Denial

Here's a project I've brought back from a recent visit to my parents in New Zealand. My father bought this many years ago as a source of spare parts for our VIC 20:
I shall be attempting at some point to fire it up and see what we get. All the chips other than the RAM are in sockets. Something isn't quite working, so the first job is to track that down.

This is a board 324001 to which active cooling was applied to the voltage regulator and rectifier bridge.

Today a much better solution is to replace the regulator (usually an LM323 or TO-3 7805) with a modern switching type.

Gyro Gearloose even designed a TO-3 format switcher specifically for this application.

I'll have to mention that to my father then. I think it was he that added that cooling fan to the circuit board. Although the board was used mainly for spare parts, he did try to boot it up in isolation a few times. His career was as an Electronics Design Engineer. His knowledge of analog electronics is equal to the best I'd say, starting back in the 50s, but he didn't end up working with digital much. He retired many years ago.

You might notice that there is a small section of this VIC 20 circuit board that has been cut off on the right hand side. I think it was like that when we got it. I forget now what chips we used from this board to fix the main machine, but looking at the stickers, it would appear that one of the 6522 chips is dead, so that probably came from the main machine. I think that recently he replaced most of the easily acquirable 7400 series chips, and the 6561. It's still not particularly happy though, so my project, when I get time, is to diagnose it further and see if we can work out what chips are dead. I'll probably start with the obvious one that has the big green sticker saying "Dead" on it.

Kind of seems obvious.

lance.ewing wrote:I'll probably start with the obvious one that has the big green sticker saying "Dead" on it.

Kind of seems obvious.

I suppose that's a good start.

I'm dabbling with this project again as well. Firing it up shows garbage on the screen. The normal white background with cyan border is there but the screen is full of random characters in random colours. Although I now have ready and waiting replacement chips for everything on there apart from the ROMs, I'm reluctant to go through replacing everything if I don't need to. I have tried swapping the 6561 but that didn't make a difference. I also replaced the 6522 that had a sticker on it implying it was dead. No difference either, but I wasn't really expecting there to be.

I've purchased one of the diagnostic carts from Tynemouth software. It appears to be saying there is something wrong but unfortunately I can't read exactly what its saying because the screen is obviously still full of garbage. But there are characters here and there that are correct, and when compared with example screen shots and videos from the diagnostic cart output, it seems to suggest something is bad. There is a B and a D in a position on the screen where an example screen shot had the word BAD. Not surprising there's something bad, but still no closer to being able to identify what chip (or chips) it is.

The following web page caught my eye a few weeks back and I started thinking that maybe it is the RAM chips.

http://swut.net/vic20.html

The approach taken using an Arduino to write to the RAM chips seemed like something I might be able to get working. So I've lifted the 6502 out of its socket and now have an Arduino Mega plugged in to the address bus, data bus, and the R/W line. May need to use the clock output pin as well. Just beginning to play around with it and thought I'd try adjusting the 6561 control registers as a way to iron out some writing logic. Well the address is hitting the right place. I'm getting colour on the screen when trying to change 0x900F, but the timing is clearly wrong at the moment. Its a good first step.

lance.ewing wrote:...it was he that added that cooling fan to the circuit board.

It reminds me of the cooling fan that Ray Carlsen added to one of my Plus/4's.

Truly,
Robert Bernardo
Fresno Commodore User Group
http://www.dickestel.com/fcug.htm
July 29-30 Commodore Vegas Expo v13 -
http://www.portcommodore.com/commvex

It is a RAM or multiplexer issue

mrr19121970 wrote:It is a RAM or multiplexer issue

AFAIK, the VIC-20 doesn't use multiplexers. The C64 has dynamic RAM so it does.

I correct myself "74LS245 TRI-STATE LOGIC"

eslapion wrote:This is a board 324001 to which active cooling was applied to the voltage regulator and rectifier bridge.

Today a much better solution is to replace the regulator (usually an LM323 or TO-3 7805) with a modern switching type.

Gyro Gearloose even designed a TO-3 format switcher specifically for this application.

