• Please review our updated Terms and Rules here

Neglected PET needing some love

I worked in a CRT re-gunning plant for a while, the toxicity levels were fairly low. Generally dislodged CRT phosphor tends to be flakey rather than dust because of the Barium Silicate binders and the aluminium backing. The same with the internal Aquadag. It is ok to handle the electron gun structures, they are made of high purity metal. You don't want to do that with new ones as it can leave finger marks from sweat. Like many things an alarmist approach and paranoia about chemical toxicity is not warranted for a broken CRT, no more than any common garden vacuum tube. The main thing is avoid cutting yourself on the glass, sweep up and dispose of any fragments & flakes wear gloves and/or wash your hands. If you insist on poisoning yourself, you could if you tried very hard, but you would have to get all the flakes and butter them onto your sandwich, so I would suggest not doing that. If you are taking CRT's in and out of sets, its worth wearing some polycarbonate safety glasses. And, many of the metal bands on small CRT's 12" or less are mounting bands, not "implosion bands" and they can be removed with little risk, that is if the glass does not get scratched in the process. I have removed these many times adapting various CRT's into VDU's.

The main place where human Chromium toxicity turns up is at the Electroplaters (for obvious reasons of chronic exposure). Then there are metal allergies. As I recall Men have a higher incidence than women to Chrome allergy, and in Women I think it was Nickel. Thought possibly due to the backs on wrist watches containing Chromium.

Often it is the case than common sense is better than any amount of fear. In all cases of problems involving poisoning and chemical toxicity, with certain super lethal exceptions requiring only minuscule amounts, it is always a combinations of the factors of dose amount, route of administration and duration or chronicity of exposure that determine the outcome. This is why when people work daily in industrial situations, with ongoing exposure over many years there can be major problems. But an occasional contact with the substances are little of a worry. A classic example of this is Carbon Tetrachloride.
 
Last edited:
Does anyone here have experience working with the RGB2HDMI? I have been trying to search google for how to take an already-existing profile file and put it on the card such that I can select it in the UI. All of my searches (including those on the official RGB2HDMI wiki) go into great lengths about how to make a profile within the RGB2HDMI menu system but what I have is a profile someone else already made and I put it on the SD card under "Profiles" but it doesn't show up when I cycle through. Is there some master list of profile files that I need to add an entry on?

I think this pet is working. I made a cable to connect the user port to the RGB2HDMI following along with the video I linked earlier (
). In the description of that video is a link to profiles. I also went through my settings and matched them to the settings he shows in the video but they didn't work for me, it could be due to a different version of PET.

Anyway when I flip the power switch I get the PET startup chime and I can just barely make out a "ready" flickering on my screen. So I think the pet is working but I just am too stupid today to figure out how to load in a custom profile so before I get too frustrated thought I'd ask here.
 
Okay never mind I figured out how to add the profile, it needed to be placed under Profiles/6-12_BIT_RGB not just Profiles
But it looks like the pet has issues after all, unless I just messed up making the cable but it's a pretty simple cable. Here is what I'm seeing:

I'll poke around on the scope to see if I see a sync signal coming out. Does this (in the video ) look familiar to anyone?
 
A sync signal from where exactly?

There are no synch signals coming from the PET - just the H and V drive signals.

So, you are correct, check the PETs H, V and video signals for sense first...

Dave
 
A sync signal from where exactly?

There are no synch signals coming from the PET - just the H and V drive signals.

So, you are correct, check the PETs H, V and video signals for sense first...

Dave
yeah I'm seeing a signal on the h.sync, v.sync, and video signal pin (all on the user port).
of course it's kind of hard to tell if the signals are correct or not on my old analog scope but I don't see anything obviously wrong (obvious to me anyway)
 
Can you identify the approximate frequency of the H and V signals on your oscilloscope?

The signals should both be high and go low for a short period of time.

Dave
 
I'll get it back in the scope and post some pictures tomorrow morning. i should really invest in a more modern scope.
 
