Hello, and welcome back to The 8-Bit Guy.
In this episode, my plan is to restore this old Commodore PET to a fully functional state.
Now, it has essentially 3 problems that need to be dealt with.
The first problem is it's a little bit dirty.
And, it is a metal case, which is not something I deal with very often.
There's not going to be any retrobriting going on here, because this is actually painted
metal.
So, I'm going to clean it up, but the two problems is that one is the screen is a little
bit dim.
I mean, you can read what's on the screen.
And, there is a brightness control back here in the back, but it's all the way as bright
as it will go, and I feel like the screen should be brighter.
So, I'm going to have to take the CRT part apart and see if there's anything I can
do to brighten it up.
The 3rd thing is the keyboard.
Essentially, almost none of the keys work.
A couple of them work, and like all of the rest of them are completely dead.
So that's something I'm going to have to look into fixing as well.
So, let's get started.
First, I thought I'd tackle the easy part, cleaning it.
Oddly enough, the camera really doesn't want to show the dirt on this computer, for
some reason.
But there's a fine black gunk just all over this thing, and my usual chemicals like windex
and alcohol seemed to make no effect at all.
When I do a closeup you can actually see some of the gunk I'm talking about.
My next test would be baking soda, as I've had great luck with it on stuff like this
in the past on plastic.
And sure enough, I could see an immediate result.
Which means I'd have to end up scrubbing the entire computer with baking soda.
This is probably the largest cleaning project I've ever done with baking soda.
But it works, and I'm a little disappointed the camera doesn't show as sharp of a contrast
between the areas I've cleaned and not cleaned.
However, looking at the top of this monitor, you can see where I've cleaned one side,
but not the other.
The difference is even more dramatic in person.
But, suffice it to say, the computer is starting to look at lot better now.
And you can see all of the black stuff on these old paper towels I'm using.
And here's the finished product.
I still haven't cleaned the keys, because I may be pulling those off anyway shortly
for troubleshooting.
And the corner of the main decal is peeling off, which needs to be fixed.
But otherwise, this thing looks great!
The PET opens like the hood of a car.
Actually, maybe more similar to a semi truck's cab.
The inside is pretty dusty.
I managed to blow most of the dust out with some compressed air.
But some of it was just stuck in place and I had to scrub it loose with a brush.
Now, it's time to tackle the keyboard problem.
Before I begin, I think it's important to show you how the keyboard on this computer
actually works, that way you'll better understand my troubleshooting processes.
When I unplug the keyboard, you'll notice there are a lot of wires, in fact, there are
17 wires total.
But there are a lot more than 17 keys on the keyboard.
So you may be wondering how this works.
If we take a look at the keyboard connector, it can be broken up into two parts.
You have output lines, that send signals to the keyboard.
And then input lines that receive signals back from the keyboard.
These are laid out in a grid like this.
And then, at the intersection of each input and output line, there is a key.
Now, I can't tell you at this point which keys are where on this grid.
The problem is, I managed to find a schematic of the PET and a keyboard matrix layout.
The trouble is, the PET had several different possible keyboards, including the business
keyboard, the home keyboard, and even the cash register style keyboard.
This diagram here does not match the keyboard I have on my PET.
But the specific keys don't matter that much just yet.
So, back to how this works.
The computer sends rapid pulses down each of the output lines one at a time, like this.
If you press a key, eventually the line with that key on it will be pulsed, and the signal
will travel back of the computer on one of the input lines.
From the computer's perspective, it's just making a connection between these two
pins on the keyboard connector, and it has an internal map of the keys, so it knows which
one you pressed.
Now, at the logic board level there are two chips that are mostly involved in running
this operation.
One is the 6520 PIA chip, and the other is a binary to decimal decoder chip.
The decoder drives the output lines.
The purpose of the decoder is that the computer can use only 4 output lines and send a binary
signal, and the decoder chip will translate that into a single pin like this.
This same process could be done without the decoder chip, but it would use 10 output lines
on the PIA chip instead of 4, so that frees up more I/O on the PIA chip for other things.