Yes he did. But he is also curious, what happened to the one he sent you? Hm??????????????????????????????????????????????????????????????????????

Gyro Gearloose wrote:Yes he did. But he is also curious, what happened to the one he sent you? Hm??????????????????????????????????????????????????????????????????????

900 PLAnktons, more than 100 Behr-Bonz and 80 TOLB modules later...

eslapion wrote:

Gyro Gearloose wrote:Yes he did. But he is also curious, what happened to the one he sent you? Hm??????????????????????????????????????????????????????????????????????

900 PLAnktons, more than 100 Behr-Bonz and 80 TOLB modules later...

I've analyzed your justification and decided to accept it... for now.

So you've been keeping busy huh?

Having made a few steps backwards, we appear to now be making forward progress. It turns out that there was a broken data line track in the area of the RAM chips. All the RAM chips had sockets added and chips replaced as part of tracking this through. I should have just done connectivity tests in the first place. Would have saved a bit of time.

I've now got a white background with cyan border and no text on start up. I purchased a diagnostic cartridge a while back, so now that the broken track is repaired, the output of this cart is readable (previously it was garbage). This cart appears to be telling me that the BASIC ROM is BAD. So that's the next thing to replace.

I now have a flashing cursor, thanks to Dave at Tynemouth Software for the BASIC ROM and adapter, and thanks to his diagnostic cart for identifying the ROM as the issue.

Now it needs a keyboard, a case, probably a few other bits and pieces. All it was at the start was the circuit board I posted the picture of at the top of this topic, and not even a 100% whole circuit board. It's missing a joystick port. I've already ordered a case and keyboard, in fact the case has just arrived. When I was searching online, all I could find at the time was a C64C case. So I'm going to have a go at creating a VIC-20C like what @Clockmeister built here:

http://www.users.on.net/~clockmeister/other/VIC20C/

Should be fun.

So I think I have everything I need now to put this all together over this weekend. I've already modified the C64C case to allow the cartridge to slot in. The keyboard arrived and appears to be fully functional. I needed some keyboard brackets, which I've ordered and will hopefully arrive today (they tried yesterday but I wasn't in.... actually, it just arrived as I was still typing this post).

Last weekend I did some messy but apparently operational soldering to attach 7 wires to where the joystick data lines had been cut off at the end of the board. I'm planning to connect those to a joystick port at some point this weekend. Connectivity tests from the ends of the attached wires back to the various chips shows that they all connected to where they should be.

Unfortunately these adapters for the ROMs aren't going to fit in the case. In addition to the BASIC ROM adapter, I also purchased a JiffyDOS/VIC Kernal adapter board with switch attached. Thought it was a great idea until I did a dry run through last night of what I'd be doing today. There's no way it is going to fit under the keyboard. This is the older and longer mother board.

I've solved the problem for the BASIC ROM by using an original BASIC ROM. I happened to purchase a couple of "not working" VIC20CR motherboards since my last post, mainly for parts if I needed them while getting this one going. I grabbed the 6502 from one of them (as my one I broke a leg when diagnosing the earlier issues with the broken track and had to do a patch up job) and the 6502 is working fine, as is the BASIC ROM. I'm guessing there are a lot of chips on these other boards that are perfectly fine, but at least one of the circuit boards looks like the underneath has been damp or something. Hmmm, anyway that's the board I took the 6502 out of and it was fine. So should be a good source of various parts should I need them. I might even try to get the board that's in better shape working after I've finished with the current longer board project.

But back to the current project. The BASIC ROM is back to normal height, but I've got no hope of fitting the JiffyDOS adapter. A shame really. Has anyone resolved this problem with the longer board? (Edit: Or is this because I'm using the C64C case? I guess it probably is). I was thinking that maybe I could run some wires from the socket to the back of the case somewhere where I place the adapter and chip. I guess I could give this a go. I've already burnt a 27128 with the 16K image all ready to go, so would be a shame not to use it.

I was worried that the fan (as seen in the image attached to the first post) was going to cause problems for the keyboard as well. Looks like it is going to fit, only just. Now I'm thinking that maybe I need to drill a few holes in to the case immediately above the fan.