andrewb
Well-known member
@lilypad19 , where did you get that ribbon cable interconnect kit?
He made it. That was something I wanted to do for years but never got around to it. But he came up with the same idea and actually built it. Now I send everyone looking for SW repairs to him, because I hated working on them.
Just as a reference, you can't get all of the chips needed from Digikey and Mouser, a lot are simply not available. Especially the math box chips.
The chips that aren't available new are typically the ones that are (or should be) socketed. PROMs, 2901's, DAC's, etc. That's the stuff you want socketed, and you want to test any that you buy (whether old stock or new from China) before you use them.
Regular TTL chips, as long as they come from Mouser/Digikey/etc, are rarely bad, and you're relatively safe not testing every one, and soldering them in.
I like @andrewb and admire his work. That said I knew this thread would trigger his "you should have just repaired an original one". Yes you can buy a board set for $500-600 and then send it out for repair. But what is learned? Building your own set even if it had to be repaired afterwards is a fantastic learning experience. Myself, I feel that the money spent is like like paying for school.
The thing is, soldering a bunch of parts to a board doesn't teach you that much.
If you want to learn to solder/desolder, watch a couple of youtube vids and practice on a junk board, where you can make mistakes, pull traces, and actually get experience doing things the wrong way (which is important), while you figure out the right way.
If you really want to learn *troubleshooting*, a repro board is a terrible way to do it, because there are likely multiple problems, and that makes figuring things out exponentially harder. Original boardsets tend to have fewer things wrong (a majority of them tend to have one core thing that is preventing them from booting), and they are easier things to figure out.
Troubleshooting one of these repro boards is more advanced, because there are usually multiple things wrong at the same time, and is usually involves fake parts, and/or incorrectly installed parts, neither of which you're going to have on an original board, which you know was working at one point.
If your goal is learning, the best thing is to buy a dead original boardset and dive into trying to figure it out. That's how I learned. I studied the archives here, Vectorlist, RepairFAQ.org, sci.electronics.repair, FTP archives of repair docs, and various repair weblogs for years, while I lurked here.
And then I've spent 8 years *practicing*. Repair is a craft, like playing a musical instrument, and you only get better by sucking at it at first, then continually doing more of it. That's what I love about this hobby, as even after many years and hundreds of boards I am still learning new things and improving my skills. I still get boards that kick my ass. But I stick with every one until I figure it out, and the hardest ones end up teaching me the most.
There is so much knowledge out there that people have taken the time to write down, that hardly anyone reads or looks for anymore. Mountains of info literally at your fingertips, but people just don't want to spend the energy to google and study it. They'd rather go to Facebook and ask their friends (who don't know anything). And now it's going to be an AI that gives you shit information. But skills can't be downloaded. They have to be practiced.