As for the input lines, the essentially go straight to the PIA chip.
So, that leaves us with essentially 4 things that could be wrong with the keyboard.
It could be a logic board problem with the 6520, or the decoder chip.
It could be a wiring problem between the board and the keyboard.
Or last, it could be the switches in the keys themselves, which seems unlikely considering
the number of dead keys.
One idea I had was to use a wire with alligator clips on each end, and just attach it to one
of the output lines.
Then just rub the other side across some of the input pins and see what happened.
And this was the result.
So, I think this can pretty much assure us that the logic board is working fine and the
problem must lie somewhere else.
But, just to be sure, I wanted to do a similar test.
I put this towel here just so that I could lower the computer down so I could type on
the keyboard, but at the same time have the keyboard wire hanging out the side without
getting crushed.
I took this spare pin header and shoved it down into the connector so that I could connect
my alligator clips to it.
I set my multi meter on continuity mode so that I could hear if there is any connection
made, then connected it to the alligator clips like this.
I then proceeded to press every single key on the keyboard, but I never hear any beep.
I tried other areas on the connector and repeated the test, but found nothing.
So there's definitely some problem with the keyboard.
At this point I decided it was time to remove the keyboard.
The best I could tell it was held in by these screws on the bottom.
Most of what I'm doing at this point is unknown territory for me because there is
so little documentation or discussion online about this model.
In fact, after removing the keyboard, I had an interesting revelation about this computer.
With the keyboard out, I could see that this computer is actually made of plastic, not
metal.
But I'm not crazy for thinking that it was metal, and you'll see why.
While this particular computer appears to be plastic everywhere, the older model PET,
which I have here is in fact metal, everywhere.
However, I'm still pretty sure the plastic is painted, so it's not bare plastic like
most computers of the 80s.
Anyway, back to repairing this keyboard.
I actually didn't even realize that the keyboard had another connector here, because
I couldn't really see it very well when it was inside of the computer.
So, this is another possible point of failure.
I don't see any corrosion here, but I thought I should do another test.
I reconnected the keyboard and then used my multi-meter in continuity mode to test the
connection between the actual keyboard, and the other end of this cable.
By doing this, I can verify that a good connection is being made, and that the cable itself is
good.
And, every pin tested perfect, so there's no problem here.
Time to disassemble the keyboard.
It appears the shift-lock key is actually soldered in with wires, so I'll need to
desolder these before going any further.
Once that's done, I can start removing all of these screws.
And, here it is.
The circuit board part looks pretty typical from the era.
But, it is in good shape.
I don't see any corrosion or broken traces or anything that could be causing a problem.
On the other side, we just have these little rubberized carbon contacts, which are quite
common.
I'm not really expecting any problems, but I thought I would just double check the connection
between the keys and the connector here.
Everything seems fine, so there's not a whole lot left to do besides clean the board
and the carbon contacts.
After that, a little test was in order.
I mapped out the zero key and connected my multimeter to the correct contacts.
But, I still don't get anything.
Interestingly enough, if I move the multi meter over to resistance mode instead, it
does in fact register some connectivity when I press the key.
But apparently the resistance must be too high for the computer to recognize this.
I used a flat tipped screwdriver to test again so I could determine if the problem was with
the board or those carbon contacts.
And I think I have my answer.
At this point I decided to try some of this de-oxit to see if I could improve the conductivity
that way.
Testing again, I didn't really see any difference.
So I turned to google and many people suggested trying a pencil to add some carbon back to
the rubber pad.
Tried that, but it didn't work either.
Although, there does seem to be less resistance now.
I decided to buy this repair kit, which is made exactly for this purpose, and it wasn't
cheap either.
It cost me nearly $30.
I figured it would be easier if I just took all of the keys off, so I could get to the
little plungers more easily.
It actually turns out these keys on the PET aren't very hard to pull off and it was
easier to do them by hand than using the key puller.
So here's all of the little plungers that need repair.
OK, let me open this thing up.
