datassette/cas-10r.gif), in the footnotes for the schematic, it says "the BIAS should be +1.7 +- 0.005V at 5V power supply" - what does this mean? I looked quickly online and it says "bias" is the base voltage for an alternating signal. So does that mean if I check the 5V lead, and it's not close to +1.7V, then this might be my problem? I think I understand this concept on a signal line (the signal oscillates around this voltage), but I don't understand how you could correctly have less than +5V at the 5V terminal.

I might try the head alignment program using my C64, as that seems to make the process pretty easy. Seems strange that an alignment issue would cause this, as I would think then it could read its own recordings (as dave_m suggested earlier). Unless the head itself is just bad (as KC9 suggested).

 

[KC9UDX]

If it's far enough out of alignment, it won't clearly read nor write, because the track will be off the tape, or, muddying itself as it records.

I was going to suggest checking the bias, but I'm not sure that will come into play in your situation. But it can't hurt to check it. I'm not sure what they're trying to tell you, but you can usually measure the bias right at the tape head leads.

That big slide switch usually switches the head between the record and play circuitry. Those can be problematic, and it may need to be cleaned. You can check it first with an ohmmeter based on the schematic. For some reason I thought you already did that, but I must be thinking of someone else in another thread.

 

[KC9UDX]

Something I forgot to mention earlier, is that if swapping heads is a problem, you can still do it, without actually swapping heads. You can desolder the head leads on both units and run wires between them to cause the head in one to be connected to the electronics in the other one. You'll have to have power to both of them, and press play on both units, but it sounds like you can do that.

I've done that in the distant past as a way to be certain that a head is bad before tackling replacement in a complicated transport (usually read "eight track").

 

[Holmes]

If it's far enough out of alignment, it won't clearly read nor write, because the track will be off the tape, or, muddying itself as it records.

I was going to suggest checking the bias, but I'm not sure that will come into play in your situation. But it can't hurt to check it. I'm not sure what they're trying to tell you, but you can usually measure the bias right at the tape head leads.

That big slide switch usually switches the head between the record and play circuitry. Those can be problematic, and it may need to be cleaned. You can check it first with an ohmmeter based on the schematic. For some reason I thought you already did that, but I must be thinking of someone else in another thread.

OK, I see.

So to check the bias, I just put the ohmmeter on the tape R/W head leads and press play with no cassette in there, right? OK, that makes more sense than putting them on the +5/GND connections.

I seem to remember you mentioned something about a large switch that changed the cassette player from play to record, and this was waaay back when I started on this adventure, so that may be why you thought I already checked it. I guess that's another thing on my list to check.

Something I forgot to mention earlier, is that if swapping heads is a problem, you can still do it, without actually swapping heads. You can desolder the head leads on both units and run wires between them to cause the head in one to be connected to the electronics in the other one. You'll have to have power to both of them, and press play on both units, but it sounds like you can do that.

I've done that in the distant past as a way to be certain that a head is bad before tackling replacement in a complicated transport (usually read "eight track").

This sounds really cool. I can't believe it would work, but I guess there's no reason for it not to. It sounds like an elegant way to isolate just the head but still run it through the datasette electronics. Again, cool.

I still have the "donor" C2N datasette which I took the drive belt from, so I can also use this as the read/write head donor as well, if that turns out to be my problem. Hopefully the parts are interchangeable. Cassette heads are pretty generic parts, I would imagine.

Lots of other things to look into. Thanks!