• Please review our updated Terms and Rules here

Commodore 3016 scrambled screen

Well... Now i switched on my 3032. It was working and second time today when i switched it on i get this screen.
So this one is broken again. Did the NOP test. All ok. Used Pettester, but this screen is only screen i get.
 

Attachments

  • IMG_9744.jpg
    IMG_9744.jpg
    881.4 KB · Views: 7
If the NOP generator works, but PETTESTER doesn't, I would use the NOP generator and check that the ROMs and the video RAM are being selected via the 4:16 address decoder.

Dave
 
What do you observe on the ROM data lines?

Do you see 'sensible' TTL signals or any bus contention?

Do you also see the video RAM and video buffers being selected?

Dave
 
You will not be able to get them stable I am afraid. What you are looking for are 'strange' voltages >0.8V and <3.5V.

This is where 'gut feeling' and experience comes into play...

Dave
 
I think he means on any data or address line (probably more so the address line). Looking for voltages outside the TTL operation window so that they can cause random effects. >0.8 and <3.5V aren't a 0 or a 1 so the IC interpreting them can choose at random.
 
>>> Do you mean the voltages on the datalines?

Yes. Either a device cannot drive a data line properly, so you get a 'strange' voltage, or multiple devices drive the data lines concurrently. Where a data line is driven both high and low at the same time, you can also get strange voltages.

The strange voltages only last a short while as the concurrent devices are selected and then deselected.

Dave
 
I check D0-D7 and this is what i get: (with NOP)
D0 2.6-3V
D1 3,6V
D2 3V
D3 2,5V
D4 3V
D5 2V
D6 3,2V
D7 4.1V

The datalines are 'jumping' from 0V to the Voltage off the datalines i mentioned
 
Some of those data lines look a bit low to me (especially D5). D5 is within the invalid area.

However, before we move on, can you explain exactly how you measured the 2 Volts.

Dave
 
Please do NOT use a multimeter. The reading is irrelevant.

Use your oscilloscope to measure the voltage - that is the point of the oscilloscope (it displays voltage against time).

If you measure the voltage using a multimeter - all it is doing is 'averaging' the HIGH and LOW voltages based upon the time the signal is HIGH or LOW for.

I am obviously not explaining the process well am I...

The oscilloscope should be indicating a voltage corresponding to a HIGH or LOW. If two things are driven at the same time, you will observe a voltage that is not a HIGH or a LOW. We are trying to observe these events with the oscilloscope.

Dave
 
A PET is available with two (2) different types of video circuits:

1. Without a Cathode Ray Tube Controller (CRTC) chip.

2. With a CRTC.

The former type will give a random display (as you have) on power-up. This indicates that the majority of the VDU circuitry is working.

The problem is either the CPU not working (no clock, no reset, or a faulty CPU), the CPU halting (due to faulty ROMs or the inability to select or read the correct data from the ROMs), or the inability of the CPU/ROMs to successfully write data into the video RAM.

We have to be quite religious, in this case, identifying what the issue could be, by ruling out what it can't be.

If the PET has a CRTC, the likelihood is that the screen will be black on startup due to the requirement for the CPU to program the CRTC to get anything sensible out of it.

Since you are having problems measuring things on the data bus, can I suggest reverting to my PETTESTER, and we can work through the potential scenarios one at a time?

Let me know when you have done this.

Dave
 
Excellent.

Use your oscilloscope and check for a 1 MHz clock on pin 39 of the 6502 CPU.

Check that pins 2, 4 and 6 of the CPU are permanently HIGH.

Check that pin 40 of the CPU is LOW when you turn the power on to the PET and goes HIGH after about 0.5 to 1 seconds.

Use your oscilloscope and check for pulses on pin 7 of the CPU.

Report back.

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