- VCF South West - June 14 - 16, Davidson-Gundy Alumni Center at University of Texas at Dallas
- VCF West - Aug 2 - 3, Computer History Museum, Mountain View, CA
- VCF Midwest - Sept 7 - 8 2024, Schaumburg, IL
- VCF SoCal - Mid February 2025, Location TBD, Southern CA
- VCF East - April 2025, Infoage Museum, Wall NJ
-
Please review our updated Terms and Rules here
Cbm 2001 Pet strange boot
- Thread starter Desperado
- Start date
ajgriff
Experienced Member
Correct so now you know what you're looking for. 
Alan
Alan
Desperado
Veteran Member
- Joined
- Nov 25, 2017
- Messages
- 6,827
yes but where can i looking for this fault?Correct so now you know what you're looking for.
Alan
ajgriff
Experienced Member
You're the one with the board infront of you. Time for you to decide on the next steps in the light of the comprehensive advice you've been given.
Alan
Alan
daver2
10k Member
As Alan has stated, we know what it isn’t - it isn’t a faulty UD2 (as we had thought all along) so turn off the soldering iron!
It is one of two things (in order of probability):
1. A partial short circuit (or tracking) from the PCB track associated with UD2 pin 9 to somewhere else. Despite your statements to the contrary, I believe this to be the more likely scenario.
2. One of the two inverter gates connected to UD2 pin 9 has gone faulty. This scenario is much less likely than point 1.
So, back to the oscilloscope and let’s be more upbeat (and less desperation) about our fault finding.
Dave
It is one of two things (in order of probability):
1. A partial short circuit (or tracking) from the PCB track associated with UD2 pin 9 to somewhere else. Despite your statements to the contrary, I believe this to be the more likely scenario.
2. One of the two inverter gates connected to UD2 pin 9 has gone faulty. This scenario is much less likely than point 1.
So, back to the oscilloscope and let’s be more upbeat (and less desperation) about our fault finding.
Dave
Hugo Holden
Veteran Member
The scope will help you for a while. You can re-solder pin 9.
You are looking to have the scope CH1 on pin 9 to get the most stable lock you can and using ch2 as your probe. The two places to probe around are the jumper block, especially its right hand side where the pin 9 connection enters at P and around the IC's pin 1 and pin 2 A4 and pins 10,11 & 12 A3, to see if there is an interfering signal that matches what you see on Ch1(pin 9,D2).
It raises the question as to whether it is likely a gate, like A4 for example could develop signal cross talk between two inputs pin 1 and pin 2 alone, it doesn't happen that I have seen.....yet, not impossible but a rare one, more often the gates go open.
There is one situation though, where you might not find a matching interfering pulse because , under a certain circumstance, it could be composed of all , or a mix, the pulses feeding all the IC pins of IC's A4 and A3:
IC's inputs can appear to have severe interactions of all inputs, if the +5V power supply to the IC has failed, or become disconnected.
But, we know that the IC A4 is powered by 5V because:
1) you have checked that 5V supply and
2) it runs from the same +5V regulator as U2......still it could pay to check it again at the IC pins, where they enter the IC body with the meter just in case of a dry solder join to pin 14 of A4 or pin 14 of A3.
It is possible that inside a chip package, the IC die could lose its connection to +5V power and that would also do it and create a failure mode of interacting inputs.
So , if, with the scope, you cannot find an exact single matching interfering pulse, and IC's A4 and A3 have definite 5V power on their pin 14's, as the pin enters the IC plastic body, then I would suspect that either IC 3 or IC4 had lost power to its die internally.
The practical way to tell is to just solder suck and free Pin 11 of A3 and check the interference, if ok solder it back up and then check Pin 1 of A4 the same way.
I would guess if it is not IC' A3 or A4 in some of the ways above being defective, then likely it is corrosion etc under the switch block.
You are looking to have the scope CH1 on pin 9 to get the most stable lock you can and using ch2 as your probe. The two places to probe around are the jumper block, especially its right hand side where the pin 9 connection enters at P and around the IC's pin 1 and pin 2 A4 and pins 10,11 & 12 A3, to see if there is an interfering signal that matches what you see on Ch1(pin 9,D2).
It raises the question as to whether it is likely a gate, like A4 for example could develop signal cross talk between two inputs pin 1 and pin 2 alone, it doesn't happen that I have seen.....yet, not impossible but a rare one, more often the gates go open.
There is one situation though, where you might not find a matching interfering pulse because , under a certain circumstance, it could be composed of all , or a mix, the pulses feeding all the IC pins of IC's A4 and A3:
IC's inputs can appear to have severe interactions of all inputs, if the +5V power supply to the IC has failed, or become disconnected.
