• Please review our updated Terms and Rules here

Commodore 3016 scrambled screen

OK, but you didn't get any errors - that is the main thing!

The longer you run it, the more thoroughly you thrash the DRAM...

So the DRAM problem turned out to be a duff H4 - yes?

Dave
 
You need to look at the HDRIVE and VDRIVE signals with an oscilloscope and see whether both of them - or just VDRIVE - is jittery on the oscilloscope.

My guess would be only the VDRIVE signal is affected - in which case it gives us a place to look - being H8 and G10 on http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/2001N/320349-6.gif (but this is pure supposition at this moment until you get us a measurement).

Are any of these ICs socketed I am asking myself?

The other thing I observed on your 'jumping jack video' is that not all of your ROM checksums are returning correctly! ROM D for example is returning 0000!

Do you have all of the correct ROMs in the PET (apart from the EDIT ROM socket that contains my PETTESTER) and do the checksums match what is in my PETTESTER manual for the version of BASIC you are running?

Try not to replace the ROMs too many times, you may (read will) damage the IC sockets... Leave the PETTESTER in place and get as much information from it as possible - especially the DRAM memory test, ROM checksums and keyboard test.

If a ROM is dead - this will cause the BASIC to enter the debug monitor. Also, if PIA C7 is dead - this will cause the same behaviour. PIA C7 is also involved with the keyboard. So testing the keyboard with the PETTESTER is a good idea...

Dave
 
Last edited:
>>> With replacement rom i see the normal screen. (only jumping)

Excellent. Getting better!

This is what I mean about using the PETTESTER tool to the maximum and clearing as many faults out of the system in one go.

I only noticed that myself when you posted a picture of the jumping screen.

>>> I will test H8 and G10 tommorow.

Can you look at the HDRIVE and VDRIVE signals first though?

Dave
 
The vertical jumping is pretty erratic. Any small phase disturbances in the V drive can cause this and the bouncing you see is because of decaying oscillations caused by the positive feedback around the vertical scan amplifier.

In this case though, I wonder if it might be something really simple we have seen on the Forum before that disturbs the vertical scan amplifier. A noisy wiper connection on the Height pot on the VDU board. Spray it with some cleaner and exercise it a little and see if that has any effect. One clue is that there are fairly significant V scan amplitude (height) disturbances.
 
@Hugo Holden, back in post #131 (if I understand correctly) the monitor electronics has been swapped and the problem still persists. On that basis, the assumption is a bad connection on the main logic board (or wiring harness?).

Dave
 
To be honest, I am finding it difficult to see any disturbance on the oscilloscope trace for either the VDRIVE or HDRIVE signal.

I would have chosen to use a slower timebase - to get a lot of cycles on the oscilloscope screen. That would help (in my opinion) to observe any disturbances.

The HDRIVE trace indicates a stable 64 us 15.6 kHz signal.

The VDRIVE trace has this information absent - as there is not a single cycle available to the oscilloscope (based upon the timebase setting) for the oscilloscope to compute the value from.

Can you point out what you think is a disturbance?

This is not something as simple as a duff solder joint on the connector to the monitor (from the main logic board) or a bad connection in the interconnecting cable is it?

Dave
 
Last edited:
@daver2 I see what you are saying about the swapped VDU, but in the case of the jumping that has been video recorded, there are large changes in the V scan amplitude even to the extent that the opening Commodore text is sitting in a lower than usual position. Look at the scan height in post #152, I don't think that is likely caused by a defective V drive unless its amplitude was changing wildly and it would ve dead easy to see on the scope. There might have been two different "jumpings", but the one in the video looks to me that it is likely in the VDU and likely the height pot or a physical connection issue with the v drive signal. It looks like the height pot is the most likely culprit, since we have also seen this before.

The other thing that can do it is if a UVeprom is used instead of the original character ROM and it drags down the /INIT line, but that just makes the scan bounce, not change height.
 
Last edited:
Don't forget the video connector itself (J7).

Is this what the video display screen looked like when it was 'normal'?

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