• Please review our updated Terms and Rules here

Cassette interface issue: can't load

C

Christoffer

Experienced Member
Joined
Oct 15, 2014
Messages
299
Location
Denmark
I'm having a problem with a 6802 based SBC i recently bought. More specifically, I can't load from cassette.
I have no problem recording to audio tapes, but when I enter the "load" routine, nothing happens.
I can play back the tape all I want, but there's neither a "it worked" promt, or a "fail" message, as there should be.
I've listened to the tape, and I can hear both the 30 second beginning and the end sound, as well as the data package between.

It's a very simple Kansas City Standard interface, and the hardware is little more than a buffer and a high pass filter on the "ear" / receive line.

Has anyone had similar issues? I'm a complete noob with cassette storage, so I don't know of any "tricks" - Could my cassette deck be playback-ing at a different speed than recording?

Any info, kinks and quirks would be much appreciated!

--Christoffer
 
Are you using something like a portable cassette deck?

Panasonic_RQ_2102_RQ_2102_Portable_Cassette_Recorder_684454.jpg


Make sure your volume isn't too high or too low. My ZX81 is extremely picky about the volume level.
 
One very much like it, a DUX (Phillips brand?) one.
It has both a "tone" and a "volume" - Am I correct in that the tone doesn't matter as long as you don't change it between save/load?
Maybe I should just try at many different volumes.

--Christoffer
 
Christoffer said:
Am I correct in that the tone doesn't matter as long as you don't change it between save/load?
Click to expand...

The 'Tone' control only ever affects what happens on playback, not record. Recorders generally try to record what they are recording 'dry' (flat, with no processing) except in the case of systems like Dolby where the high frequencies are intentionally boosted during record and reduced by the same amount on playback. That won't be an issue with this simple cassette recorder.

In general for these retro computers the signal you want from the tape is a mid to high frequency tone or pair of tones around 1-2 Khz so things will generally work best with the tone control set to the high / trebly end.

Incidentally, I like your sig comment about Rainbow ribbon cable - so much nicer than the modern, depressing, flat grey stuff.
 
Cheerful wiring is a widely overlooked design criteria in a lot of products.
Okay, so I found one problem, at least: I was recording from the absolute start of the tape, missing the first 2 seconds or so to the blank, uncoated tape. D'oh.
Anyway, same problem persists. I'll try the treble-y tone setting. The sound out line is just driven by a MC14584 inverter, and that's what picks up the sound on the "in" line.
Should I match the line out/input signal strengths? - The manual states "generally, louder is better" but that doesn't sound very technical.

Other possible issues:
- The inverter driver could have a dead gate.
- The 6821 PIA input in use could be dead (unlikely, since the entire rest of the chip works (it drives the displays and keypad)
- The "recieve" routine in ROM could be corrupted (does these things happen or did I just make it up?)
- My cassette recorder is crap and/or noisy

Should probably try saving to a computer, and playing back to eliminate the tape recorder as the culprit.

Another possibility entirely, is that the cassette interface is really bad and more of a token feature than something expected to be used. Again, never heard about anybody using this SBC before.

Thanks for the interest!

--Christoffer
 
Okay, I suppose the interface isn't necessarily poorly designed, it is VERY similar to the Heathkit ET-3400 expansion module one.
And as that also use a 6802, I think I'll try finding a known good ET-3400 program sound file and play that to the cassette input.

--Christoffer
 
Christoffer said:
Cheerful wiring is a widely overlooked design criteria in a lot of products.
Okay, so I found one problem, at least: I was recording from the absolute start of the tape, missing the first 2 seconds or so to the blank, uncoated tape. D'oh.
Click to expand...

Easily done, at least you worked it out so that's one less thing to worry about.


Christoffer said:
The sound out line is just driven by a MC14584 inverter, and that's what picks up the sound on the "in" line.
Click to expand...

... This is two different gates in the one MC14584 package, I take it? It is entirely possible for just one gate in a logic IC to be damaged, especially if that gate is exposed to the outside world either directly or indirectly. It's also perfectly possible for just one port pin on a VIA / PIA / PIO to be damaged and for the rest of the device to be working perfectly.

Christoffer said:
Should I match the line out/input signal strengths?
Click to expand...

Usually, no. The cassette-out signal is normally designed to go into the MIC input of a recorder. The signal generated by a dynamic microphone is tiny, just a few millivolts, so the cassette output signal is usually only a few millivolts as well, so it seems to the recorder as though the sound is coming from a microphone.

It could be that in this case your cassette interface outputs audio at a much higher level designed to go into the line-in input of a recorder, but that would be unusual because, while virtually all portable cassette recorders had an input socket for an external microphone, relatively few of them had a line-level input as well.

Conversely, the input on the tape interface is expecting the audio to be coming from an audio power amplifier which would normally be driving the internal speaker in the cassette unit, so the expected input level to the cassette interface IN input will be much, much higher than the audio level it outputs from its tape-out output.

Have you got a scope or a logic probe available to you? It would be interesting to see what's on the line between the inverter and the PIA port pin at various volume settings while the tape is playing.