But, we know that the IC A4 is powered by 5V because:
1) you have checked that 5V supply and
2) it runs from the same +5V regulator as U2......still it could pay to check it again at the IC pins, where they enter the IC body with the meter just in case of a dry solder join to pin 14 of A4 or pin 14 of A3.
It is possible that inside a chip package, the IC die could lose its connection to +5V power and that would also do it and create a failure mode of interacting inputs.
So , if, with the scope, you cannot find an exact single matching interfering pulse, and IC's A4 and A3 have definite 5V power on their pin 14's, as the pin enters the IC plastic body, then I would suspect that either IC 3 or IC4 had lost power to its die internally.
The practical way to tell is to just solder suck and free Pin 11 of A3 and check the interference, if ok solder it back up and then check Pin 1 of A4 the same way.
I would guess if it is not IC' A3 or A4 in some of the ways above being defective, then likely it is corrosion etc under the switch block.
daver2
10k Member
Yes, internal loss of the 5V supply on one of those ICs would cause utter 'carnage'. It would be worth checking where the other inverter gates of the same package are used. Looking at the output pins of the other gates in each package may identify a similar issue. This may narrow the search down a bit?
Dave
Dave
daver2
10k Member
Read Hugo's Post above...
He is telling you exactly what I would be looking for next and how to go about it.
All you have to do is read and do a bit of research.
Dave
He is telling you exactly what I would be looking for next and how to go about it.
All you have to do is read and do a bit of research.
Dave
Hugo Holden
Veteran Member
Desperado, you have similar troubles with hardware problems that I have with software ones. But, in this case you are very close to solving this particular problem.
If you have a list of what to do, it will probably work out for you, to at least get past this problem so you can move on with the rest of the repair.
So follow this list:
1) check with the meter pin 14 of IC A3 and A4 both have +5V on them where the pin enters the plastic IC package. If not find out why and fix that. If that is ok move to step 2:
2) Solder suck and free pin 11 of A3 (make 100% sure it is disconnected with the meter). Check with the scope on pin 9 of UD2, is the interference gone ? if so replace IC A3. If not re-solder the pin 11 and go to step 3:
3) Solder suck and free pin 1 of A4. (make 100% sure it is disconnected with the meter) Check with the scope again on pin 9 UD2, if the interference is gone ? if yes replace IC A4. If the interference is still there, re-solder pin 1 of A4 and go to step 4:
4) Since at this point if the checks above were passed, it would appear that D2, A3 and A4 are all ok, and really the only "likely" thing left is corrosion under the switch block, that block would have to be removed & replaced.
It pays to remember, it is the pcb that has the value, not the many IC's or switch blocks that are easily (in most cases) replaced. I used a very small pair of sharp nosed cutters to cut the switch block up so I could remove its pins, just one at a time, so as to avoid pcb damage. If you decide to free them with the solder sucker instead and do it in one go, make sure to add a lot of fresh solder first and be 100% certain that each pin of the switch block is actually free in its hole, before you attempt to lift the block off the pcb, or you might damage the pcb. The soldering there looks very grey and heavily oxidized, you might have to scrape the surface of it a little to get the fresh solder to take.
( but....one other point; I assume that because you would have been looking around the track work that connects to pin 9 of UD2 desperately searching for problems along its pathway with an Eagle-Eye, and you would have noted from the schematic that this track also connects to the large expansion connector on the side of the pcb, that you would have also very closely inspected both sides of the expansion connector for any obvious signs of shorts or corrosion. If I assumed anything though, I might be a very bad hardware detective, because real detectives operate on a three law platform: assume nothing, trust nobody and check everything)
Please report your findings as you go through the steps.
If you have a list of what to do, it will probably work out for you, to at least get past this problem so you can move on with the rest of the repair.
So follow this list:
1) check with the meter pin 14 of IC A3 and A4 both have +5V on them where the pin enters the plastic IC package. If not find out why and fix that. If that is ok move to step 2:
2) Solder suck and free pin 11 of A3 (make 100% sure it is disconnected with the meter). Check with the scope on pin 9 of UD2, is the interference gone ? if so replace IC A3. If not re-solder the pin 11 and go to step 3:
3) Solder suck and free pin 1 of A4. (make 100% sure it is disconnected with the meter) Check with the scope again on pin 9 UD2, if the interference is gone ? if yes replace IC A4. If the interference is still there, re-solder pin 1 of A4 and go to step 4:
4) Since at this point if the checks above were passed, it would appear that D2, A3 and A4 are all ok, and really the only "likely" thing left is corrosion under the switch block, that block would have to be removed & replaced.
