Casio AS-A 1969 Nixie Calculator - Division Issue

[DrAlis]

Dear all, I got this calculator that works flawlessly except for a intermittent (but most of the time) division issue. Entry/Plus/Minus/Multiply work reliably.

Attached you find the manual in German, schematics of the very similar successor of the model (AS-B,121-B, slightly higher integration but similar ICs), fotos of the board layout of top and bottom PCB with components, nixie display after some failed division (might also be during infinite loop actually).

I am looking for a particular place on the board to resolder or component to swap. I tried following the KDIV signal in the schematic but got lost quickly. Looking forward to seeing some Genius do magic. Thank you.

 

[daver2]

Ah, that calculator schematic looks very familiar!

Here is the reset / power on clear circuit:


When you power up the calculator, there should be a pulse from each of the signals nINIT and POC.

nINIT should initialise the calculator logic.

POC is in parallel with the 'C' key - to 'clear' the calculator.

Monitor both of these signals when you turn on the calculator and see what they both do.

If there is no delay between powering up and the signal changing state (after a period of time) then there is probably a 'flaky' component here somewhere - we just have to find out where...

I hesitate to say a voltage level, because this calculator works from a negative power supply, and my brain is not in a good place at the moment working with negative logic voltages - I have been working all day with positive logic voltages (+5V and +12V) so I am bound to make a mistake!

If you can identify the value of some of the components, we can simulate the circuit and see what should happen...

For the intermittent division problem you will have to isolate the device with freezer spray and a hair dryer.

Dave

 

[DrAlis]

Hi Dave, thanks for the analyis.

You jumped to the other casio as-a (based on my other post) which I am confident has several issues. Well, any idea where this init circuit is physically? I got no idea where to look on the PCB.

Coming back to this casio as-a. this one only has the division issue. The freezer/heater thingy sounds good but i doubt i can operate it open and play with temperature enough because to check i got to power cycle and press 5 keys at least. also focus on the ics with the heat or diodes or quadrants or what?

I was hoping that since the fault ONLY affects division we can narrow it down to a hand full of things from the schematic and then start freezing/heating. there cant be that much circuitry that is unused for add/substract/multiply/display etc. what do you think?

 

[daver2]

You will have to do a bit of hunting.

The POC signal is connected to the 'C' key. So follow one side of the 'C' key (not the side that is common to all of the other keys) and see where it goes. We are not interested in ICs - we are interested in 3-legged transistors

Ditto for the nINIT signal. This is connected to IC 1-1B1 pin 15 (if you can find that IC).

Dave

 

[DrAlis]

Thanks! For the power init I understand to look for transistors.

But for the division error (I like to fokus on that one as it is more reproduceable and not part of a multi-issue board) the question was what components/traces to focus on.

 

[DrAlis]

Alright, reporting back of a many many hour hunt. Intermittent faults are the worst.

I was convinced at least 20 times that I found the problem, just to realize after resoldering that I did not. I believed the problem was located on the top right corner of the lower PCB after dozens of reheat and cools while frantically keying stuff in blindly on an upside down keypad with 200V+ sitting on the open board. I ended up redoing around a hundred solder joints, just to find the problem to reappear. Sometimes inbetween board lost function even more up to completely. Then I was convinced an IC is at fault, particularly the NEC PD13C in that area. Then, I wasn't sure if maybe some nearby diodes are the problem. Then, I thought maybe the issue appears not based on the local heat but based on the duration it was switched on.

Then I decided to give up.

In a last desperate attempt I soaked the top right corner in isopropyl a second time. Everything had an isoprop bath already. And WTF up to now the problem seems to be gone. I am so confused. Why? I already cleaned it before! In the desperate human attempt to make sense of a incomprehensible complex world, I tell myself that there was some conductive or capacitive dust or dirt stuck somewhere that I did not catch in the first round. But success so far, with a terrible process. I am still amazed how much beating that thing can take. There have been so many crappy solder joints and I surely added some while removing some.

And now division works and the other things I broke on the way as well (again). Thanks.

 

[DrAlis]

Thanks for the much needed upvote after this odysey. Unhappy to report today, the fault came back after I could not reproduce it for a full two day. I guess, I'll run over this calculator with a truck. Cheers!

 

[daver2]

Statistics I am afraid...

Dave

 

[DrAlis]

Yeah, I put so much love and effort in but it was not honored. The intermittent nature and lack of spare ics and knowledge makes it not feasable. it is a pretty piece of tech and the linear reed switches and pretty nixie display make it a beautiful vintage piece. if someone in the eu wants to take this on let me know i ll gladly sponsor shipping for a date.

 

[wrm]

Thanks for the much needed upvote after this odysey. Unhappy to report today, the fault came back after I could not reproduce it for a full two day. I guess, I'll run over this calculator with a truck. Cheers!

Don't do that, it's still a display piece and history and stuff. They don't make 'em anymore...

Where in .eu are you?

 

[DrAlis]

Don't worry I am not going to hurt this pretty piece of tech. Just expressing my frustration about failing to repair despite feeling so close to succeeding. I am in southern Germany.