B3-18A Soviet Engineering Calculator

[mvno_subscriber]

First of all, sorry if this is the wrong forum. It's certainly something portable, and I guess it was quite the ace back in its day..


A Russian colleague of mine is moving on to a new position, so I thought I should find him gift. I ended up getting hold of a B3-18 revision A, the first Russian engineering calculator (original is from '76, this is a revised version from '78 ).

This utilized a special microprocessor and cost a couple of months salary for an engineer at the time. I'm refurbishing it to make sure it doesn't act too cranky when my friend gets it (and also I want to look inside of course! :D)

So it came in a leather case w/manual (gotta have that, documentation is half the fun!) and a power adapter. Amazingly it still holds, but either there are some loose connections in the calculator, or the power is wonky. It might seem to fail when having been on for a while.



I really wonder what that "octopus" copper wiring is.

The way it handles square numbers is rather interesting. When pressing x^y, it calculates some kind of intermediate number, and the result of the calculation is approximate. For example, this is me typing 2^8:



Not sure if that kind of behavior is special to this calculator, or if it was common at the time -- but I'd love to know :D

 

[roberttx]

The intermediate number is the natural log of 2.

 

[shattered]

I really wonder what that "octopus" copper wiring is.

It's a transformer, T1 on the schematic.

 

[Dwight Elvey]

e^(8*ln(2)) = 2^8 = 256
I just did it on my Casio.
Dwight

 

[mvno_subscriber]

It's a transformer, T1 on the schematic.

If it's a transformer, then it seems quite broken - what could have gotten it to untangle in such a fashion?

 

[krebizfan]

If it's a transformer, then it seems quite broken - what could have gotten it to untangle in such a fashion?

Each of the separated wires looks to have been inserted into a hole in the board so it looks quite deliberate. Probably slightly cheaper than having extra wires properly connecting everything.

 

[hornbetw]

That transformer probably makes the High Voltage for the VFD display.

Tom

 

[mvno_subscriber]

Ah! That makes sense, thanks. Looking at the capacitors, would it be probable to assume they're gone or weak? They look sealed, but I have no idea about the quality of them. The calculator is acting a bit up, and normally I would replace them regardless, except they're so cool.. I can't really test them unless I take them out, and if I do that I might just as well put in something that'll last a bit longer..

 

[shattered]

Looking at the capacitors, would it be probable to assume they're gone or weak? They look sealed, but I have no idea about the quality of them.

I've no direct experience with electronics repair, but advice on the Internet is to replace these specific models (K50-6, K50-16) always

 

[Dwight Elvey]

If it's a transformer, then it seems quite broken - what could have gotten it to untangle in such a fashion?

The transformer isn't broken. If it were, there would be no display.
If you wiggle the wires, you may be able to break it, though.
Dwight

 

[mvno_subscriber]

So I've replaced all the capacitors, except one that puzzles me. All the other ones had polarity markings both on the capacitor and the board, but this one doesn't. It's markings are similar to the others, but missing polarity. Capacitance meter says 12µF (reading is "10M" which I assume stands for 10 micro). Taken a picture of one of the others, 25V 10µF side by side.



According to the Internet, I might be able to connect two 20µF in series (back to back) to make a non-polarized 10µF cap - anyone got any experience with this?

 

[Agent Orange]

So I've replaced all the capacitors, except one that puzzles me. All the other ones had polarity markings both on the capacitor and the board, but this one doesn't. It's markings are similar to the others, but missing polarity. Capacitance meter says 12µF (reading is "10M" which I assume stands for 10 micro). Taken a picture of one of the others, 25V 10µF side by side.

View attachment 37281

According to the Internet, I might be able to connect two 20µF in series to make a non-polarized 10µF cap - anyone got any experience with this?

The rule is capacitors in series add like resistors in parallel and vice-versa. So, you seem to be correct.

 

[KC9UDX]

Yes but due to the way capacitors age, it's generally not a good idea to do that. A non-polarised 10mmF (sorry, no mu on this keyboard) should be readily available.

 

[mvno_subscriber]

I guess I have other problems right now. Refitted all caps but the unit seems rather dead. No light on display. Except when turning off, then the display flashes briefly (just a bunch of small 0's). I located and fixed a soldering short, but still no luck. It's only a 5V circuit, so I wouldn't think anything could go wrong.. but who knows. It's 23:30 here now, so maybe I better sleep on it.