• Please review our updated Terms and Rules here

Cbm 2001 Pet strange boot

YES, YES, YES!!!

I think that will do for tonight. You have done well.

A5/6 and A4/8 was a check for a single logic gate working correctly - but it was also a test for the oscilloscope and the oscilloscope operator :)!

Have you looked at the PET schematic (#1) to see what A4/8 is? It is an inverter - inverting the output from A5/6!

The channel #1 trace for A4/8 should be the EXACT inverse of the trace for A5/6. Double check that for yourself - and then I will explain tomorrow what we are looking at and why we have gone to all this trouble!

Dave
 
daver2 said:
YES, YES, YES!!!

I think that will do for tonight. You have done well.

A5/6 and A4/8 was a check for a single logic gate working correctly - but it was also a test for the oscilloscope and the oscilloscope operator :)!

Have you looked at the PET schematic (#1) to see what A4/8 is? It is an inverter - inverting the output from A5/6!

The channel #1 trace for A4/8 should be the EXACT inverse of the trace for A5/6. Double check that for yourself - and then I will explain tomorrow what we are looking at and why we have gone to all this trouble!

Dave
Click to expand...
Yes i follow schematic!! Thanks so much for your precious help! Have a good night!
 
Desperado's scope skills are improving all the time.Well done !

Now it is a matter to figure out if E9 & E10 are working properly. Passing the data in the correct direction from the CPU's data bus to Bus BD0 to BD7, and at other times from the BD0 to BD7 bus back to the CPU when E9 & E10 pins 1 & 19 are taken low respectively.

Desperado... give it some thought about how you could set up the scope to confirm that E9 & E10 are working properly, or not.
 
Sorry, I was somewhat more busy with work today than I was expecting.

I also had to go out tonight...

We will reconvene tomorrow.

Dave
 
Sorry, I have just spent the last 5 hours searching through documents at work...

I am now "dead beat" to do anything at all!

Dave
 
daver2 said:
I feel better now I have gone for a long drive to clear my head!

We will start in earnest again tomorrow...

Dave
Click to expand...
I hope you are feeling better.

One interesting thing about being in the position you are in, is that it can seem to be overwhelming, with everyone asking you questions. It can seem to add up, if you sum up all of the questions being asked at once.

If only one could focus on just the one question at a time, it would reduce the stress and the burden of it.

If you think of one question at a time, from one person, it gets much more manageable.

It is an interesting thing to consider, that it is not a burden at all to be asked the questions and be asked for help, but actually a privilege, because they ask you for help and they trust you for the right answer.

I was thinking about how Albert Einstein must have responded to many questions. If he got one from some unknown person out there, struggling with a formula or similar problem in Physics, I bet he would have stayed up all night trying to help them. This sort of thing is the response of a true professional, a Teacher and an Educator, which I think you are.
 
Ok,

Are we ready to go again?

We know that the E9/E10 buffer enable signal seems OK that is generated by A5/6. What we don't know is whether the data bus buffers E9/E10 are actually behaving themselves though.

The data bus from the CPU directly drives the PIA and VIA devices and the ROMs. The CPU indirectly drives the main RAM and video RAM via the buffers E9/E10.

At this point in time we don't actually care about the I/O devices, main RAM or video RAM. All we care about is correctly reading and executing code from ROM UD9. Nothing more, nothing less. This means that I don't actually care whether the buffers E9/E10 are broken - only that they don't interfere with the data bus when they shouldn't.

My plan is (with the NOP generator in circuit) to try and trigger the oscilloscope with one probe when ROM UD9 is selected and look for data bus contentention with the other probe.

The first thing is to NOT bother with the channel 1 signal measurement, but to concentrate on getting a stable trigger point first. Then we can take our measurements.

So, the first thing is to attach a small, stiff wire somewhere convenient on the PCB track between UD9/20 and UD2/17. Then (using your oscilloscope channel 2) with the trigger Edge set to (-)) try and get a stable trace.

You should observe a long period of inactivity (high level) as the NOP generator scans from addresses $0000 to $EFFF, and then 4,096 short, low-going pulses (no, don't count them...) from addresses $F000 to $FFFF as the ROM UD9 is accessed.

Don't forget to adjust the trigger level though.

Over to you...

Dave
 
Hugo,

Thanks for the kind words of encouragement.

I do the same thing for a job as well (keeping a large 1980s computer system running in a mission-critical application) so sometimes I just want to sit in front of the TV and vegetate at night time!

I am also trying to train a 'new generation' of young engineers to take over. I want to retire!

Dave
 
daver2 said:
Hugo,

Thanks for the kind words of encouragement.

I do the same thing for a job as well (keeping a large 1980s computer system running in a mission-critical application) so sometimes I just want to sit in front of the TV and vegetate at night time!

I am also trying to train a 'new generation' of young engineers to take over. I want to retire!

Dave
Click to expand...
Lets hope Raspberry PI's, Arduinos and a touch of Python can cope with it.

I think that you have the advantage that with machine code you are "closer to the machine".
 
I am desperate!!
The only signal that can i see on Ud9 pin 20 with Ch2 is this :(

 
That is 50 Hz mains pickup by the looks of it.

The first thing is to switch channel #1 to GND rather than DC to ensure it is not pickup from the other channel that is not connected.

Where is your probe GND connected to?

What is the peak to peak voltage (count the vertical divisions from the peak to the trough of the wave and multiply by the voltage/division setting).

Let's work the problem through - no need to get desperate. All part of the learning experience!

Dave
 
Desperado said:
I am desperate!!
The only signal that can i see on Ud9 pin 20 with Ch2 is this :(

Click to expand...
That might be good in terms of fault finding (unless you forgot to turn the computer on).

What you are seeing looks like line power borne interference, which suggests that UD9 pin 20 is open circuit (floating). If this is the case it means that there is a broken connection between the trackwork and the actual UD9 pin 20. Perhaps the socket is not making a connection to the IC pin (assuming you put the probe on the actual IC pin where it enters the IC body.

Check for continuity with the meter between UD9 pin 20 and UD2 pin 17 (SELF).
 
You must log in or register to reply here.
Back
Top