This might seem like a stupid question but just to confirm the user port is the one in the middle right? I've never used a pet before and there are two 24 pin edge connectors on the back. I tried to find a user manual but all I could find are user port pinouts, repair manuals, etc... nothing so simple as "this is what this port is". Also what is the other 24 pin edge connector? Is it for a floppy drive?

I did some scoping today but I'm not sure still images would do justice to what I'm seeing so I decided to just set up the camera and point it at the scope and keep in the shot the connector I'm probing. I have a diagram of which pin on the connector is what in the shot as well, a little hard to make it it's kind of washed out so here's the diagram:

H. SyncV. Sync(Not used)(Not used)(Not used)
GroundGround(Red, Not used)Green(Blue, Not used)

I have already probed the user port directly to confirm that what I'm seeing on the connector is the same and on the expected pins but to make it easier to make out what I'm probing in this video I probed the connector instead.

There are a couple of good captures of the v.sync which I grabbed stills for (below) but for the h.sync and video signals I couldn't find one still frame that would be a good representation of what I'm seeing. I think this is because my scope's time axis only has set values and I can't look between them so like i can set it to 20 uSec/cm, 10 uSec/cm, 5 uSec/cm, etc... but I can't set it to say 15, or 7 or anything in-between.

pet_vsync.jpg


Link to video: (still uploading at time of writing)
 
The assembly drawing http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/8032/8032030-6.gif is your friend here.

J2 is the USER port.

Pin 2 = VIDEO.
Pin 1 = GND.
Pin 9 = V DRIVE.
Pin 10 = H DRIVE.

See here http://www.6502.org/users/andre/petindex/pinouts.html#io for the actual pinout arrangement for the USER PORT.

J3 is a cassette port. Note that there is a second cassette port on the side of the PET also.

J1 is the IEEE488 port (for interfacing to printers, disk drives, tape drives, volt meters, oscilloscopes, plotters, etc. etc. etc.

Dave
 
The assembly drawing http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/8032/8032030-6.gif is your friend here.

J2 is the USER port.

Pin 2 = VIDEO.
Pin 1 = GND.
Pin 9 = V DRIVE.
Pin 10 = H DRIVE.

See here http://www.6502.org/users/andre/petindex/pinouts.html#io for the actual pinout arrangement for the USER PORT.

J3 is a cassette port. Note that there is a second cassette port on the side of the PET also.

J1 is the IEEE488 port (for interfacing to printers, disk drives, tape drives, volt meters, oscilloscopes, plotters, etc. etc. etc.

Dave

Okay thanks. I did see that first image but there's nothing on that diagram that says that J2 is the user port, just that it's J2. I guess I needed that 2nd page. The text "Arrangement of pins plug J2" below "USER - PORT" is the only indication that the middle port (which is J2) is the user port. But that's what I'm saying there's tons of documentation and technical details right down to the metal but something really basic that one might expect in the first few pages of a standard owner's manual is not so easy to find.
 
The associated schematic containing the USER port details it is J2 and the assembly drawing identifies where (physically) J2 is on the board.

There is a logical method...

The USER manual would also detail this information. It does, honestly...

I would suggest downloading all of the relevant manuals from the Zimmers website related to your particular machine.

Dave
 
The h and v drive signals out of the PET are not h and v sync signals though the vertial one is close to ,but still wider than a negative going v sync pulse would be.It is equivalent to a negative going blanking pulse that has the same width as vertical flyback.
The h drive pulse is much wider than an h sync pulse, it has the duty cycle suited to driving an h output transistor.
Initially there has to be a small circuit that shortens the h drive pulse, and mixes it with the v drive pulse in the correct proportions and polarity and mixes that with the video signal to create a composite video signal, then it can be fed to the composite to HDMI converter box for good results, or fed directly to a composite vdu.
I assume you are using the converter circuit with the logic gates?
 
