Fluke 8125A military DMM teardown

[tggzzz]

I've got test equipment with LCDs, CRTs, LEDs, Panaplex, and Dekatron displays, but no Nixies - until now. (N.B. this "completist" mentality is actually under my control since I will ignore any attempts to get me interested in even more esoteric displays. Honest.)

Then by chance I bumbled across a military DMM with what looked to be nixies on fleabay. Since the vendor had omitted almost all sensible search terms and marked it "for collection only", I put in an excessively low bid and got it for half that price. I went to the end of the road (literally) in the back of beyond (not much of an exaggeration) to pick it up from the old sawmill in a boatyard that was closing down, and since the petrol costs dominated the puchase price so I gave him my maximum bid price.

The vendor said it was working but it didn't have any leads. That was a concern because the mains socket was strange, being on the front panel and secured by a screwthread. Oh well, I'm sure I could recover my money by stripping it and flogging the nixies.

So, I returned home and almost used a hose to get the crud off this thing's outside, but thought better of it.



The outside of the cover has a spring-loaded handle (like a wimpy mousetrap) that is protected from knocks by some more metal. The pressure equalisation air vent is also visible. Presumably they they are concerned that during air transport the case might imitate elderflower fizz in a plastic bottle, or that higher external air pressure could prevent the cover being opened. Closing the vent keeps water out.



The cover is held on by catches that will really annoy mechanical engineers. Mechie's hate it when people think they can make everything from bent metal and wire, but here's proof you can make milspec catches from bent metal. They work surprisingly well, and won't get knocked open as easily as traditional over-centre clamp mechanisms.



The inside of the cover has a piece of carefully shaped plastic which will knock the power switch into the off position when the cover is closed. There are similar things on my Tek 1502 TDRs, and in both cases it will avoid unnecessary careless discharge of the battery.



Taking off the cover reveals the front panel, with the mains socket at bottom right.



Even the fuseholders are protected against water ingress. Even so, I again resisted the option of removing the crud with water.



Delving around in the cover reveals some mouldy old probes (yuck), and a mains lead (yay!). Now I know why someone was able to believe it might be working.



Taking the guts out of the frame showed that it was beautifully clean inside, no signs of NiCd battery leakage. It also neatly demonstrates the old "right to repair" ethos: there are silkscreened instructions on how to dismantle it correctly and on the calibration points.



So let's plug it in and see what happens... it lights up, but not all in right places - no nixies.



Oh well, onward...

[tggzzz]

The inside of the green case shows, if you look carefully and know what you are looking for, a 1" gap between the inner frame (and electronics) and the green outer case. (It would be easy to see in a stereoscopic photo, but very few people have suitable red/blue anaglyph viewers to hand!). That gap prevents "unauthorised repair by kicking" or, more charitably, reduces damage when run over by a car. Under the perforated metal grille there are paper bags of dessicant, and there is a humidity indicator visible.



Removing the metal box on the top of the guts shows us the battery compartment is clean; maybe the 21.6V 1.2Ah battery will work and I won't have to replace the 18 cells.



But then since the battery is from 1981 and last charged in 1982 (yes there is a printed record!), maybe not



Anyway, now we can see the interior in all its glory. Shiny.



But when we look more closely, it looks like someone has been bodging. Odd heatshrink, jumper leads clipped and with solder blobs. That would be a real pain, since there isn't a manual let alone a schematic.




But then maybe not, since the bodged jumpers are inserted into dedicated PCB holes and there are other intact jumpers.



Looking around at more of the guts, there are the usual blue Philips electrolytics and some wet slug tants. None of them show any leakage. I also find a large 250mA fuse that has blown. I don't have one in my stock cupboard, but there is an intact fuse clipped into the cover.

There's no manual for the 8125A, so for the next steps I have to use the manual for the 8120A, which appears to be a bench DMM rather than a muddy field DMM. Fortunately it is pretty similar in most (but not all) respects.

It turns out that the bodged jumper leads are designed in bodges, a variant of select-on-test components. The ones that have been fiddled with are for the DMM zero, for when the DC zero trimpot isn't sufficient!

So, replace the fuse, turn on, apply a 9V input and .... bingo!



The lighthouse has almost disappeared; the bulb limits currents through the NiCd battery. It looks like that was shorted so I removed it, and the current is back to normal.

I fiddle around with the calibration and find that the jumpers that were cut shouldn't have been cut. Typical.

Trying it with 1/10/100k resistors also gives good measurements, and I can't be bothered to try it with AC volts.

So, while not in wonderful calibration, it is certainly working well enough for me be satisfied that I have nixies, and to call it a day. That also means I can ignore recapping and buying NiCd cells.

Since it is military and my house is full, at least I can legitimately keep it in the garden shed

[bd139]

Very nice meter that. Nice to see some good old "no expense spared" design. Thanks for the detailed write up. 

[das_strobel]

Cool meter! Thanks for the tear down. Were you able to determine the approx build date based on some date codes? Looks surely like from the 1960s considering the Nixies.

I hope I never find something like this. I will need to keep it but I already have enough of this old stuff and no room.

[tggzzz]

Cool meter! Thanks for the tear down. Were you able to determine the approx build date based on some date codes? Looks surely like from the 1960s considering the Nixies.

I hope I never find something like this. I will need to keep it but I already have enough of this old stuff and no room.

Quite

Early 80s. Battery was made in 1981, and the few ICs are consistent with that. But the overall design and construction does feel like the 60s, using capacitors and FETs to store bits

[das_strobel]

Yes, the design is probably much older. Early 70's maybe. Considering the ICs 60's is probably a bit too early. By the beginning of the 80's LEDs were already commonly used,I think. I have a Tek 475A scope with a DM44 built in 1978 that has 7 segment LEDs.

