Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

How does this 60's transistor calculator function?

Status
Not open for further replies.

The Guru

New Member
Hello all,

Recently I've been looking into old calculators using pre-LCD displays.

I've always loved the look of nixie/VFD tubes or those old, small, LED displays, and after some research, I stumbled upon this particular video:

As the owner pulls the awesome-looking thing apart, it seems as though every tube has a separate board.
I assume every such board has a display driver circuit on it to convert binary to decimal.

What surprises me, is that besides the display boards, there seem to be only two moderately large, yet single-sided boards, and one similar in size to the display boards, yet the calculator can perform addition, subtraction, multiplication and division!

Could anyone please explain to me in rough lines how this beast works? Or perhaps point me to a page with more information?
I have a basic understanding of digital logic (adders, 2's complement, the logic gates).

I'm very interested in logic devices like these built out of discrete transistors, so if anyone has any similar information, it would be greatly appreciated if you shared!

Thanks in advance,

Jeroen - The Guru
 
Last edited by a moderator:
That probably cost some BIG BUX in 1966. Very labor intensive indeed to assemble that.

As for how it works, if you take a good look at the circuit boards, they're filled with essss-loads of diodes. They were using diodes to implement AND and OR gates, and buffering with transistors. The circuit itself probably didn't operate in binary, as this is a specific function that would more easily been handled in BCD directly without the conversions.
 
Beckman used to make nixie tube modules in the '60's. I don't recall them looking like yours, but I haven't seen them all either.
 
Can you imagine what that beast cost, all to do what a $1 calculator can do today. ;)

Edit: Here's some info on it.
 
Last edited:
Our chemistry department had one that looked much like the one pictured. There was a metal rod around the perimeter inside the case. That functioned as delay line for data that were introduced as vibrations in one end. There were obviously other ways for memory.

John
 
A good example can be found in the old HP 5245L Frequency Counter Manual. A classic 50 MHz mainframe counter of its day. Each nixie tube has a driver board. That was the first counter I worked on. Another good example that pre dates the Nixie tube is the very old HP 524 counter that uses valves or vacuum tubes. The linked to manual describes the theory of operation very well. Pretty cool old stuff.

Ron
 
I've made a few nixie clocks, but never messed with nixie calculators.
 
Thank you all for the replies!:p

The circuit itself probably didn't operate in binary, as this is a specific function that would more easily been handled in BCD directly without the conversions.
Hmm, that would make sense.


Thank you crutschow for the link!
This info confirms that the machine uses BCD.

And thank you Reloadron! A fantastic document as far as I can tell! I will make sure to give it a complete read soon.

Phew, this is going to take some time... I'll make sure to post a thread if I manage to finish a calculator:p
 
Is this the same machine as threaded on the vintage radio forum?
 
I worked on computers built like this calculator. I remember a 3x4 inch PCB labeled "2-D FF" that was two flip flops.

I can't see well but I think each tubes is driven by two PCBs. I think there is a BCD adder / shifter one one board and the other is a BCD to 1 or 10 driver.


The machine only needs to add, (not--add is subtraction) and (shift--add) multiplies. Many of the boards have the same function so I would shift the boards around for testing.

A hand full of diodes + resistor = gate. AND, OR
a transistor + resistors = inverter
Two transistors + diodes + resistors = simple FF

Because these gates have a race condition there are often RCs to cause time constants for delays or edge triggering.

I had bad problems with the connectors between boards. I needed to clean the gold contacts many times. The wires between boards have a bad problem of breaking.

Thank God we now can design a 1,000,000 transistor project and have it fit with in 1x1 inch.
 
A good example can be found in the old HP 5245L Frequency Counter Manual. A classic 50 MHz mainframe counter of its day. Each nixie tube has a driver board. That was the first counter I worked on. Another good example that pre dates the Nixie tube is the very old HP 524 counter that uses valves or vacuum tubes. The linked to manual describes the theory of operation very well. Pretty cool old stuff.

Ron

This is a fantastic document!
Thanks for posting.

It shows the exquisite complexity of an instrument that nowdays could perhaps be constructed with an IC or two...although not reading 50 Mhz.
 
It shows the exquisite complexity of an instrument that nowdays could perhaps be constructed with an IC or two...although not reading 50 Mhz.

Attached is a data sheet of a 2.5ghz divide by 4 that makes a good first stage.

Altera makes FPGAs (Stratix III) that could easily do all the logic for a 500mhz counter in one IC.
Its memory can be clocked at 600mhz. A 32x32 multiply takes 2nS. I have used older/slower parts with 750mhz serial IO. A 16 digit counter takes only a tiny piece of a Stratix FPGA.
 

Attachments

  • MC10EL33-D.PDF
    146.7 KB · Views: 218
Thanks for the update Ron... indeed single ICs now have more features and functions that full equipment racks from the 60s.

I'm fascinated however, by equipment from the early 1950 to 1960s, prior to the ICs becoming mainstream. To build something like that, it required a design engineering expertise and manufacturing build quality that just boggles my mind.

I remember visiting as a student a local TV station. The sync generator, which supplied the V and H signals for all the cameras equipment, was a 6 foot rack with perhaps 80 tubes. And this was still B&W, no chroma generation or insertion yet!
 
If you delve into something like an arm processor you'd find a lot more going on, modern technology to an extent hides the gears turning.
 
I'm fascinated however, by equipment from the early 1950 to 1960s, prior to the ICs becoming mainstream. To build something like that, it required a design engineering expertise and manufacturing build quality that just boggles my mind.

Hold on to your hat. The future is coming. What will 2050 look line?.....unless some fool Fs it up.
 
Hold on to your hat. The future is coming. What will 2050 look line?.....unless some fool Fs it up.

Beats me but born in 1950 in 2050 I would be 100 years old. Don't plan on being around that long. :)

Ron
 
We'll probably be a third world by then, probably just as well I wont be around either and if I am I'll probably be round the twist.
 
I can't see well but I think each tubes is driven by two PCBs. I think there is a BCD adder / shifter one one board and the other is a BCD to 1 or 10 driver.


The machine only needs to add, (not--add is subtraction) and (shift--add) multiplies. Many of the boards have the same function so I would shift the boards around for testing.

A hand full of diodes + resistor = gate. AND, OR
a transistor + resistors = inverter
Two transistors + diodes + resistors = simple FF

Hmm that certainly would make sense... Thank you very much!
I don't own this specific calculator though, sadly.

Thank God we now can design a 1,000,000 transistor project and have it fit with in 1x1 inch.

LOL:p
 
I'm fascinated however, by equipment from the early 1950 to 1960s, prior to the ICs becoming mainstream. To build something like that, it required a design engineering expertise and manufacturing build quality that just boggles my mind.

You and me both!
 
Status
Not open for further replies.

Latest threads

Back
Top