Hello, and welcome to another episode of the 8-Bit Guy.
So, last month I received this very generous donation from David Leech of a vintage Commodore
calculator.
Now, I've never repaired a calculator before, but this one doesn't work.
I probably wouldn't even care that it didn't work except for the fact that it has the Commodore
name on it and of course so it has some historical significance, at least from my perspective.
So, I'm going to take it apart and I don't know, see if we can fix it.
So Commodore was a calculator company before they made computers.
And I think this was made around 1972, it's a little dusty but otherwise in good shape.
However, as you can see when I plug this thing into the wall and power it on, it doesn't
do anything at all.
Now this unit does not have the option to run on batteries, it is powered only from
120V.
So, I'm going to go ahead and take it apart and hopefully it will be some easy problem
with the power supply inside.
Well, that's a very interesting board, all of the traces look hand drawn.
I'm accustomed to seeing computerized layouts.
You can definitely tell there is a 28-pin DIP style chip of some sort on the other side.
And there's a fuse, which might even be the culprit, along with a transformer.
But otherwise, not much to see here on this side.
Before I go any further, I should probably test this fuse with my multimeter by putting
it in continuity test mode.
And it does seem that the fuse is good.
So, I'll go ahead and remove this board.
There are a couple of flat tipped screws here, but then I ran into these little oddballs.
They're little hex screws but I don't have a socket that small.
I could spend the next 3 days trying to find the right socket, or I could just see if maybe
some needle nose pliers will do the job.
And it looks like they will.
And here's the little screw.
OK, now we can finally get a look at this side.
One thing I noticed is that this foam is literally just disintegrating just by touching it.
And it looks like the heart of this is a Texas Instruments chip, which is no surprise because
I know Commodore was buying from them in the early days.
If you ever watch any documentaries with Jack Tramiel, one funny story he likes to tell
is how TI tried to start raising the price of the chips they sold to Commodore so that
they could make Commodore uncompetitive in the calculator business.
So, one of the things that Commodore had to do was buy their own chip fab, known as MOS
technologies.
What's funny about that is when TI tried to enter the personal computer market a few
years later, Commodore had essentially turned the tables on them.
So, time to do something about this foam as it's just getting everywhere.
So the next thing I wanted to do was turn this thing back on and poke around with my
multimeter and check the voltages.
Now, one nice thing is they've labelled a lot of stuff on the bottom of the board,
including several different voltages that are expected to be present.
And poking around, I did manage to find voltages pretty close to the numbers stated.
So, I wasn't entirely sure why it wasn't working.
But then, for no explainable reason, I eventually noticed the calculator came to life.
So yeah, there it is.
However, I noticed that some of the keys worked, some keys didn't work at all, and others
sort of worked intermittently.
So the first thing I wanted to figure out was whether the problem was mechanical, such
as with the keys themselves, or electrical such as trace or resistor or something like
that.
It's also the possibilie the chip itself may be fried, thus it may not be possible
to repair at all.
So, the first thing I wanted to do was see if I could find a schematic diagram for this
calculator, which I couldn't find.
But, I did manage to find a pinout of the main chip from Texas Instruments.
So, this is the information I found, and if you aren't an expert in electronics, this
probably looks a bit like Egyptian hieroglyphics.
So, part of what I like to do here on my channel is try to explain this stuff so that it's
easier for people to understand.
So, let's see if I can make sense of this.
The chip has 28 pins, but we really only care about the input lines for the keys.
And there are 11 of them.
However, you can see that there are more than 11 keys.
Also keep in mind that the switch at the top right has multiple positions so that's like
having several more additional keys.
So how do they read all of these keys with only 11 input lines?
Well, they use a matrix.
Here's how that works.
There are 4 other lines, they have labelled them KN for keyboard number that would be
like numerical keys like zero through 9.
KO for keyboard operator.
That would be things like plus, minus, multiply, and divide.
And then KP for keyboard position, this has to do with the little slider switch.
And lastly KQ, and I have no idea what the Q stands for, but it only handles the one
locking key.
So, the ones in blue here are input lines.
And the other 4 in red are output lines.
Now, I'll show you in slow motion what happens probably hundreds of times per second.
The chip will send out pulses, rotating between these 4 lines.
When you happen to push a key, such as 3, it will connect KN with D3.
And when the pulse comes through KN, this connection will be detected and so it will
know you have pressed a 3.
So, the next thing I did was map out all of the keys that worked.
So the keys in green here work perfectly.
The ones in orange sort of work sometimes, and the ones here in red don't work at all.
So the next thing I needed to do was try to figure out if there was any rhyme or reason
to the bad keys.
Take Key number 2, for example.
It depends on lines KN and D2 in order for that key to work.
If there is a problem with either of those lines, then 2 won't work.
However, it will affect other keys too.
For example, if line D2 is faulty, it will also affect multiply and one of the settings
on the slide switch.
So, it is certainly possible there is an electric problem with D2 since multiply is also not
working.
If KN were bad, then it would affect pretty much every number on the calculator.
So, we can assume that one works.
However, this doesn't explain everything because for example, the 9 and zero keys work,
but CE and C do not.