Last edited:
But it looks like the pet has issues after all, unless I just messed up making the cable but it's a pretty simple cable. Here is what I'm seeing:

I'll poke around on the scope to see if I see a sync signal coming out. Does this (in the video ) look familiar to anyone?
I have a RGB to HDMI board and it works well with the PET 8032 timing signals from the User Port to my inexpensive 15" flat screen HDMI monitor. But I had to play around a little with Chuck Hutchins' configuration settings a little.

First try, with his settings, it would not sync and I got a messed up screen something like yours with rolling patterns. On the Main Menu, I changed Scaling to 'Interpolate 4:3 /Soft'. I have no idea what that means but it seemed to fix everything for my monitor. Don't be shy about using the Reset button on the gadget. It seems to get smarter with each try. It is a pretty smart board.

Can you list the profile settings you are using and I will compare with my file. Mine is very close to Chuck's.

Also don't 'Save' the configuration until you get one that works. Keep Chuck's as the baseline as his will be close until you tweak your monitor in.
 
The h and v drive signals out of the PET are not h and v sync signals though the vertial one is close to ,but still wider than a negative going v sync pulse would be.It is equivalent to a negative going blanking pulse that has the same width as vertical flyback.
The h drive pulse is much wider than an h sync pulse, it has the duty cycle suited to driving an h output transistor.
Initially there has to be a small circuit that shortens the h drive pulse, and mixes it with the v drive pulse in the correct proportions and polarity and mixes that with the video signal to create a composite video signal, then it can be fed to the composite to HDMI converter box for good results, or fed directly to a composite vdu.
I assume you are using the converter circuit with the logic gates?
Hi Hugo,
It turns out the the RGB to HDMI board will work with almost any TTL (+5V) video interface. MDA, CGA, PET and many other video interfaces can be converted into a good HDMI signal. It is a very smart gadget with a comprehensive configuration menu. Although this can mean a lot of experimentation and calculations regarding the signals. See this geometry menu as an example. There are several other configuration menus also. Thankfully we had Chuck Hutchins to figure it out.
 

Attachments

  • IMG_2264.JPG.jpeg
    IMG_2264.JPG.jpeg
    152.6 KB · Views: 11
Last edited:
I have a RGB to HDMI board and it works well with the PET 8032 timing signals from the User Port to my inexpensive 15" flat screen HDMI monitor. But I had to play around a little with Chuck Hutchins' configuration settings a little.

First try, with his settings, it would not sync and I got a messed up screen something like yours with rolling patterns. On the Main Menu, I changed Scaling to 'Interpolate 4:3 /Soft'. I have no idea what that means but it seemed to fix everything for my monitor. Don't be shy about using the Reset button on the gadget. It seems to get smarter with each try. It is a pretty smart board.

Can you list the profile settings you are using and I will compare with my file. Mine is very close to Chuck's.

Also don't 'Save' the configuration until you get one that works. Keep Chuck's as the baseline as his will be close until you tweak your monitor in.
thanks I'll try changing the scaling setting tomorrow morning and play around some more with the settings.
 
Hi Hugo,
It turns out the the RGB to HDMI board will work with almost any TTL (+5V) video interface. MDA, CGA, PET and many other video interfaces can be converted into a good HDMI signal. It is a very smart gadget with a comprehensive configuration menu. Although this can mean a lot of experimentation and calculations regarding the signals. See this geometry menu as an example. There are several other configuration menus also. Thankfully we had Chuck Hutchins to figure it out.
When you say Video Interfaces, I'm not sure what you mean. In the case of the PET, in your case for your PET, where are you acquiring the three signals (H & V and video) ..is it from the signals feeding the original VDU, or are you getting them from elsewhere on the PET main board ?

In the case of the signals that feed the PET VDU, these are not standard "sync" signals, but drive signals. They are better modified into the form of sync signals before they are used. For example, some VDU's (of the CRT variety) protest if the V sync pulse is too wide.

Have you made the composite adapter with the three NOR gates ?
 
