You’re fortunate that the connectors are standard; 22 contacts sounds familiar. Back then, diallyl phthalate was one of the best plastics for connector bodies, and it might still be. It was a bright green, although I never learned whether that was simply a good dye, or the inherent color of the polymer. Gold is not a luxury in such connectors! Except in unusual situations (iirc it catalyzes airborne chemical compounds, but those aren’t commonplace), it remains clean and essentially ensures good contact. The plating is very thin, on the order of microinches.

By now, there must be more to say, or else the task was almost forbidding…I’ll re-read your message.

I do hope you’ll have a working machine!

This sort of design approach was, it seems safe to say, what Bob Ragen had much in mind when he conceived of the “utterly-serial” EC-130 calculator. By storing internal numbers in the delay line, he saved oodles of discrete components.
True, cores economize to some degree (one flip-flop per bit is just about hopeless), but core memory requires a fair amount of circuitry, a great deal more than required by the delay line.

This is not meant to disparage the Busicom machine; it seems to have been typical of a ‘textbook” approach to design.

Off-topic, mostly — I well remember the almost-tiny Sony Sobax calculators that were displayed at the ’64-’65 World’s Fair in NYC (Queens). They used Nixies.

Best regards,
[nb]