10000hr Battery Life Bench Multimeter

[EEVblog]

This really belongs in the test equipment group:

[floobydust]

Feasible but I find it's a MYTH that batteries last decades. Their lifetime spec is theoretical, based on self-discharge not their pissing, leaking package. Hello Duracell "10-year guarantee (5-year guarantee for 9V)".

I deployed Ultralife 9V LiMnO₂ (1st generation) 1,200mAh in a fleet of Fluke 87's for low temperature below -18°C use. They lasted from 5-12 years depending on multimeter use. After about 12 years, they die and balloon out quite a bit- but never leaked. The latest are CAD $40 each and appear to have changed from plastic to SS case.

Tadiran LiSOCL₂ claiming proven (up to) 40 year life ... they leaked after about 8 years. Laser-welded SS case corrodes and they spew electrolyte all over a PC board and eats traces. Their lifetime claims are not in datasheets, only in marketing wank. Not a believer.

I get 2,000hrs on a 9v one of my old DMM's and it's a treat.

[bdunham7]

https://multimeterwarehouse.com/m9803r.html

[retiredcaps]

Dave's meter is an APPA 205.

Regarding the JRC NJU9210FF, it's been used in Appa 105 rebranded Tek DMM 252.

https://www.eevblog.com/forum/testgear/iso-tech-(appa)-idm105-dmm/

https://www.eevblog.com/forum/testgear/tektronix-dmm-252/

https://www.eevblog.com/forum/repair/tek-dmm252-multimeter-wrong-voltage-and-current-measurement/

[SilverSolder]

Feasible but I find it's a MYTH that batteries last decades. Their lifetime spec is theoretical, based on self-discharge not their pissing, leaking package. Hello Duracell "10-year guarantee (5-year guarantee for 9V)".

I deployed Ultralife 9V LiMnO₂ (1st generation) 1,200mAh in a fleet of Fluke 87's for low temperature below -18°C use. They lasted from 5-12 years depending on multimeter use. After about 12 years, they die and balloon out quite a bit- but never leaked. The latest are CAD $40 each and appear to have changed from plastic to SS case.

Tadiran LiSOCL₂ claiming proven (up to) 40 year life ... they leaked after about 8 years. Laser-welded SS case corrodes and they spew electrolyte all over a PC board and eats traces. Their lifetime claims are not in datasheets, only in marketing wank. Not a believer.

I get 2,000hrs on a 9v one of my old DMM's and it's a treat.

Lithium batteries do have good durability -  I use them everywhere I can now...  (i.e. when rechargeables are not appropriate).   I started using them about 8 years ago, and I haven't yet had to change one that died before it was used up.  No leaks either.  I'm a Lithium fanboy now.

[MadTux]

Well, battery only needs to survive like 50 charge charge cycles. Then you're dead anyway 

[floobydust]

Lithium batteries do have good durability -  I use them everywhere I can now...  (i.e. when rechargeables are not appropriate).   I started using them about 8 years ago, and I haven't yet had to change one that died before it was used up.  No leaks either.  I'm a Lithium fanboy now.

What brand/size/chemistry?
For primary lithium batteries, I'm having good success with Energizer Ultimate Lithium, using them in cold weather rated -40°C to 60°C.
L91 AA 3,500mAh Lithium/Iron Disulfide (Li/FeS₂) BUT high voltage ~1.8-1.7V typ. about $3 ea.
L522 9V ?mAh Lithium/Iron Disulfide (Li/FeS₂) voltage ~8.5V typ.
Marketing wank :"Energizer Ultimate Lithium™ batteries are 100% leak proof* *Based on standard use", "Holds power in storage for up to 20 years (AA/AAA)"

Realistically, Eneloops are great. Charge them once or twice a year, about $5 ea. AA

Why isn't there a solar panel and renewable approach instead of this 10,000hr battery being proposed? 

[MBY]

Funny, I'm currently in a project myself to do exactly what Dave is thinking! My choice of "decent but cheap" bench DMM is the [Noname/Victor] DM8145, a 80,000 count bench DMM with a LOT of empty space. https://www.banggood.com/DM8145-4-7-or-8-Bench-top-Multimeter-1000V-20A-80000-Counts-Digital-Multimeter-tester-Auto-Range-Multimetro-Digital-Voltmeter-Ohm-p-1687590.html?cur_warehouse=CZ&rmmds=search

I think this meter is ideal, as it has a real power switch with an arm traveling to the back where the simple iron core transformer PSU is.

I've ordered some micro switches that has yet to arrive. My plan is to mount it in such way that it's actuated by the power switch arm and my goal is to have a seamless switchover between battery and mains use. If connected to the mains, the mains charges the battery, and if power is lost/removed, the battery powers the meter. And the option to use primary cells. Probably with a switch or a jumper. High voltage tolerance is the aim, I think, so one can wack 18650, alkaline AAs, PP3 or whatever in. Some old bench meters often had a battery option, usually like 2-6 Pb-acid D-cells (the Cyclone thing). I think such option is available to e.g. Keithley 177 and 179 DMMs and such.

I have an old "Beckmann" multimeter powered by a couple of D-cells, and I think it has the same dry cell batteries as when I got it, in the early 90s. I's very similar to the bench fluke you showed. No mains, just batteries.

