My Klyball 600D build

[Thomas]

I would suggest to test, if the cassette/laptop playback options are working as intended.
Attached a file to "LOAD" under BASIC. It goes for a while and shows, if the data can be read correctly.

test.JPG (12.03 KiB) Viewed 13517 times

If characters are missing or wrong, you can adjust volume or other parameters until in scrolls without any hick-ups. The center letter should increment every line. If this is working, than any other data loaded into the 600D build should work.

Here the BAS and WAV file for experimenting
Thomas

CASS_TEST.zip

(184.85 KiB) Downloaded 234 times

[ronin47]

I would suggest to test, if the cassette/laptop playback options are working as intended.
Attached a file to "LOAD" under BASIC. It goes for a while and shows, if the data can be read correctly.

test.JPG

If characters are missing or wrong, you can adjust volume or other parameters until in scrolls without any hick-ups. The center letter should increment every line. If this is working, than any other data loaded into the 600D build should work.

Here the BAS and WAV file for experimenting
Thomas
CASS_TEST.zip

That's cool, thank you

I'll get back with the results.

[ronin47]

I've been playing around with this emulator: https://www.pcjs.org/machines/osi/c1p/

At least I now know what I should see on screen when stuff is loading.

I'll respond back after I've had some more time to test.

[ronin47]

OK, so nothing is working as I would expect it to going by the results from the emulator.

Although my pulse width is correctly set up according to the service manual, all I get when I try to load is garbage on screen (Random characters).

I think what I need to try next is adjusting the pulse width to different spots in the range to see if that helps, and if not, then it's back to the schematics to see what else could be faulty.

[ronin47]

OK, so tweaking the pulse width upper and lower down the suggested range made no difference.

I've double checked the values of the caps at C11 and C55 and they are fine, as are R53, R57 & R63a.

When testing with the method suggested at in the Service Manual (Continually printing U to screen with pins 9 & 10 connected to set pulse width) I see this:

IMG20240610114401.jpg (3.64 MiB) Viewed 13474 times

The yellow trace is pin 5 of U69 (Pin 1 of U63 matches), the blue is pin 2 of the 6850 (U14) and is the same as is seen at pin 5 of U63.

Similarly this trace is when trying to load the cassette test program:

IMG20240610114522.jpg (3.8 MiB) Viewed 13474 times

Does this look correct, or does the blue trace seem wrong (It does to me as not all the pulses arrive at the 6850)

If anybody has any thoughts I'd be grateful.

[bxdanny]

Since you are connecting the tape output (AUX, J2 pin 9) to the tape input (J2 pin 10), the signals when running that test program won't make a lot of sense unless the output side is working.

I'd suggest adding a couple of POKEs to the test program to produce a continuous 150 Hz square wave (1/300 sec low, 1/300 sec high) at the TxData output of the ACIA, winding up with something like this:

10 SAVE
20 POKE 61440,21
30 PRINT "U";
40 POKE 14,0
50 GOTO 30

[The POKE in line 20 sets the ACIA to produce one stop bit (8N1) instead of two, and the POKE in line 40 keeps the character POSition cleared to 0, preventing a carriage return and linefeed (with 10 nulls in between) from being inserted after every 72 characters.]

You can then follow the signals around from TxData (6850 pin 6) to RxData (6850 pin 2), where the original 150 Hz square wave should be reproduced. I'm not sure exactly what all the intermediate signals should be, but I know the audio signal at that link between J2 pins 9 and 10 should show 1200 Hz when TxData is at logic 0, and 2400 Hz when it is at logic 1.

[ronin47]

Since you are connecting the tape output (AUX, J2 pin 9) to the tape input (J2 pin 10), the signals when running that test program won't make a lot of sense unless the output side is working.

I'd suggest adding a couple of POKEs to the test program to produce a continuous 150 Hz square wave (1/300 sec low, 1/300 sec high) at the TxData output of the ACIA, winding up with something like this:

10 SAVE
20 POKE 61440,21
30 PRINT "U";
40 POKE 14,0
50 GOTO 30

[The POKE in line 20 sets the ACIA to produce one stop bit (8N1) instead of two, and the POKE in line 40 keeps the character POSition cleared to 0, preventing a carriage return and linefeed (with 10 nulls in between) from being inserted after every 72 characters.]

You can then follow the signals around from TxData (6850 pin 6) to RxData (6850 pin 2), where the original 150 Hz square wave should be reproduced. I'm not sure exactly what all the intermediate signals should be, but I know the audio signal at that link between J2 pins 9 and 10 should show 1200 Hz when TxData is at logic 0, and 2400 Hz when it is at logic 1.

Cool thanks, I'll give that a try.

[ronin47]

OK, so I did what you suggested bxdanny and these are some of the results.

In all of these, pin 6 of the 6850 is the yellow trace, and we can see 150Hz

This is the input to U64, pin 2, blue trace:

IMG20240610162311.jpg (4.19 MiB) Viewed 13452 times

It's been through an inverter, but it's 150Hz as expected.


Next up is pin 11 of U64, so the output:

IMG20240610162449.jpg (4.21 MiB) Viewed 13452 times

I believe we have 1200hz for logic 0 and 2400Hz for logic 1. However the blue trace is slightly shifted over by a small amount.

Finally this is the output of the op amp on the tape in side:

IMG20240610160420.jpg (3.8 MiB) Viewed 13452 times

This looks OK but I now need to trace it through to pin 2n on the 6850, see what that looks like.
And again this looks ok, though mildly shifted to the right again. I believe we have 1200hz for logic 0 and 2400Hz for logic 1 here too.

[bxdanny]

The shifting is normal, I was expecting it actually. But I wonder what those sharp spikes are, and if they make any difference. I guess they are well out of the frequency range of what a tape could record anyway, so I think you should be able to record tapes now that others will be able to load. But if anyone can explain those spikes, please do.

Good luck now with the input side. You didn't show the end result at the 6850 input (pin 2), but I guess it doesn't yet match the output (pin 6), apart from a slight delay, which is what it should do. If it did, you'd be able to load tapes.

[ronin47]

OK, well that's been fun and games.

I've ended up doing a swap around for different brands of U62, U63 and U69!!!

Not because they didn't work, but they didn't work together. It was lucky I had a few spares of all of them in different makes.

They all test fine in my TL866 though. Big thanks to @bxdanny for the test program that generates the 150Hz square wave, it's been very useful..

This is the result:

IMG20240610223507.jpg (4.23 MiB) Viewed 13440 times

Yellow is the output (Pin 6) of the 6850 (U14) and blue is the input (Pin 2) of the 6850.

This looks better, and while the blue is shifted, as you say @bxdanny, it should be shifted.

There's more news too, this was an attempted load from my digital voice recorder, to which I also saved the program direct:

IMG20240610225927.jpg (6.92 MiB) Viewed 13440 times

It loaded partially

It was odd, the nine of 'A's came in before I pressed play, but with the digital recorder disconnected I see no such 'noise'

Now it hung the machine, but I think the issue now is the crappy digital voice recorder I have, so I'll try again tomorrow with a better audio source and some known good wav files.

I think I may need a little adjustment, but this is the first real progress I've had getting any data back into the machine, and it does look like it may be saving too