Military tech probably used rock-solid and field proven design for much longer.

[Dr. Frank]

Thanks for sharing this nice piece of equipment..
It seems to be from about wk 10 / 1975, that's the date code faintly visible on one of these logic ICs.
Finding younger date codes indicates that this box had been repaired several times 'in the field', which is very common for military equipment, also these bodges may originate from that.
Probably there's also a civil version of this DMM. (edit: seems to be similar to the 8100A, which dates from 1969 - 1972)
Frank

[tggzzz]

Yes, the design is probably much older. Early 70's maybe. Considering the ICs 60's is probably a bit too early. By the beginning of the 80's LEDs were already commonly used,I think. I have a Tek 475A scope with a DM44 built in 1978 that has 7 segment LEDs.

Military tech probably used rock-solid and field proven design for much longer.

I don't disagree, but in ~1974 I built the first digital clock anybody in the road had seen, with 0.7" LEDs. It still exists, and I recently turned it into a Vetinari clock https://entertaininghacks.wordpress.com/category/digital-clock/

[bd139]

Nice. All those lockfits though 

[tggzzz]

Thanks for sharing this nice piece of equipment..
It seems to be from about wk 10 / 1975, that's the date code faintly visible on one of these logic ICs.
Finding younger date codes indicates that this box had been repaired several times 'in the field', which is very common for military equipment, also these bodges may originate from that.
Probably there's also a civil version of this DMM. (edit: seems to be similar to the 8100A, which dates from 1969 - 1972)
Frank

Even before posting, I irritated myself by not noting the datecodes

Looking at a higher resolution photo show two ICs are 7510 and 7629, a tant is 7643, but I can't read the 7447 datecode. I'm going to be opening it up again tonight to show people at the local hackspace (and to reinsert some "HumiSorb" bags), so I'll check then and add a photo of the case interior.

EDIT: the 7447 is also '75 vintage. It looks like the NiCd battery was replaced in '82 and was probably never used - certainly the written "recharge record" is blank.

[tggzzz]

Nice. All those lockfits though 

At least they are all the same. Most of the stuff I built then used random components removed from defunct circuit boards. I still have some of them

[bd139]

Good point. I remember having to do that as well. Even worse those bloody bi-pak UMUT transistors were all I could afford.

[Gyro]

They were pretty concerned about it getting too cold -  Big lamp in the rear, array of power resistors in the base, low temperature indicator at the front!

Hopefully there's a thermostat for desert conditions somewhere.

[tggzzz]

A few extra pictures which I'll also retrofit into the first two posts...

The outside of the cover has a spring-loaded handle (like a wimpy mousetrap) that is protected from knocks by some more bent metal. The pressure equalisation air vent is also visible.



The inside of the cover has a piece of carefully shaped plastic which will knock the power switch into the off position when the cover is closed. There are similar things on my Tek 1502 TDRs, and in both cases it will avoid unnecessary careless discharge of the battery.



The inside of the case shows, if you look carefully and know what you are looking for, a 1" gap between the inner frame (and electronics) and the green outer case. (It would be easy to see in a stereoscopic photo, but very few people have suitable red/blue anaglyph viewers to hand). That gap prevents unauthorised repair by kicking or, more charitably, reduces damage when run over by a car. Under the perforated metal grille there are paper bags of dessicant, and there is a humidity indicator visible.

[bd139]

That’s a pretty well designed bit of kit. I’ve seen stuff that was supposed to be in active field service that was built worse. Husky Hunter comes to mind.

[tggzzz]

The 8125A weighs 20lbs/9kg. Makes me appreciate the virtues of my main handheld DMM, a Fluke 25. But that is nixie-less

The person that sold the 8125 to me said the previous owner bought it as a bench DMM - and realised the mistake when he received it!

[windsmurf]

What's the purpose of the light bulb in the back?

[bd139]

The 8125A weighs 20lbs/9kg. Makes me appreciate the virtues of my main handheld DMM, a Fluke 25. But that is nixie-less

The person that sold the 8125 to me said the previous owner bought it as a bench DMM - and realised the mistake when he received it!

Yes it’s definitely a “back of a Land Rover” DMM that one.

I am still tempted by a Fluke 25.

What's the purpose of the light bulb in the back?

Redneck charge controller.

[tggzzz]

What's the purpose of the light bulb in the back?

It is a short-circuit current limiter, a traditional technique based on the bulb's non-linear I-V curve.

In one picture it was fully illuminated because the NiCd battery was connected - and was a short circuit. Removing that battery/short means the lamp is only dimly illuminated in normal operation.

[Cubdriver]

Sweet score!  Thanks for the tear down photos.  Nice piece of kit you snagged there.

-Pat

[bitseeker]

What a cool find, tggzzz! The mainboard layout looks like one of the two variants in an 8100A. The front panel is very 8100A, also (no current functions). However, the model number would lead one to believe it's based on an 8120A.

Perhaps the 8125A was made after the 8120A, but since it doesn't have current measurement, it used the components from the 8100A.

[tonyalbus]

Found also one!!



Do you have the manual, schematics ? i did find the calibreation docs, if you are interested...

Tony

[tggzzz]

Do you have the manual, schematics ? i did find the calibreation docs, if you are interested...

I have two scanned documents: the 121 page "8120a digital multimeter" manual from August 1972, and the 12 page "TB 9-4935-365-35" military calibration manual.

[tonyalbus]

Do you have the manual, schematics ? i did find the calibreation docs, if you are interested...

I have two scanned documents: the 121 page "8120a digital multimeter" manual from August 1972, and the 12 page "TB 9-4935-365-35" military calibration manual.

Thats great i'll send you a DM

[tonyalbus]

Thanks !!  received