So the next thing I thought I should try was actually making the connection myself using
a piece of wire up against the legs of the chip to see if I could make a 2.
And sure enough, it works.
So we know the chip itself is working.
In fact, I was able to pretty much make any number I wanted to.
I turned it back over and started examining the traces to see if I could find any lack
of conductivity between the chip and the actual switches, and I couldn't.
So that meant the problem bust be with the switches themselves.
One of the problems with these switches is that they are connected to the board using
a method that would require damaging them to remove.
So, my only hope was that I could disassemble the switches from the top.
This foam was also on the verge of disintegration, so I removed it.
I think it's only purpose is to make the buttons a bit quieter when typing.
I could probably create something to replace this if it turns out to be necessary.
Well, I'm not sure if I'm going to be able to repair this, but it will be an interesting
learning experience.
The way I see it, I can't make this thing any worse than it is now and it's not a
particularly expensive calculator as I've seen similar ones on eBay for $25.
I started with the number 2 button because I know for a fact that one is totally dead.
I wasn't sure if these tops would come off or not, but with some very careful prying
I did manage to pop it off.
Surprisingly, I found a spring inside covered in more of that disintegrated foam.
I think they actually stuck a small piece inside the spring, probably to make it stiffer.
And you can see there's a lot of debris left in there from that foam.
I wanted to see if I could get any contact from that at all, and with some effort, I
was able to.
So that definitely confirms the switch as the culprit.
So, it looks like there are these little metal contacts at the bottom, and I could carefully
remove them.
Now you can see that when you push down on the metal it makes contact between the ring
and that center pin.
I'll just blow that debris out of there.
I wanted to clean the copper inside, but it wasn't in an ideal location for using chemicals
like vinegar or baking soda to clean any oxidation off.
I didn't want to put anything down inside that wouldn't evaporate on its own.
I tried getting a toothbrush down in there, with no luck.
Eventually I realized I could use a screwdriver and a paper towel soaked in alcohol.
I rinsed off this spring and used a toothbrush to try to get as much of that old foam out
of there as possible.
I'm not sure if that foam is required but we're about to find out.
I put the metal contact back down in there, along with the spring and tested it.
I don't know what happened to the footage I thought I took of me testing the 2 key,
but it worked.
So, I decided to go ahead and repeat the same procedure to all of the rest of the keys,
even the ones that were still working.
OK, so I just have the bare minimum installed, and I'm going to test all of the buttons
now.
And they seem to be working, so that means I can finish re-installing the tops of these
things.
Unfortunately, after I put them back on, none of the keys would work at that point.
I suspected one of the keys was making permanent contact, as if you were holding one of them
down, and since the calculator is only designed to read one key at a time, that could cause
a problem.
Fortunately, I was able to take the information I learned about the layout of this chip and
use my multimeter to figure out which key it was.
And it turns out it was the 8 key.
So, back off with the little top piece and I worked to compress the spring a little more.
I had actually had a problem with this one making good contact so I stretched the spring
earlier to give it more force, and now it looks like I gave it too much.
I'm also going to try bending the little metal tab just slightly so that it requires
a bit more force before it flattens out.
Time to test again, and this time all of the buttons are working.
At this point the only thing left to do was clean this thing up some and re-assemble everything.
And here's the final result.
It looks good, and seems to work properly now too!
Also it does have this little stand on the bottom that changes the angle of the calculator
if you so desire.
And well, so here it is.
This is definitely the oldest piece of technology that I have ever worked on probably ever,
since it is from 1972.
In fact, this was produced 3 years before I was even born.
One mystery still remains though, which is why did it not work at all when I first received
it and then suddenly started working?
And, I only have one possible theory to explain that, which is the fuse.
Even though the fuse is good, one of the things I did when I was playing around is I took
the fuse out and put it back in.
It's possible the copper contacts had become corroded and maybe just the act of taking
the fuse out and putting it back in might have created a better contact situation between
them.
And I'm just guessing.
That's the only thing I've been able to think of.
And so, this old Commodore calculator is a perfect introduction to the miniseries I'm
going to be doing on Commodore computers.
Starting of course next with the PET.
And, this particular one is from 1977, just a few years after this calculator was made.
This particular one is not mine.
I'm actually borrowing it from The Obsolete Geek and it actually has the original keyboard
and tape drive built in.
And this one was recently donated to me and it's from a couple of years later because
it has the real keyboard on it.
The interesting thing is most of the keys on this keyboard do not work.
So, it's sort of a coincidence that I was working on this calculator which also had
a problem with bad key mechanisms, and I'm really curious to see, when I open this thing
up if they are similar in any way.
So, you know, maybe what I learned here will go forward to help me fix this thing.
So, anyway, that about wraps it up for this, so stick around for the PET episode, and thanks
for watching!
For more infomation >> Cancer | كانسر | 2007 اربط حزامك ياكلاش - Duration: 4:04. 
For more infomation >> HISTORY OF TERROR The case of the ghost of Edinburgh Castle MYSTERIES OF THE WORLD - Duration: 5:03. 
For more infomation >> 100 Dolares Cada Mes | 2Captcha- Día (26) | 26/Julio/2017 - Duration: 2:03. 
Không có nhận xét nào:
Đăng nhận xét