• Please review our updated Terms and Rules here

[HELP] Commodore PET CBM 3008 3032 Screen Issue

Hi Dave, and hope you had a marvelous xmas. Ok. So refreshed I took up the battle with this time sink again. So the transformer indeed seems to work. The mainboard alone sucks 30w which seems fine. Part of the sparkle seems found: The monitor assembly sucks 300w plus and i found heat damage at two left ones of the rectifier diodes, see picture. Any of you schematics wizes know now what is or could be the issues? Many thanks, Alex
 

Attachments

  • C273544A-5F89-42AD-8094-EE863166C619.jpeg
    C273544A-5F89-42AD-8094-EE863166C619.jpeg
    720.3 KB · Views: 12
Before you ask ;) here is Pin 1 vs Pin 2 of the power input to the screen pcb. Seems like 5v ac if i am reading this right.
 

Attachments

  • EDB2941A-7785-451F-A40A-7469D44EE41D.jpeg
    EDB2941A-7785-451F-A40A-7469D44EE41D.jpeg
    90.3 KB · Views: 14
  • 551937C7-6E80-4D66-BAA6-2732FBB1DF94.jpeg
    551937C7-6E80-4D66-BAA6-2732FBB1DF94.jpeg
    84.5 KB · Views: 14
I looked at the zimmer schematics and cant even find what pin 1 and 2 are supposed to be... but looking at the heat damage i would say some or allnof the the diodes gone short and need replacement. what type of diode do i need. find no info at schematic. sorry for the basic questions. i never planned to go that deep in the pet repair 🤷
 
The fotos I posted use the left oszi port and the probe set to 10x. Setting it to 1x gets me a v rms along 17.4v which is close to my original measurements and make more sense. If i use channel 2 ie the right port of the oszi it grows to peak voltage of 250v. i find this very confusing. i did look at the 4 page (!) manual for the oszi but it is absolutely a chinese joke and useless for someone without oszi training
 
Well done identifying the damage.

I wouldn’t bother using an oscilloscope to read the secondary voltage of the transformer. Sledge hammer to crack a Christmas nut! I would use a multimeter (set to ac volts) with the faulty monitor disconnected from the transformer.

If you have the oscilloscope probes set to x10, you reduce the voltage that is passed to the oscilloscope by a factor of 10. So whatever voltage reading you measure on the oscilloscope, you have to multiply it by 10.

Yes, Chinese manuals are cr*p... You need to download and read a good (old) book on using an oscilloscope. You can then apply what you learn to your oscilloscope. If you are confused regarding how to effectively use the oscilloscope, then I would treat the measurements as “suspect” until you get a bit of experience under your belt. An oscilloscope is a very useful tool, but you need to know how to operate it.

I suspect that one of the diodes has gone short circuit - causing the other to fail in the process. I had exactly the same fault many years ago, but on a very BIG bridge rectifier (30A)...

If you look at the equivalent schematics for the PET 2001 you will observe that the diodes are specified as 1N5402 (3A) or 1N4002 (1A). The voltage regulator is rated at 1A. I would be inclined to use the 1N5402 devices as replacements though.

Make sure you mark the polarity of each diode before removing them. I would be inclined to remove the two faulty devices and test them out of circuit AND test the remaining two diodes to ensure they are still functional. Either that, or replace all four (4) devices at the same time.

Warning: Putting the diodes in the wrong way when they are replaced could cause further fireworks and destruction.

It is also just possible that the voltage regulator input side has gone short circuit. Possible, but a low probability.

I assume you know how to test diodes?

Dave
 
Incidentally, your diodes are identified on the schematic as part number V06CS. The V06C datasheet identifies this device as rated at 1.1A. I seem to have noted down on my sheet of paper 1.5A from somewhere.

Dave
 
I found some diodes around: 1N4004 400V 1A Rectifier. Do you know if they would work as replacements for V06CS? Also why are the diodes mounted on these stelts lifting them 3cm of the board? Thanks, Alex
 
1N4004 diodes should be fine.

The 'stilts' are there for a couple of reasons. Pick one:

1. They prevent the PCB from getting damaged if the diode fails and gets hot.
2. They permit the free flow of air around the diode, thus helping it to remain as cool as possible.
3. The diodes are the 'fusible' element if things go wrong - and the 'stilt' enables them to be desoldered and replaced easily without damaging the PCB in the process.

They are the reasons I can think of...

Dave
 
Update, 4x 1N4004 diodes are soldered in and we are back to working state with one of my PETs. Yeah!!! I feel I did quite an ugly job in soldering :( as I was to lazy to remove the PCB from the Tube for the 20th time... It works though ;)

The interframe wobble is still there on the top left part of the screen, but I was accepting that by now. The wobble amplifies with increasing brightness as does image size generally. I suppose that is not a strong diagnostic hint. Analytically hunting that problem down seems too much in my mind because I think even the scope patterns would look mostly alright. If there is any "blind" replacements you recommend, I'd surely consider as the parts cost very little.

For the nonlinear one I have to start finding inductivities... I have some understanding of Rs/Cs/ICs and what not but these Lsssssssss.... No idea how to find equivalents. I will start my hunt :) Thanks
 
Well that's good (but not about the ugly soldering though)...

Now we have your PET working again - can you go back and read the few posts from #75 onwards regarding the scan coils, capacitors and the like?

You were going to post a video fo the 'wobble' after the scan coils were centred and (hopefully) you had adjusted the height and width.

Dave
 
Thanks Dave,

so when going back I think I could replace L1 and L2 (both Miller) as per your schematics link http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/2001/video-layout.gif
as they might have broken cores or other issues. For these two parts I am clueless what part to get today.

The caps in your suspect list are already changed. So that has me thinking about Hugo's HOT behavior and any additional resistance that would squeeze x-axis. But not having a probe allowing me to measure there I am not sure what components to switch. Kind regards, Alex
 
And L1 and L2 sit on the Video PCB, right? Dave, you mentioned they were part of the yoke as per post #76, but I think they are not. They are right were they are supposed to be on the layout drawing. Or am I misunderstanding something? Thanks, Alex
 
Are we back to the condition where the raster matches the photo in the original post ?

And now the power supply is known to be ok ?

If that is the case, the best move would be a scope recording with a x10 probe to look at the collector voltage of the HOT. To find out if both the damper diode and HOT were conducting properly during scan time. The flyback peak must be about right or the EHT would be way off, which it is not.
 
I see where the confusion has occurred now...

In post #65 I just mention the scan coils.

In a slightly later post you mention L1 and L2.

I haven't worked out that we are not talking about the same thing!

I meant the physical scan coils around the neck of the CRT that cause the electron beam to be deflected.


Dave
 
Hi,

IF you have access to an oscilloscope, you can compare signals from same part points between the two machines and when you find something different then you're hot on the trail of the fault.

Having a good working system setup identically as the faulting system usually makes repairs fast and easy.

That's the value of having a nice cheap $250 oscilloscope ... it can save you a ton of money, time and effort.
 
Back
Top