Gee, there's not as much in here as I was expecting.
The instructions say to mix the smaller bottle into the larger one.
OK, that's done.
It's actually kind of hard to mix this stuff up because there's so little in here and
most of it is on the sides and not the bottom.
Well, here goes my first attempt to paint this stuff on.
OK, well that's not too hard.
Of course, as I was working on these, I had to wonder if I was wasting my time because
I still don't know for sure if this was going to work.
To make matters worse the instructions said to wait 24 hours before using the keys.
OK, they are all done, so tomorrow I'll get to find out if it worked!
OK, so it's been 24 hours and I've got the meter hooked back up and I decided just
to take one of these plungers and test it right on the circuit board.
And it works.
So this is giving me some confidence that this may actually get this PET working again.
Of course, now I have to re-assemble the keyboard.
For my first test, I decided to run the cable out of the PET and just test right there on
the bench.
That way if I need to fix something else, it will be quicker than a complete teardown
again.
And here goes.
I do believe we have a winner!
The keyboard is working!
Now I can move on to some other issues.
While working on the keyboard I noticed this EPROM that was missing the sticker over the
top.
That's not good because if any UV light were to get in there, it could corrupt the
EPROM.
I found the little paper sticker laying on the bottom of the case.
I noticed it had a SYS command written on it, which suggests to me this is some sort
of aftermarket software.
I wanted to test that, so I typed in the SYS command on the PET now that it has a working
keyboard.
So it did something, but I don't fully understand what this software does.
It must just add some features to BASIC.
Anyway, I decided to clean the gunk off of the EPROM with some alcohol.
That looks much better!
Then, I printed out a better label with the same information.
There, that's much better.
Maybe some day I'll figure out what it is for.
The last thing I needed to do was increase the brightness.
Originally I thought I'd need a major disassembly to do that, but I noticed if I shined a flashlight
through the back that there were some extra potentiometers there.
One of them was labelled sub-brightness, so I found that I could stick a tiny screwdriver
through the back slot and rotate that.
And sure enough, I was able to increase the brightness to a level that is much better
now.
All right, so the PET is one of those computers that I would actually not recommend for beginner
hobbyists.
I mean, for one thing they're very expensive.
They usually run several hundred Dollars.
But, the other thing is, they're pretty rare.
And so finding parts for them, I mean even something as small as a keycap or something
can be very difficult to locate.
Also, it's very difficult to find a schematic for them.
And yes, you can go on google and you can get schematics for the PET.
I even found some myself.
The trouble is, the PET was only in production a few years and went through several design
changes during that time.
So, a lot of the documentation and schematics that you do find don't match up what you
might have.
So it's just because it was never much of a popular computer with enthusiasts like the
Commodore 64 has been for example, it's just not that well documented.
However, at least with this one I feel I've made some really good progress with this video.
There's still a couple of things that need to be fixed on this computer.
For one, is this little corner piece of the decal.
I still need to epoxy that back on or find some kind of glue or something to put that
back.
That'll be easy, I'll take care of that later.
The other problem is actually with the keyboard.
And there's two problems with the keyboard, you might have missed both of them.
One is, if you look at the keys, you know, they're supposed to have these little graphics
characters that are on the front face of the keys.
But, many of the little graphics characters are scratched off for whatever reason.
It doesn't really impact the computer that much, so I'm not really going to worry about
that.
I'm not sure there'd be any way to fix it anyway.
But the more serious problem is actually with the space bar.
It works fine if you push the space bar right in the center, But, if you push it on the
side, well, it doesn't work so well there.
And that's actually due to a little broken plastic tab, and I've thought of several
different ways that I might be able to fix that.
So, I'm going to work on that later since I think I've spent enough time on this last
week as it is.
But, I think I can fix the space bar too.
So, anyway, for those that were disappointed that I didn't show more information about
what you could do with the PET, well never fear because that will be the focus of an
upcoming video next month.
So stick around for that, and thanks for watching!
Không có nhận xét nào:
Đăng nhận xét