CBM 8032 - What does the startup chirp actually mean?

[dave_m]

So the level is noisy +4.78 Volts. Not shorted low. Now we have to determine if the 6845 is being programmed correctly by the CPU or if there is still an issue on one of the data lines. We might need daver2 to whip up some test code. Also have we tried a new 6502 CPU?

[daver2]

Is the test with the original PET EDIT ROM in service or my PETTESTER?

If you install my PETTESTER EPROM into the EDIT ROM socket - we know exactly what should be sent to the CRTC and we can work from there.

Another recent thread had a faulty ROM that was upsetting data line D0. Any data line fault would corrupt the initialisation of the CRTC and give us a 'funny' picture at best. I am not saying this is what it is - but just one candidate. Although this was a pretty specific fault and never seen before - so the probability of it being similar on your machine should be quite low.

Dave

[MikeP]

Hi Daver2, the scope reading is with your PETTESTER EPROM.

Hi Dave_M, Sorry, I misinterpreted the scope output as being negative voltage, (Still a lot to learn). I have not tried a new 6502, I have a 6510 from a Commodore 64, but is that compatible? If not, I can buy a 6502 - are the new ones OK or do I need to buy vintage?

Thanks Guys !!!!!

[dave_m]

Hi Dave_M, Sorry, I misinterpreted the scope output as being negative voltage, (Still a lot to learn). I have not tried a new 6502, I have a 6510 from a Commodore 64, but is that compatible? If not, I can buy a 6502 - are the new ones OK or do I need to buy vintage?

Mike,
I'm not familiar with your scope, but I assume the L on the parameters on the right indicate the average level of 4.87 V. Also the yellow pip on the right of the plot may be the ground line. There is a lot of noise on this signal. How are you grounding the probe?

I would you to get a vintage 6502 from Commodore, Rockwell and equivalent. I am not familiar with the Western Deign Center chips. There may be a version those that is compatible with the old NMOS +5 V types. Someone here may know.

[MikeP]

Hi Dave_M, The L is the trigger level and the yellow pip is also the trigger level displayed on the output. The level of the Vertical Drive signal is closer to +5.13V (4.87V was where I had the trigger set).

I had the probe grounded on one of the J4 Ground pins, but I also grounded it on J7 Pin 7 which did not make any difference. I will look for and buy a vintage 6502.

Thanks
Mike

[dave_m]

Hi Dave_M, The L is the trigger level and the yellow pip is also the trigger level displayed on the output. The level of the Vertical Drive signal is closer to +5.13V (4.87V was where I had the trigger set).

OK, thanks for the correction about the display. However the trigger lever is set way too high. For TTL it should be set about 2 V. You were triggering on noise.

[MikeP]

Hi Dave_M,

Here is the scope reading with the trigger at 2 volts. I had to zoom out in order to set the trigger that low, so now you cannot see the noise.

Corrected_Trigger_at_2_volts.png

[dave_m]

Definitely no skinny negative pulses. Apparently the 6845 is not being correctly initialized despite us seeing some chip selects. Let's ponder the next step...

[daver2]

I am guessing your oscilloscope doesn't have any logic analyser capability?

What we need to do is to capture the writes to the CRTC registers and see what the data bus has on there at the time.

What I am thinking of is a simple program that lives in the EDIT ROM (replacing my PETTEST) that outputs the values $00 to $FF to the CRTC address register continuously.

This won't program the CRTC at all - but it will permit us to see the 'state' of the data bus.

This is just a thought at the moment - need to think how to trigger the oscilloscope so we only see the data bus during the CRTC WRITE cycle and not at any other time. Possibly when CRTC /CS (pin 25) is LOW. D7 should then indicate a 'slow' clock, with D6 being twice as fast and so forth down to D0 being the fastest.

The other possibility is to output $00 followed by $FF or $AA followed by $55.

Thoughts?

This would be my first attempt (will only work with BASIC 4 though).

A9 00 8D 80 E8 38 69 00 4C 00 E0

ORG $E000
LDA #0
STA $E880 ; CRTC
SEC
ADC #0
JMP $E000 ; LOOP

Dave

[dave_m]

I could burn the binary data into a 2516 and ship it to Mike.

This code will increment the data bus and test the CRTC address register, but it will not exercise the odd address ($E801) needed to write data to the 6845 registers in case that is the issue. Should we add that? Or maybe you do not want to write nonsense into the 6845?

we could sync scope to R/W low and chip select. Or if we want to initialize (and trust) the VIA, we could pulse a line on the User Port for the scope. I wish we could read the 6845 registers. That would be a simple check.

P.S. Shouldn't the jump be to $E802?