• Please review our updated Terms and Rules here

Sol-20 questions

falter

Veteran Member
Joined
Jan 22, 2011
Messages
6,573
Location
Vancouver, BC
One of the machines near the top of my acquire list is a Sol-20. The last one I saw for sale went for about $1000... is that pretty typical? Are they comparable in rarity to the Lisa, etc? Havent seen one up in a while now.

Also is there anything to be on the lookout for with these? Things to be desired or avoided?
 
One of the machines near the top of my acquire list is a Sol-20. The last one I saw for sale went for about $1000... is that pretty typical? Are they comparable in rarity to the Lisa, etc? Havent seen one up in a while now.

Also is there anything to be on the lookout for with these? Things to be desired or avoided?


These days, if you got one for $1000, it wouldn't be running and would be a rust bucket.... The Sol-20 is a great system, but they are not the easiest to get and keep running because they are an "integrated" system not like most S100 systems where you could "simplify" the system to figure out what is going on.

If you check eBay history, there hasn't been a working one sold in a couple of years... The cosmetically decent ones have been going for around 2k, with out proof of running...

There are exceptions, my second Sol-20, I grabbed on a buy it now for $500 (It was up for less than couple of minutes when I saw it and grabbed it), cosmetically it was perfect because I don't think it ever worked. It must have been built and shoved in a box when the person couldn't get it working. It took me a long time and required some investment in equipment I didn't own like a logic analyzer and an 8080 ICE to sort it out which turned out to be a defect, there was short under the PCB I had to fix with a razor blade. There are more than a few Sol-20 that never ran back in the day that were purchased as kits, maybe they had similar quality issues. This system also had a factory modification having to do with the transformer having a brown out setting that could cause a problem in the system if you didn't have enough S100 cards plugged in to draw enough current. The factory modification had a minor problem I had to fix.

My first Sol-20 didn't have much electronically wrong with it other than a bad 74H00 and some bad 2102 ram chips, which all was easy to track down with the schematics and an oscilloscope. I also had a bad big capacitor in the PowerSupply which stumped a couple of people who tried helping me out, but in the end it was a simple fix. Cosmetically it was a basket-case. It was completely rusted, and scratched up. I completely tore it down and disassembled it down to every screw. I had to repaint the green on the fan, touchup and clean up the case, wire brush with an air tool the entire chassis, replace the plexiglass, replace the logo strip with a new one and replace some of the card guides. I also sanded down the sides and re-finished them and repaired a couple of other minor cosmetics things as I recall. Oh and it was missing the fuse holder back which turns out to be a very difficult to find a replacement, but I did.

Both needed Keyboard rebuilds, and a calibration of the PLL for the cassette interface. I also performed all the ECN changes for my boards which were Rev D. Finally I replaced every single tantalum cap on the board and in the power supply.

I also revived a Sol-20 at the MARCH museum at the InfoAge Science center. For that one I used an IC tester one afternoon on every chip and simply swapped out any bad chips. It was a factory built unit so it was much easier to be confident if something went wrong it was a chip or something like that. I also had to rebuild the keyboard, calibrate the PLL and replace one of the big caps in the power supply.

Cheers,
Corey
 
I've been doing a lot of power cycles on my SOL-20 recently as I've been trying to bootstrap Northstar DOS. I read Corey's message above and felt quite smug that my SOL has been problem free for quite some time. About 30 mins later I powered it up only to be greeted by a blank screen!!!

ARRRRGHHHH!!!!!

Anyway I have the SOL up and running again now.
It was the 74LS63 @ U40 in the video circuit...I was lucky to find it so quickly. Less than an hour. It's great having everything socketed, because if you can get a general idea where the problem is using a logic probe, and have a good supply of logic chips, I just do quick chip swaps until I find the problem. It's cheating a little, and doesn't always work, but when it does, it's fast and efficient.

Philip
 
Thanks.. yeah if I read correctly there is a much higher proportion of Sols that were built as kits than other computers that were available either way.

Personally I'd be happy with even a non working one. I enjoy the challenge of repair. Hopefully one will cone up fairly soon while i still have some capital in my museum budget.
 
I've been doing a lot of power cycles on my SOL-20 recently as I've been trying to bootstrap Northstar DOS. I read Corey's message above and felt quite smug that my SOL has been problem free for quite some time. About 30 mins later I powered it up only to be greeted by a blank screen!!!

ARRRRGHHHH!!!!!

Anyway I have the SOL up and running again now.
It was the 74LS63 @ U40 in the video circuit...I was lucky to find it so quickly. Less than an hour. It's great having everything socketed, because if you can get a general idea where the problem is using a logic probe, and have a good supply of logic chips, I just do quick chip swaps until I find the problem. It's cheating a little, and doesn't always work, but when it does, it's fast and efficient.

Philip

Philip,
Did you mean 74LS163? I am troubleshooting same exact problem except that the screen goes blank only when the keyboard is plugged in, otherwise I see only 909090 pattern (which I believe is a separate problem)...

