Digitec United Systems Corporation 262B Multimeter Repair

[Prime]

Hi.

I'm hoping someone can help me ID these chips and possibly offer up a manual for this one.  It's a battery powered Nixie Tube Multimetermeter.

The meter turns on and does measure but the tubes are erratic and I'm missing a digit in the value.

Swapping the chips did seem to gain me another digit so I suspect one of them might be faulty.

Can anyone ID these?

Sent from my HTC One M9 using Tapatalk

[SvanGool]

A very long shot: https://www.eevblog.com/forum/reviews/lets-see-your-nixie-tube-equipment/msg992615/#msg992615
If you look at the third picture from the top at the up-standing PCB: IC4 is an SGS 7517.
SGS and Fairchild had a partnership in the period of 1960-1968.
That at least broadens your search scope.

Maybe "Gyro" can help you further with this, he/she is still very active on the EEVblog fora.

[Gyro]

Wow, you have a sharp eye and encyclopedic memory SvanGool!   

Unfortunately not quite right though, digging out the counter shows that IC4 is actually an SGS 995879 the datecode is 6910 though - that whole row are decade dividers by the way. All the ICs are Fairchild / SGS 99xxxx series logic, date codes 6910 to 6933, so it pre-dates the OP's meter by a couple of years. For reference the BCD to Nixie drivers are 996079. The logic supply is +3.8V suggesting RTL logic(?).

Fairchild 75 series does ring a bell though - I'm sure I've seen it somewhere, so hopefully someone has data archived somewhere.

@Prime:

Which ICs are you swapping? It looks fairly clear that the F 7510 chips are the Nixie drivers, almost certainly BCD input, with the F7516s being decade counters (dividers). Depending on the swapping results, this will indicate whether you have a faulty driver or counter.

Also checking the supply voltage on the 7516 (probably pin 14) will help to establish what logic type. If it's +5V then it gives a better chance of finding a TTL substitute. Note, watch yourself on the approx 200V Nixie supply.

By careful (power off) tracing, you may be able to get a reasonable guess of pinouts - there should be 4 bits of BCD data running from the counters to the drivers and a carry in/out signal running between the counters (highest frequency input on the least significant digit). Verify supply and ground pins etc.

Also check the sockets carefully and look for poor solder joints (although from your description, it does sound like a duff chip).

Hopefully others with long memories will come along too!


P.S. You probably also want to do a 'Report to Moderator' and ask a Mod to move (please don't duplicate) your thread to the repair section, it really ought to be there rather than Metrology - and you might get more hits there too.

[IconicPCB]

I have Fairchild 1972 logic catalogue.. an orange red coloured tome dealing with 99xxx chips but no reference to any 75series part numbers.

[Gyro]

Thanks for the pointer IconicPCB.

I found the 99 series (the numbers are truncated to 4 digits by the way) in the 1969 Fairchild Logic (Ch1-5) at bitsavers.org... http://www.bitsavers.org/components/fairchild/_dataBooks/. There's a 1972 TTL databook there too, but no mention of the OP's 75xx in either of them that I can see. Looking at the 1969 one, there's a combination of DTL, RTL and various forms of transistor logic in the same book with part numbers jumping all over the place - it was a real mess back then!

@Prime:

In addition to the bitsavers.org archive that I linked above, the other useful place is Archive.org: https://archive.org/search.php?query=fairchild . There are a bunch of old Fairchild databooks in there too. A quick scan didn't show up any 75 series but you probably want to go through them more carefully. Don't give up hope just yet, there must be a databook somewhere.

[IconicPCB]

I also have a 73 Fairchild collection of digital app notes.

[kfuelectron]

Yeah, bumping this ancient post.

I just came into possession of this exact meter. Like yours, it was in a non-functioning state. This nixie tubes fluctuated wildly and sometimes had non-responsive digits. The meter did not register any kind of reading, no matter what setting I tried.

As standard, like any repair, I disassembled the unit down to the individual boards and inspected each side carefully. No PCB discoloration. Traces intact. Good solder joints. No burned components. Hmm, slightly discolored driver sockets, a white powdery oxidation?

After taking pictures, I pulled all socketed ICs, and noticed the pins had the same powdery oxidation as noted before. All I can think is some kind of reaction of dissimilar metals in the IC pins and socket contacts. Well, I cleaned all the pins with a fiberglass brush, applied a light coat of deoxit, put everything back together.

Guess what? It works perfectly! Surprisingly accurate too.

Long story short, clean the socketed IC pins, they are old and have oxidized. Maybe yours will start working too?

Good luck.

[emanarias]

The ICs marked 996079 are Fairchild 960 decimal decoder and driver.  Similar to, but a different pinout from, a 74141

995879 are Fairchild 958 BCD Decade Counter with Reset, comparable in function to a CD4510

I have the manual to this meter, and can scan it if you would like.

[IconicPCB]

A possible culprit one of the old style carbon composition resistors. They tend to go high impedance.
Just check their resistance value.
You may be lucky.