No.137 - Keithley 440 Digital Picoammeter Repair & Test

Donated by Dan from England, this Picoammeter is in need of some calibration on it's Picoamp range.
And after some re-calibration & testing it suddenly stopped working, so needed repair also!
Here's Dan's EEVBlog thread on the unit before it came to me:
www.eevblog.com/forum/repair/...
#electronicscreators
Visit my website for lots more - www.ianjohnston.com
Help me make more of these videos:
Pledge using Patreon - / iansjohnston
Donate using PayPal - www.paypal.me/IanSJohnston
Follow me on Twitter - / iansjohnston

[@IanScottJohnston]

Still to do: The 225, 486 and 440 need to be run for a while before further CAL/evaluation etc. The 486 needs a triax connector which I don't have yet. Possibly replace the potentiometers inside the 440 with a small round Pcb (sympathetic design) on which I can fit 10-turn pots. 440: Update: So my replacement darlington pair transistors arrived (new old stock apparently!) so I switched out that TIP126 I fitted, also the amplifier transistor Q113.......WOW! the tempco on those old transistors is pretty bad. -15v rail was -14.5v at power up and slowly crept up......heat from soldering! Seems stable at -15v now though. Oh yeah, and the main darlington transistor that was fitted to the 440 when I got it......was a BFY51, an NPN transistor....ouch!

[@danielbriggs2]

Excellent work Ian! I’m glad you were able to put my 440 to some use, repair and cal the picoamp range. +share the Nixie display for others to appreciate. Keep up the good work! Dan

[@TheEmbeddedHobbyist]

my 88 year old mum who was watching over my shoulder asked what it did, I pointed out it was a current meter! her reply to that was can it measure sultanas. 🙂

[@IanScottJohnston]

Asda don't stock Pico Sultanas........might try for Nano's next time we're in.

[@petarpetrov6388]

Love this acsent! Especialy cler pronounced "R". Good job too, very explanatory! Greetings from Bulgaria!

[@IanScottJohnston]

Thanks Petar.

[@PeterCut62]

Awesome vid as usual. Thank you Ian.

[@GadgetReviewVideos]

I ran across a Keithley 480 Picoammeter years ago a repaired it. It’s surprising accurate even compared to some high end modern equipment and beats out the mid range equipment. Also it look neat on the bench. I was able to find the manual that included parts list and schematic for it.

[@Runco990]

I REALLY love to see the engineering of old devices. Clever stuff! 👍

[@MiracleMAX]

Great job as usual Ian. I love repairing electronics and keeping them out of landfill as well. At the moment I am working on a 55 inch LED TV that shuts down after 20 minutes. Has got me scratching my brain a bit, but I'm sure I'll get there in the end. I like your diagnostic techniques and analysis of schematics as well. Keep up the good work! 👍😊

[@SeanBZA]

Those old ceramic case transistors were always a mystery, some batches would fail fast because of contamination, and some would not. But well done to getting it right. The power supply having slightly different resistor values depends on the actual long tailed pair in use, if you had a batch with low gain you needed the changed value to sort of compensate for the base current, but if they had high gain (later production where the high gain bin was generally the most common bin, so the lower gain bands and the unbanded could be anything including high gain) then using identical resistors ensured a very close match in tracking between the rails. that would have been a part selected on production, where the transistors were gain checked, and then a decision made if the run needed to have the changed resistor to get the rails within tolerance, or they were selected in production and made up into packs accordingly, for the assembly line to use. Auto insertion machines and reels of assorted axial components were common, and there were a few who used a single reel, and your component was different as you went down, but an auto inserter would always place them in the same order.

[@IanScottJohnston]

Update: So my replacement darlington pair transistors arrived (new old stock apparently!) so I switched out that TIP126 I fitted, also the amplifier transistor Q113.......WOW! the tempco on those old transistors is pretty bad. -15v rail was -14.5v at power up and slowly crept up......heat from soldering! Seems stable at -15v now though.

[@IanScottJohnston]

Oh yeah, and the main darlington transistor that was fitted to the 440 when I got it......was a BFY51, an NPN transistor....ouch!

[@SeanBZA]

@@IanScottJohnston Well it sort of worked there....... Probably as a zener series regulator due to emitter base break down.....

[@davidv1289]

Really interesting repair video, Ian. Thank you! Finding calibration standards for these instruments is a challenge for hobbyists. I am currently repairing a Keithly 619A dual channel electrometer and Keithly provides a calibration procedure for the 20 nA range that uses a 1.9 volt source (PDVS2Mini - of course) and a 100 Mohm resistor to provide 19 nA. Perhaps you could use a similar method on the picoamp range of your 440. My biggest challenge has been coming up with a 19 Gohm shielded and guarded resistor to calibrate the 20 Gohm range. Regards, David

[@IanScottJohnston]

Cool idea using the pdvs2mini. My 486 could probably use with recal also.