It pays to remember, it is the pcb that has the value, not the many IC's or switch blocks that are easily (in most cases) replaced. I used a very small pair of sharp nosed cutters to cut the switch block up so I could remove its pins, just one at a time, so as to avoid pcb damage. If you decide to free them with the solder sucker instead and do it in one go, make sure to add a lot of fresh solder first and be 100% certain that each pin of the switch block is actually free in its hole, before you attempt to lift the block off the pcb, or you might damage the pcb. The soldering there looks very grey and heavily oxidized, you might have to scrape the surface of it a little to get the fresh solder to take.
( but....one other point; I assume that because you would have been looking around the track work that connects to pin 9 of UD2 desperately searching for problems along its pathway with an Eagle-Eye, and you would have noted from the schematic that this track also connects to the large expansion connector on the side of the pcb, that you would have also very closely inspected both sides of the expansion connector for any obvious signs of shorts or corrosion. If I assumed anything though, I might be a very bad hardware detective, because real detectives operate on a three law platform: assume nothing, trust nobody and check everything)
Please report your findings as you go through the steps.
Gary C
Veteran Member
put your scope on BA14 on channel 1 and get it to trigger with a stable waveform (adjust range and timebase until you get a good, readable trace)
Then with that connected, put channel 2 onto pin 9 of D2 and see if the 'corrupt' part of the pin 9 wave form occurs at the same time as the signal of BA14
If not, then do the same with BA0 to BA15
Then with that connected, put channel 2 onto pin 9 of D2 and see if the 'corrupt' part of the pin 9 wave form occurs at the same time as the signal of BA14
If not, then do the same with BA0 to BA15
Desperado
Veteran Member
- Joined
- Nov 25, 2017
- Messages
- 6,827
Ok, i have +5 V on pin 14 of A3 and A4
Unfortunately when i removed pin11 of A3, the leg broke
I have however tried with the oscilloscope on pin9 of Ud2 and i have always the interference....
Now i remove A3 because i need to replace
Hugo Holden
Veteran Member
Ok that means A3 was not the cause of the fault.Ok, i have +5 V on pin 14 of A3 and A4
Unfortunately when i removed pin11 of A3, the leg broke
I have however tried with the oscilloscope on pin9 of Ud2 and i have always the interference....
Now i remove A3 because i need to replace
It might be a language translation thing, but you were not supposed to "removed pin 11 of A3". You were supposed to solder suck the pin and free it up in the hole so it was disconnected from the hole & tracks.
What do you mean pin 11 of A3 broke ? did you bend & tear the pin out of the side of the A3 IC package ? Why could you not solder it back up again ? Was it corroded where it enters the IC package ? I'm not sure what you mean by "leg broke"
If you have not been doing what I asked, with the solder sucker, to disconnect the IC pins, then you would be better just snipping them off, with cutters, flush with the pcb above the pad, then at least you could easily solder them back up later.
Last edited:
Desperado
Veteran Member
- Joined
- Nov 25, 2017
- Messages
- 6,827
Yes, exactly! I solder suck the pin but it always touched with the pin so I tried to remove it from the hole and it broke.
Now i soldered a new ic!
Hugo Holden
Veteran Member
In that case, for the next test, disconnecting pin 1 of A4, just snip the IC pin to be disconnected above the pcb pad with sharp cutters and the small gap it creates can be bridged with solder later if you need to reconnect it.
Desperado
Veteran Member
- Joined
- Nov 25, 2017
- Messages
- 6,827
Yes!!!! I cutted A4 pin 1 and now, signal on UD2 pin 9 is correct!!
Hugo Holden
Veteran Member
Ok, A4 was defective, please replace it.
Likely it had an internal 5V supply failure to the die and it was causing interactions at its inputs, not a very common fault. We are slowly eliminating the Ghosts from your haunted PET.
daver2
10k Member
It should be a 74LS00 device, not an S device.
Before you remove A4 from the PCB, I would strongly advise you to use this as a training exercise with your oscilloscope and monitor (the now healthy) signal on UD2 pin 9 on CH1 (it should be easy to trigger on it now with it working) and probe the pins of A4 with the other channel of the oscilloscope to see what the signals look like.
You should have been able to diagnose this fault using the oscilloscope alone. If not, it is still a very valuable process for fault finding.
I have never come across a fault like this myself.
Dave
Before you remove A4 from the PCB, I would strongly advise you to use this as a training exercise with your oscilloscope and monitor (the now healthy) signal on UD2 pin 9 on CH1 (it should be easy to trigger on it now with it working) and probe the pins of A4 with the other channel of the oscilloscope to see what the signals look like.
You should have been able to diagnose this fault using the oscilloscope alone. If not, it is still a very valuable process for fault finding.
I have never come across a fault like this myself.
Dave
Desperado
Veteran Member
- Joined
- Nov 25, 2017
- Messages
- 6,827
Unfortunately i already removed A4 but i haven't spare parts