Looking for MG-17G for calculator restoration

[Paolo Cravero]

Hello.

I think people here like to read stories, so here's mine before the request for help.

Recently I added a 1973 calculator to my small collection: Inno-hit K-80W. It mounts 9 Rodan MG-17G tubes, that are 7-segment cold-cathode displays, a transition between full-digit Nixie and 7-segment VFD. A sort of single-digit panaplex. 1973 is kind-of late for this kind of displays, but that's the datecode on the Texas Instruments calcolator-on-a-chip.

The calculator came as non working. The reason was that batteries had leaked and there was no way to power it up. Interestingly this device worked either with 6V DC or 4x C-sized batteries, and even recharged them! That's an easy fix, also the leak did not seem to have destroyed parts of the PCB.

Once it came to life an electronic problem showed up: digits were not displaying correctly. Basically segments A and F were lighting up together. Curiously, tapping here and there the whole thing, it restored normal operation. After reverse-engineering the PCB I came to the conclusion that there must be a short into a tube since the traces for A and F are never running parallel. With the right ambient light I could see that A&F were dimmer than other segments when F was not supposed to light up. The DVM confirmed the short.

I was afraid of having a bent segment that would have been a nightmare to locate. But looking closely at the tubes I noticed that two of them were missing the top ring: manufacturer difference or fault? Note that all digits were working properly besides the short! Remember those handheld games where you had to control a small ball into a labyrinth (maze?) ? I played that game with the calculator and both rings appeared from behind the tubes and one is now balancing near its original location and the second slid down on the front (visible on the least significant digit in one of the pictures). It could stay like this, as long as the calculator is not moved around.

Else, has someone here some MG-17G awaiting for a proper circuit and housing? I need at least two, but I would got for three to have one extra spare. Please email off-list.

The only reference to these Rodan displays on neonixie is from 2007!

[Nick]

Hi - welcome to the group.

Are you sure this isn't a driver fault or an issue with contamination on the PCB causing leakage?

I have both NOS and ex-equipment MG-17Gs, but I am 3,500 miles from my storeroom! If you can wait until April, I can certainly help with spare tubes.

Cheers

Nick

[Paolo Cravero]

Hi.

Hi - welcome to the group.

Thanks, but I've been around this list for a couple of years or more :) We might have never contributed to the same thread, though.
 

 

Are you sure this isn't a driver fault or an issue with contamination on the PCB causing leakage?

Unfortunately it is not a driver fault (checked with oscilloscope segment-by-segment comparison). I would have preferred to replace a 3-pin device rather than align a long-legged 9-pin component!

PCB looks clean as I inspected it with a magnifier. I think that the calculator was laying on a side when the leak occurred and it stayed mostly in the battery compartment. I found no traces of previous attempts to clean it.

I began suspecting a problem in the tubes when I could cure the fault by gently tapping tubes or the PCB near the tubes area. Tapping/moving other components resulted in no changes instead. The final confirmation came when I was holding the powered-up device and I got a small shock from 160V, so I let it roll on the desk. After this unexpected movement tapping didn't restore normal operation, segments stayed shorted together. DVM confirmed a 0 ohm short between A and F segments. Then I did the maze game, the two rings emerged from the back of their tubes and 3 and 7 look normal again.

This calculator was meant to be portable, and it came in a nice case with handle: back in the old days probably it hasn't been sitting on a desk all the time and the ring connection could have been stressed.

 

 

I have both NOS and ex-equipment MG-17Gs, but I am 3,500 miles from my storeroom! If you can wait until April, I can certainly help with spare tubes.

Thanks! No hurry: the device is supposed to stay with me for many years! And I need to locate a suitable DC socket before bringing it out to the public (office). Just let me know if I should set a reminder in my calendar.

Paolo

[Nick]

On Monday, 12 March 2018 16:17:45 UTC+4, Paolo Cravero wrote:

Hi.

Hi - welcome to the group.

Thanks, but I've been around this list for a couple of years or more :) We might have never contributed to the same thread, thoug

My bad! Apologies for that - I should have checked in the member's list - I recently accepted another member with a similar name... BTW, over 1,000 members now...

 

 

I have both NOS and ex-equipment MG-17Gs, but I am 3,500 miles from my storeroom! If you can wait until April, I can certainly help with spare tubes.

Thanks! No hurry: the device is supposed to stay with me for many years! And I need to locate a suitable DC socket before bringing it out to the public (office). Just let me know if I should set a reminder in my calendar.

Probably a GOOD plan - I'm notoriously forgetful, but I have decent stocks of these, so please remind me in the first week of April!

Cheers

Nick 

[Michail Wilson]

Thought people might like this.

I am thinking of getting the watch just because of the face ability alone.

 

https://www.facer.io/watchface/Q9MWLGtyqg

 

Michail