Briefly back on that subject of the risks of toxic chemicals, such as those from broken CRT's, it is amazing how far Occupational Health and Safety has now gone to protect people from hazardous, no yet lethal and toxic substances in the local environment.

I took a photo of some tanks holding some of this toxic stuff beside a driveway on a McDonald's takeaway drive through, you better watch out for it, it makes the metabolic acid on the Alien movies that corrodes through the hull of space craft look like child's play. If you ever see this dangerous substance, call the Hazmat team and run for your life. (see attached photo)
 

Attachments

  • OHS.jpg
    OHS.jpg
    54 KB · Views: 8
Well I tried playing with every conceivable setting in the RGB2HDMI and couldn't get it dialed in any better than in the video I posted. I noticed that the text sometimes says "ready." and sometimes is garbled like in the video. I attached the keyboard and tried hitting a bunch of enters to see if the "ready." would scroll away and nothing happened. I also typed in a simple program:
10 ? "X";
20 goto 10
run

to see if the screen filled up with X's but nothing seems to have happed.
So, aside from the RGB2HDMI issues which might be an improper sync there may be ram issues or other issues.

BTW the scaling was already set to Interpolate 4:3 /Soft but I tried cycling through the other options.

I'm wondering if it's possible that the RGB2HDMI connection/setup is working correctly and the pet is just repeatedly typing:
ready. ready. ready. ready. ready. ready. etc...

Is the PET known to have a failure mode like that?
 
Last edited:
Is the PET known to have a failure mode like that?
It appeared from the video you posted that there was no vertical lock and possibly no H lock either, so you were seeing it scroll by, that may be why it looks like its typing ready.ready.ready etc and the video of the text ready is appearing in multiple places perhaps. To make much use of the H drive pulse as a sync pulse signal it requires (as it is in the composite adapter) to be shortened to create a pulse from the correct leading edge. In the case of that composite adapter, they do it by differentiating the pulse with a low value capacitor.

Basically the negative going part of the H drive waveform which is about 15us needs to be shortened to narrower pulse of around 3 to 5uS to better resemble an H sync pulse. Of course, this is a negative going pulse in the 12" VDU (as also is the vertical drive pulse leading edge) so whatever converter unit you are feeding with those pulses has to be able to accept negative going sync pluses.

In the 12" PET VDU. when that H drive signal is low (for 15us) corresponds to when the HOT (Horizontal output transistor) is switched off and flyback starts (but that 15uS longer than the flyback period and a lot longer than an H sync pulse) and when the H drive goes high, this corresponds to the time the HOT is switched on during the active scan time. So the closest point an actual H sync pulse can be derived from, is the falling (negative going) edge of the H drive pulse which corresponds to the start of flyback in the 12" VDU.

I have just noticed looking at the 12" VDU vs the 9" one (apart from the different H scan rate), they appear to have opposite polarity H drive signals. The 9" VDU has an extra transistor that inverts the signal, and the 12" one does not. I have never owned the 12" version or the later PET with the CRTC IC, so this is interesting news. It means the design of an adapter to create a composite video, to work properly, would have to be different between the two machines, that is if one used the signals presented to the VDU's connector at least. Probably there are few if any VDU's that run on composite inputs at Horizontal frequencies close to MDA, so probably this has never been an issue and the H & V syncs are just coupled in separately or into a converter unit, but the polarity still has to be correct.
 
Last edited:
It appeared from the video you posted that there was no vertical lock and possibly no H lock either, so you were seeing it scroll by, that may be why it looks like its typing ready.ready.ready etc and the video of the text ready is appearing in multiple places perhaps.
Yeah could be. Given that sometimes it's garbage and sometimes it's "ready." makes me wonder if there's an issue with one or more of the roms. Three a socketted and three aren't. I have an eprom reader/writer but I'm not sure it'll work with these chips.

Anyone know where I can find the rom images for this model of PET so I can compare CRCs?
Edit: looks like they'd probably be here: http://www.zimmers.net/anonftp/pub/cbm/firmware/computers/pet/index.html
 
Last edited:
Back
Top