• Please review our updated Terms and Rules here

Cbm 2001 Pet strange boot

I think you ought to put the NOP generator back in and check all of the address and data pins of the UD9 ROM with your oscilloscope.

BA0 should be 250 kHz (if I remember correctly) with each higher address line decreasing in frequency by a factor of 2 (so BA1 will be 250/2 = 125 kHz).

For the data lines, just check that nothing is stuck at a '0' or '1' for the time being or is not an invalid voltage (e.g. in the range from 0.8 Volts to (say) 4.5 Volts).

Dave
 
I think you ought to put the NOP generator back in and check all of the address and data pins of the UD9 ROM with your oscilloscope.

BA0 should be 250 kHz (if I remember correctly) with each higher address line decreasing in frequency by a factor of 2 (so BA1 will be 250/2 = 125 kHz).

For the data lines, just check that nothing is stuck at a '0' or '1' for the time being or is not an invalid voltage (e.g. in the range from 0.8 Volts to (say) 4.5 Volts).

Dave
Ba0= 142Khz and i have waveform
Ba1 to Ba7= no waveform 0,2V
 
Datalines are all good except for pin 13, in this pin i have 0,5V and no waveform :(
 
Ok i checked with nop inserted and seems to be all fine except for UD9 pin 2....in this case frequency is not stable:
 
I'm thinking about offering a repair service for your board...
Bless you for volunteering. I think we are coming to the conclusion that our friend Desperado has not learned enough computer skills to fix this particular PET that has so many problems.
But as long as he wants to try fixing his computer and is learning, we will continue helping...
But we may have to step up and give daver2 a break. But truthfully, I have lost the thread after 1500 messages. We may have to get a fresh baseline and start again.
 
I think you ought to put the NOP generator back in and check all of the address and data pins of the UD9 ROM with your oscilloscope.

BA0 should be 250 kHz (if I remember correctly) with each higher address line decreasing in frequency by a factor of 2 (so BA1 will be 250/2 = 125 kHz).

For the data lines, just check that nothing is stuck at a '0' or '1' for the time being or is not an invalid voltage (e.g. in the range from 0.8 Volts to (say) 4.5 Volts).

Dave

Here's the frequencies I got with my NOP generator in place.

Address line Pin Frequency
A0 9 250Khz
A1 10 125Khz
A2 11 62.5Khz
A3 12 31.25Khz
A4 13 15.63Khz
A5 14 7.813Khz
A6 15 3.096Khz
A7 16 1.953Khz
A8 17 976.6Hz
A9 18 488.3Hz
A10 19 244.3Hz
A11 20 122.1Hz
A12 22 61.05Hz
A13 23 30.53Hz
A14 24 15.26Hz
A15 25 7.632Hz
 
Ok i checked with nop inserted and seems to be all fine except for UD9 pin 2....in this case frequency is not stable:
You mean UD2 pin 9 right ?

With the NOP running:

Do the pulses look ok on pins 23,22,21 & 20 of UD2 ?

I think you also replaced IC A3, but even so , can you check the output of IC A3 pin 10 on the scope, to see if it is an inverted (upside down version) of the pulse on pin 9 UD2 ?
 
You mean UD2 pin 9 right ?

With the NOP running:

Do the pulses look ok on pins 23,22,21 & 20 of UD2 ?

I think you also replaced IC A3, but even so , can you check the output of IC A3 pin 10 on the scope, to see if it is an inverted (upside down version) of the pulse on pin 9 UD2
No.. i mean Ud9 pin 2 Hugo.....
 
No.. i mean Ud9 pin 2 Hugo.....
Ok, sorry.

So this is buffered address line BA6 which originates from UC3 pin 12, and its input being pin 8, which is connected to the CPU's pin 15 address label AB6 .

Can you check IC UC3 to see if its inputs and its outputs match, for example compare pin 12 and pin 8. According to the post by ScottishColin, you should see on A6; (pin 15) of 3.096Khz with the NOP running (it should be stable).

Also if the connections (pcb track and UD9 socket) between UC3 pin 12 and UD9 pin 2 are ok the recording should be the same on those two pins.

Can you tell us more about the scope recording on your post #1588 and exactly what you did.

The pin 2 UD9 waveform started out low (I assume you had connected the scope up and the power was off to the board ?) then you switched on the pet, and saw those initial what looked like extra signals while the PET was in the process of a reset ? Then the waveform appeared to stabilize on the scope display, but your frequency counter did not get a stable measurement ? Is this what happened ?

Or was your PET already running with the NOP when you connected the scope to UD9 pin 2 and you saw those extra signals initially? If this was the case, does that happen every time you initially connect the scope to UD9 pin 2 ?
 
Last edited:
Ok, sorry.

So this is buffered address line BA6 which originates from UC3 pin 12, and its input being pin 8, which is connected to the CPU's pin 15 address label AB6 .

Can you check IC UC3 to see if its inputs and its outputs match, for example compare pin 12 and pin 8. According to the post by ScottishColin, you should see on A6; (pin 15) of 3.096Khz with the NOP running (it should be stable).

Also if the connections (pcb track and UD9 socket) between UC3 pin 12 and UD9 pin 2 are ok the recording should be the same on those two pins.

Can you tell us more about the scope recording on your post #1588 and exactly what you did.

The pin 2 UD9 waveform started out low (I assume you had connected the scope up and the power was off to the board ?) then you switched on the pet, and saw those initial what looked like extra signals while the PET was in the process of a reset ? Then the waveform appeared to stabilize on the scope display, but your frequency counter did not get a stable measurement ? Is this what happened ?

Or was your PET already running with the NOP when you connected the scope to UD9 pin 2 and you saw those extra signals initially? If this was the case, does that happen every time you initially connect the scope to UD9 pin 2 ?
Ok thanks, this evening i'll check!!

The pin 2 UD9 waveform started out low (I assume you had connected the scope up and the power was off to the board ?)
Yes!
 
Have you tested both the DATA and ADDRESS lines of the ROM like I asked?

Your issue with UD9 pin 2 is perhaps more to do with the scope trigger level rather than the signal itself. The signal on the oscilloscope trace seems relatively stable.

Remember what I said about adjusting the scope timebase to get at least one full cycle of the signal on the screen.

If the waveform is stable, but the scope frequency indication is not, you need to measure the frequency from the scope trace by counting the horizontal time divisions from leading edge to leading edge and converting the time measurement into a frequency by doing a 1/time calculation.

You are checking the voltage levels as well?

Do you get the same frequencies for the address pins as described in post #1,591? Although the pin numbers in this post refer to those on the CPU but you MUST do your testing on UD9.

If these measurements are good, we will then need to monitor UD9 pin 20 when we operate a RESET to see if we get some sort of activity on this pin. Perhaps use your logic probe for this test. BUT you need to be careful in taking this measurement because if the probe slips when you operate RESET we can get a false indication.

I will be going away on a business trip today, so I will write up a post describing where I think we are with this troublesome machine...

Dave
 
Yes, we still haven't found the root cause of the original instability and crashing - so this may still be a factor of course. I am hoping that finding this problem may identify that problem also (we can wish).

Dave
 
exactly ... we had reached a point where it was possible to see the basic screen, it stopped after a few seconds or minutes but at least you could see it :( Probably there was some component or ic that was failing and now it is definitely broken :(
 
Back
Top