[@joelkist6493]

Excellent video; thank you sir.

[@erikdenhouter]

That Q112 was build in a a safety device, a current limiter only if things go wrong. That must have saved a lot of hardware with this fault. Or it could be that there already was no more regulation from te Darlington, but... the negative voltage was more than it should be. Nice straight forward repair 😎

[@radio-ged4626]

Love those Nixie valves, it makes me wish I had a use for measuring Picoamps. Not too much call for it in Valve radio repair.

[@gjvdspam]

Nice one pico amps, counting the coulombs passing by are you haha. Nice device, I enjoyed it a lot

[@swainsa]

I have an old HP boat anchor picoammeter and the calibration procedure uses high value reference resistors and a basic power supply. Getting hold of some multi gigaohm 1% resistors might be another option for adjusting such equipment, without needing to push your 220 to its limits... or maybe try to find a Keithley 261, like I think Dave Jones has (lucky so and so).

[@IanScottJohnston]

Yes, I saw Dave has a 261, and I did look for one but the prices are horrendous.

[@Miller7747]

Not surprised by so many damaged transistors, that is what happens in DC coupled stages. One transistor dies, but it takes out several others with it.

[@TheEmbeddedHobbyist]

Well they did get the title of "fastest 3 legged fuses". :-)

[@TheDefpom]

The 225 has a "off" position for the fine tune, which should be used to set it when calibrating to ensure a consistent setting, turn it fully left and you will feel it click in.

[@IanScottJohnston]

Not if you follow the calibration procedure in the manual: "Reference" calibration "Set FINE to 10" and adjust R182 (which is the SPAN pot). What I found was that this gives a +/- ability around the calibrated value instead of locking you out trimming down the way afterwards. Seems wierd I know, but I just followed the procedure just to see where it led to.

[@TheDefpom]

@@IanScottJohnston ok, mine came set for the fine being off instead.

[@alpcns]

Beautiful equipment and a joy to watch. How do you manage not to drool constantly when seated happily between all that marvelous gear?

[@IanScottJohnston]

I must admit, I spend quite a lot of time out here....:-)

[@DrFrank-xj9bc]

Very interesting pA stuff repair.. thank you - very satisfying. Didn't you check the zero baseline, i.e. possible bias or leakage currents of the different instruments? That might have been another source of error .. and maybe have been necessary before the F.S. calibration? Interesting to see one of the later successors of the 440 pico ammeter. Keithley as well has a systematic numbering system, like HP. 2xx for current source, 4xx for pico meters.

[@IanScottJohnston]

The 225, 486 and 440 need to be run for a while before further CAL etc. Not to mention the 486 needs a triax connector which I didn't have, I was only able to 'cobble' something together temporarily for the testing. Also, I think I'd like to replace the potentiometers inside the 440 with a small round Pcb (sympathetic design) on which I can fit a 10-turn pot. I have also got to replace the TIP126 when my new, old stock, transistors arrive. So, more offline fun to be done.

[@geirendre]

1:00 Is that really 4 nixies? The board inside had just 3 nixies on it. Is the forth digit in front of the 3 nixies just a neon bulb that serves as a 1 indicator? Guess this can be refered to as a 1999 count meter 🙂 Nice repair anyway, and I just love your Scottish accent.

[@IanScottJohnston]

Later in the video I think I might have mentioned it was possibly a neon. I never really paid too much attention to it. Thanks on the accent....:-)

[@donepearce]

Ian. Nice work - what's the burden voltage like on the picoamp range?

[@IanScottJohnston]

Manual says 1mV for full scale display on all ranges.

[@victorman2227]

Its a feedback ammeter so rather small probably.

[@worroSfOretsevraH]

Hi. What type/model are those thin brown coax cables? Are they 50 Ohms? Thank you.

[@IanScottJohnston]

50ohm RG316 cable c/w BNC bought via E b a y.

[@larrybud]

Apologies if you addressed this (still watching the video) but what is the real-world use case for a device like this?

[@IanScottJohnston]

Electrometer calibration, photodetector testing & characterizing, environmental/biomedical sensor testing, measuring leakage currents in high resistance materials, semiconductor/IC testing & design........so I guess more towards science than traditional electronics.

[@keithmarlow143]

after 50 years most things start bulging...

[@IanScottJohnston]

Ho ho, not wrong there!

[@stoptheirlies]

Hi Ian, Awful lot of work to calibrate something you will probably never use! I can't think of a single case when I would need to accurately supply or measure nano-amps. Bob