SCM Cogito 240SR - Operation Fault

This is my 1965 SCM (Smith Corona Marchant) Cogito 240SR calculator. One of my favorite calculators... But it developed a fault after moving it. Bummer...

Unfortunately, as I said, it needs servicing. The machine won't load the accumulator output results (I think) into any register, even though I can manually load data into any register using the keyboard to load data into the "K" register, and the swap keys to manually swap the data between registers. The fact that it recalls ANYTHING at all suggests the delay line is okay. The machine's primary logic consists of 6 large square (about 9.25x10 inches on a side or about 23.5x25.4 cm) PC boards that consist of almost NOTHING BUT resistors, diodes, and transistors (and possibly some hybrid modules). No chips at all! There are 4 smaller PC boards that appear to be associated with the CRT display, and a small pair of PC boards that are related to the magnetostrictive delay line memory. All those boards, and it's wired with board to board plug in wires and cables. There is no simple socketed back plane. WHY can't things be simple!

I would love to map the schematic to this thing out. There are patent documents that describe the basic operation of the machine. Patent 3,518,629; Filed Feb. 6, 1964 and dated June 30, 1970, for Stanley P Frankel. I, and others, have yet to find any trace of actual service manuals or documentation. Only the very segmented diagrams included in the patent. I would LOVE to take my machine apart and map out the entire circuitry of the machine... It is... Most daunting...

A bit of trivia, Stan Frankel was involved with the Manhattan project, but lost his security clearance in the 1950s during the "Red scare" because of his views on making nuclear technology academically accessible for civilian development... The military was NOT keen on that idea. This barred him from the majority of computer related jobs, since nearly every computer application in the 50s was government or military connected. That led him to work with Marchant, who made mechanical calculators back then, to develop the Cogito 240SR electronic calculator.

Still... wow... I am utterly amazed by the skills of those computing pioneers, people who managed to come up with this stuff at a hardware level, and using such primitive technology!

Fortunately, if it's any consolation... the boards appear to have all the components very nicely laid out in a grid. No clue about the traces beneath.

I don't doubt I could map it out, given enough time and unfathomable levels of patience to do it. It's that it only uses about 3 or 4 major part types to perform logic! Even AFTER mapping out the schematic, I'd have to actually figure out and organize what diodes and resistors are grouped together, and what logic function they perform! I have no idea if there is any way to test the boards live either... They are not situated in a means that appears to make that simple. Hybrid modules are also troublesome... They are like chips with no datasheets. Joy.

At least it is still an amazing piece of history, and quite the sight to look at, even if it is not functioning at 100% anymore.

Pics of the inside of my Cogito 240SR: http://richfiles.solarbotics.net/calc/CogitoOpen.jpg http://richfiles.solarbotics.net/calc/CogitoInternal.jpg http://richfiles.solarbotics.net/calc/CogitoCRT.jpg http://richfiles.solarbotics.net/calc/CogitoPCB.jpg http://richfiles.solarbotics.net/calc/CogitoWires.jpg

Details on the Cogito 240SR http://www.oldcalculatormuseum.com/scm240sr.html

Frankel's Calculator Patent http://www.google.com/patents/US3518629

History of Stanley Frankel http://www.hp9825.com/html/stan_frankel.html

Youtube is at it again. Do your part in resisting the corruption that is destroying Youtube... And if all else fails, JUMP SHIP. I think it will be worth ALL our efforts to look at new services like Bitchute. If you too do not approve of Google's cyber bullying tactics, please consider fighting back. ... https://www.youtube.com/watch?v=l74BJeY_fLY