Christoffer said:
Should probably try saving to a computer, and playing back to eliminate the tape recorder as the culprit.
Click to expand...

I've had some success with doing this in conjunction with a ZX81, although the output drive level from the audio-out on most modern computers is usually so low that you may have to put a small audio amplifier between the computer output and the tape interface input. I experimented with an LM386 - based amplifier, which worked OK.
 
... This is two different gates in the one MC14584 package, I take it?
Click to expand...
Yes, it is. I've attached the schematics of the interface, going to the PIO - If something had to take damage from the outside world, it'd likely be the first chip in the path, I'd guess.

The guy who previously had this removed the jack sockets for some DIN-connector, with a switch bypassing the attenuator, letting one switch between line and mic out.
The standard is "-long leading tone" "data package" "long end/standby tone" as I recall, but just before the leading tone, there's a very short, and very distorted sound, is there some identifying letter it's not recording properly? It isn't every time I hear it.

Anyway, I'l rip out the buffer IC and test it. - For the sake of originality, no IC's are socketed - would you socket the replacements?

Thanks again!

--Christoffer
 
Christoffer said:
would you socket the replacements?

--Christoffer
Click to expand...

Good question.

As a technician, ICs in sockets make my life much easier just as today's surface-mount equivalents make my life an utter misery, since it is no longer possible to change a chip in three seconds.

However, since this system of yours is an interesting / historic item, and since you obviously have the confidence and skill to remove and refit the IC without doing any damage my answer in this case would be that it would be better to conserve it in its original form, so if its ICs weren't originally in sockets, I wouldn't fit them.

That said, I broke exactly that rule when I repaired a couple of completely different problems on my Atari ST recently, each one caused by the failure of two different gates within the same six-gate logic IC. In both cases the gates which had failed were directly exposed on external connections on the monitor and HDD ports respectively, so I judged that the likelihood of repeat failure was quite high and fitted the chip, originally directly soldered, in a socket to speed things up next time.

The 'third way' would be to fit a socket while you are trying to diagnose the problem and then, when you've got it working, remove the socket and refit the IC directly to restore the system to its historically correct appearance.

Re: recording, the long lead-in tone you've described is probably for the purpose of informing the recorder's Automatic Level Control what the audio level of the data itself will be. By the time the computer starts sending out the actual data, the ALC has adjusted itself to the correct recording level while the lead-in tone was in progress.
 
I always install sockets. The way I see it, it's 'period correct' anyway. Someone would have installed a socket on day one if they had to.
 
Both valid points. If I bought a microcomputer kit in the early 80's, and it didn't come with sockets, I'd definitely add them in myself, and put the expensive IC's in as the absolute last state.
But again, I have a (mid-70's) vacuum desoldering station, so it isn't eating a whole lot of time to suck out even a 40 pin IC.

Good to know it's a cassette thing. I thought it was actual data being transmitted, I'm not very familiar with the Kansas city standard format.

--Christoffer
 
The cassette interface input has a 0.01uF decoupling capacitor, feeding a lowpass filter (cutoff around 100 KHz) - presumably to avoid crosstalk from other busses.
It also has a switch, feeding the output from the tape transmit connector back into the receive. This was omitted on my board (wires directly to chassis connectors)
Does blue ceramic capacitors die a slow death like their electrolytic cousins?
 
Just a really "back to basics" question. Is your tape deck mono and not stereo? Also is your audio cable mono ? Stereo cables used as mono don't always connect very well with a mono jack and audio jacks may have a switch built in.
Another problematic part of a tape recorder is the complex internal record / play switch, a long sequence of switches. Spraying cleaner in that switch solves a lot of issues.

Larry G
 
I've used both mono and stereo cables, no difference. Maybe I should clean the tape recorder, though. - Think the tape deck is stereo.
 
If the tape deck is stereo, be sure it is only recording/playing on one channel, I think Left channel was recommended for mono. What will happen if you record the mono output of the computer to both L and R simultaneously, when you play it back on older tape decks the phase between the L/R channels can actually cancel some of the higher frequency signal, unless the deck actually has a stereo/mono switch. Signal is handled differently than a mono tape recorder shown by NeXT
 
Last edited:
That's interesting, but, if you put a mono cable in the stereo jack, it must short the "ring" channel to ground, eliminating that problem. Though, I'm not entirely sure I used "true" mono cables. I'll check that out! Thanks!
--Christoffer
 
Very sure, All ceramic capacitors on the board are of that blue type. And I have the schematics, they're definitely caps.

One feature I don't understand is the loopback between MIC and EAR, connected to the EAR's jack switch. If there's no connector in the EAR (input) jack, it closes that switch, sending MIC /output back into the input line. - That functionality was disabled by a hardware mod by the previous owner. My MIC line on the board is by the way always transmitting a tone, even when the "record to tape" routine isn't running, don't know if that's normal or not.

-Is it possible that the interface only works if you have ONE tx/rx connected to the tape deck? and not both? - If you recorded without a EAR connection, it'd also play back to the board itself, maybe it has a "verify" routine? - The manual is very unspecific about the whole cassette thing.

--Christoffer
 
You must log in or register to reply here.
Back
Top