UPDATE: confirmed....74LS163 or 25LS163 is installed in (U28,31,33,40)

I will probe this chip though eitherway...

b
 
Last edited:
Ooops, sorry for the confusion. Sloppy.

I have some more SOL-20 fixes (some for screen issues too) on my site here:
http://www.neoncluster.com/home/projects.html

Just scroll down to the 'Processor Technology SOL-20' section.

Although your keyboard/screen issue sounds nothing like anything I've had before.
Good luck.

Phil

P.S. I'll fix the 74LS163 vs 74LS63 mistake on my site too. Thanks for picking up on that.
 
Bill,

A couple of things for you to check with the keyboard.

First is your cable plugged in the right way or is it damaged? I recall when the cable is in backwards on one end, the screen will be blank because the crowbar engages in the supply. Double check what is the voltages look like when you plug the keyboard in to see if that is happening because of a short or the cable.

FYI: If you bring the Sol to VCF East, I can bring a spare keyboard to test it if you want.

Cheers,
Corey
 
I tested the 74S163's on both u28 and u40 with a TTL chip tester. They were both ok, but one seems marginal to me. It has been a while since I had a oscilloscope on the board (2012). AT the time everything was fine and matched signal for signal wth a working unit yet there was no screen display.

I reset all of the switches, which seems to have be kind of arbitrary. Now I get a blank screen regardless of the keyboard. All of the TTL on the keyboard checks out so I have to work on the keyboard board itself.

I need to get some (or borrow temp) 74ls163's and see what happens.

I have been working through the manual kind of slowly, jumping from other projects to this and back again for a few years. No rush.

BIll
 
update

I tested the 74S163's on both u28 and u40 with a TTL chip tester. They were both ok, but one seems marginal to me. It has been a while since I had an oscilloscope on the board (2012). AT the time everything I was getting one garbled line of text at the bottom of the screen. I fixed a chip (u47 / 7410) and system went from generating a line at the bottom of the screen garbled characters to a line of 9090909090's. Lee Felsenstein (in 2012)told me that the line of chars at the bottom of the screen was his "Tom Thumb" diagnostic circuit in action, sort of a diagnostic message, but I need to check my notes as to what he meant and how I can use this. In short the video is ok.

found U113 is bad and needs to be replaced. it's a 4013.

Plugging in the keyboard turns off the display. There must be a short. I will see if I can track it down. The keyboard may not have been finished, C7 was not soldered to the keyboard board. What else could be bad (I need to check)? My IC tester says the chips are OK, but there may be other problems. Lights sometimes respond to keypresses on bare board. I am certain the keyboard is plugged in correctly. I have another keyboard I can rig that will work on this system.

I need to test the RAM. Ignoring the keyboard problem for now, I am hoping that replacing the 4013 and perhaps fixing a few RAM chips will return this machine to the SOLOS prompt.

I have a copy of the manual in the processor tech section of my site.

BIll
 
Last edited:
update - 90909090 means the monitor is crashing (I believe)...meaning possibly RAM is bad. Probed U3--- and found that U5 and U16, which are next to each other, report different results than the other RAM chips. The RAM can be switched around to same effect. So, it's the board.

All but 2102 U5 and U16 Probe pin 11 - steady LO, steady HI, pulse visible
2102 U5 and U16 Probe pin 11 - no/weak LO, steady HI, pulse visible

so I took a look at the schematic, not sure yet what's wrong but I am getting closer. more to come....
 
update - 90909090 on display means the monitor is crashing (I believe)...meaning possibly RAM is bad.

All but 2102 U5 and U16 Probe pin 11 - steady LO, steady HI, pulse visible
2102 U5 and U16 Probe pin 11 - no/weak LO, steady HI, pulse visible

U5 and U16, which are next to each other, report different results than the other RAM chips. The RAM can be switched around to same effect. So, it's the board.

so I took a look at the schematic, not sure yet what's wrong but I am getting closer. more to come....
 
Maybe you have an addressing problem or an issue with the phantom circuit. Do you have a logic analyzer?

If you can capture the addressing of the first 5 or 6 instruction cycles it would let you know what is going on. You could also have a SINP problem. That is the issue I had on one of mine where SINP had a short under the solder mask that must have been a factory defect. A nice sharp razor blade fixed that.

Cheers,
Corey
 
do you have a photo of the underside of your board that I can use for comparative purposes? I am using a logic probe to locate bad signals. U5 and U16 have bad signals on pin 11 compared with the other 2102's. THat's a big clue because they're right next to each other. WOrking through the schematic.
 
Last edited:
do you have a photo of the underside of your board that I can use for comparative purposes? I am using a logic probe to locate bad signals. U5 and U16 have bad signals on pin 11 compared with the other 2102's. THat's a big clue because they're right next to each other. WOrking through the schematic.


I sent you an email with a pic of the bottom side of my sol-20
 
Back
Top