When [Kai Robinson] found himself faced with the difficult task of saving as many Mac SE’s as he possibly could, the logical but daunting answer was to recreate the Mac SE logic board for machines that would otherwise be scrapped. These machines are over 30 years old and the PRAM battery often leaks, destroying parts and traces. Given that the logic board is a simple through-hole two four-layer board, how hard could it be?
The first step was to get some reference photos so [Kai] set to desoldering everything on the board. The list of components and the age of solder made this an arduous task. Then a composite image was produced by merging images together using a scanner and some Inkscape magic. in graphics software.
Rather than simply putting the pins in the right place and re-routing all the netlists, [Kai] elected instead to do a copy, trace for trace of the original SE board. [Kai] and several others on the forum have been testing the boards and tracking down the last few bugs and kinks in the design. An unconnected pin here and an improperly impedance matched resistor there. Hopefully, soon they’ll have Gerbers and design files ready for anyone should they need a new logic board PCB.
It’s no secret that we love the Macintosh SE here at Hackaday. We’ve seen new custom cases for it and now new PCBs for it. It does cause the mind to ponder though and wonder, what’s next?
Thanks [Toru173] for sending this one in!
Tracing is probably a pretty good idea, since it would be a kick in the groin to do all that work and testing, and find out that you have bunch of interference problems that you would end up solving by… reproducing the layout that Apple engineers came up with 30 years ago.
I thought this would be about making new Mac SEs.
I traced out most or al of the Mac Plus logic board, just following traces by eye and ohmmeter. But lots of it is mystery, either Apple components (like the Woz disk controller) programmable arrays.
The analog board was easier, especially after I realized the switching supply was likely like that in the Apple II. Once I redraw my raw work like the Apple II supply (the schematic in the manual), it mostly lined up.
But this was just about replacement boards, so stripping parts to trace was viable. And mystery parts not needed, since they can just be transplanted from the bad boards.
Oh hi. Few things to clear up – I use Sprint Layout, not Inkscape. Secondly, the board is four layers, not the two initially assumed (I did mention this early on in the thread).
Also, the gerbers can be generated from the layout file attached in the thread, they’re just not posted on github yet due to nailing down final issues.
Finally, the GLU custom chip was reverse engineered and the BBU Chip is next in line.
Also… Is it not custom to give someone a heads up about posting articles they feature in? A simple hello would have sufficed. :/
That’s just how this site works, it aggregates things users submit. Happens to me, too. I’m like “Oh. Hi HaD”
How many simultaneous page loads can that nifty PHP based site of yours handle?
Who knows, it hasn’t been touched in over a decade and I’m not the host
They just don´t do that. But they introduce mistakes and approximation by not caring to read the original source.
Yeah, wait ’til you have the entire comments section screaming at your project for doing something really stupid, when it was described wrong in the article :-D
Nobody can say how long my SE/30 will remain stable. I’ve had it open, but done absolutely nothing with the board. (I changed the PRAM battery in, like, 2001.) It was booting into the screen-of-thin-and-thick-lines for a while, but now consistently boots just fine. I will use the ancient technology of keeping my fingers crossed — and be sad when that proves to be ineffectual, like so many technologies that predate Mac SE. Thing is, it will be a case-study, to count the days that it boots to a legible screen. Every day, I can snap the power button, marvel and put a naively self-deluding check-mark on the calendar. And oh yeah, it even still boots from the floppy drive. Really — the floppy drive is not sclerotic.
If you haven’t recapped it already, please do. What kills the SE/30 isn’t usually the PRAM battery– it’s the surface mount caps. Same for most late 80’s and 90’s Macs.
Yeah if you can change your motherboard capacitors (or make them changed) do it ASAP, and also please check your 20yeas PRAM battery. Also the 2 capacitors right at the front of the floppy disk drive are prone to leaking so change them too to restore a finicky/non working FDD (you can change other SMD caps on it but by experience it’s only those two in the front that are prone to die).
I for one would like to see an article about impedance matching resistors.
Came to look for the same …
https://hackaday.com/2016/02/21/retrotechtacular-transmission-lines
meybe some petition to Apple for publish source code of old os 6,7 ?
many people still love gui, fastest program and small fir to microcontroler
Why not just de-solder it all, then rub/solve off the solderstop and make copy with a scanner?
That’s almost exactly what I did. Well, aside from removing the soldermask. The thread will link to the article, where I describe the process :)
I wish there was a vendor that sold these logic boards as new parts. My SE has the thin white vertical line and I’m pretty sure it’s the logic board. Many old electronics especially Nintendo products have a market to sell these parts. It’s a shame it isn’t the same for the classic Macs.
Well, once the boards are verified working – i’ll sell them through someone, or direct from tindie, for about £29 each.
Good to hear!
Congratulations on a job well done! 😎
I really appreciate it. Thank you. Projects like this preserve our history and allows our grand grand children to relieve our days, so they may have a better understanding of our times, too.
Also, a project like this inspires and motivates others to participate.
And Kai has done it – it works!!! Awesome work, Kai
https://68kmla.org/forums/topic/60059-reverse-engineering-the-macintosh-se-pcb-custom-chips-for-11-reproduction/page/32/
Thanks for the heads up !
Thanks! I have 3 matte black boards left from the special order run left, then it’s blue boards only for the subsequent runs.
I have a Mac SE 20 that I am troubleshooting to revive. So far it has 5 bad capacitors. These are N8720 and all look the same: 1mF 50V axial capacitors. Can you suggest where online would be best to get these?
1uF Capacitors – the axials, can be bought from Mouser: https://www.mouser.co.uk/c/passive-components/capacitors/?capacitance=1%20uF&product%20type=Electrolytic%20Capacitors&termination%20style=Axial