[SilverSolder]

Lithium batteries do have good durability -  I use them everywhere I can now...  (i.e. when rechargeables are not appropriate).   I started using them about 8 years ago, and I haven't yet had to change one that died before it was used up.  No leaks either.  I'm a Lithium fanboy now.

What brand/size/chemistry?
For primary lithium batteries, I'm having good success with Energizer Ultimate Lithium, using them in cold weather rated -40°C to 60°C.
L91 AA 3,500mAh Lithium/Iron Disulfide (Li/FeS₂) BUT high voltage ~1.8-1.7V typ. about $3 ea.
L522 9V ?mAh Lithium/Iron Disulfide (Li/FeS₂) voltage ~8.5V typ.
Marketing wank :"Energizer Ultimate Lithium™ batteries are 100% leak proof* *Based on standard use", "Holds power in storage for up to 20 years (AA/AAA)"

Realistically, Eneloops are great. Charge them once or twice a year, about $5 ea. AA

Why isn't there a solar panel and renewable approach instead of this 10,000hr battery being proposed? 

Energizer Ultimate, Ultralife 9v batteries, and Eneloop 2000mAh AAs are the staple diet.

[Raj]

Would you buy a bench top meter with hand held meter specs?
Naa, I'd just find a meter who's display connect with ribbon cable. Extend the cable length and 3d print a new case for the thing.

[drussell]

My good old American Reliance model 37 also uses that same JRC NJU9210 controller IC with an external true RMS converter chip.  They're only 4000 count, but was actually a pretty slick setup back in the early 90s when I got mine.   

Mine runs on two AAA cells and it lasts for YEARS on a set of alkaline AAAs.  Definitely nice, low power consumption!

It has certainly held up and still works well today within its performance limitations. 

I actually just fully disassembled the ol' Amrel the other day and gave everything a thorough proper cleaning of the button and switch contacts, the LCD zebra strip, re-lube the main dial assembly, etc. and it is back to running basically like new. 

edit: It took quite a bit of scrubbing with SuperClean (basically sodium hydroxide based) to get the yellow holster clean, I wish I had a "before" photo, as it has been many, many years since I did a real proper cleaning of all the grease and grime from when it gets used in the garage or shop or some dirty worksite or whatever.  I was really surprised how nice and yellow and clean it came, or I suppose I was surprised how dirty the whole thing actually was?   

Oops.  I guess it was a tad overdue for proper routine maintenance? 

[drussell]

Dave's meter is an APPA 205.

Regarding the JRC NJU9210FF, it's been used in Appa 105 rebranded Tek DMM 252.
...

Aha!!!   

I'll bet my Amrel was made in Taiwan by Appa!  I always wondered who actually made those since it was a relatively short-lived product line for Amrel, who's focus was primarily high-end power supplies and electronics loads and military contract stuff.

The Amrel 37 has smooth sides, not the ribs like the Appa 100 series and the dial switch positions are in a different order, as are the buttons (like the blue select button is below the reset/on button) but it is exactly the same size as a Fluke 87 and fits an 87 holster also, etc.  My Amrel even came with a yellow holster as this was before Fluke started slapping and smacking other companies around for using yellow holsters.   



Somewhere at home I have the tilting bale (one of the little pins broke off, I need to repair that at some point) and the plastic protector that goes over the LCD that says "AMREL 37 True RMS Digital Multimeter" that I need to glue back on, but considering the abuse it's been put through in the past 30 years or so, it's holding very up well.   

I'm glad I bought it a good set of Mueller probes way back in the day, they've lasted for probably almost 25 years.  The original ones that came with the meter lasted for a few years, then I think I bought a cheap set once, then splurged on the Muellers that even came with a bunch of screw-on accessories.  That was well worth the investment.

[goaty]

This one easily lasts longer than 10000hr on one battery.

[retiredcaps]

I'll bet my Amrel was made in Taiwan by Appa!

Look on the pcb backside and see if you can spot 105 on it like in this picture fromthe ISO-TECH (APPA) IDM105 DMM thread.

[pascal_sweden]

http://www.vicimeter.com/en/product/VC8145

[drussell]

I'll bet my Amrel was made in Taiwan by Appa!

Look on the pcb backside and see if you can spot 105 on it like in this picture fromthe ISO-TECH (APPA) IDM105 DMM thread.

No, it is apparently a much cheaper design than the 105 and only says "MIC37-1G" on the board.  I don't recall seeing "APPA" anywhere on it but I'll check closer the next time I have it totally disassembled.  This one is definitely a cost-reduced variant as it doesn't have proper HRC fuses, they're actually just 5x20mm size ceramic.  I wouldn't want to go probulating high energy circuits with this one, I have other meters for that, plus you usually want a clampmeter anyway in that kind of situation.  The construction just looks very similar to the one that Dave showed in the video, though I suppose it could have been made by some other similar factory still...

I know the datecode on the AD737JR is 9246, though...  I happen to have written that down the other day when I went looking for datasheets after thinking about multimeters in another thread.   

When I saw Dave taking apart that GW Instek branded one I was thinking to myself, wow, that construction style looks a lot like the Amrel.

[wizard69]

Would you buy a bench top meter with hand held meter specs?
Naa, I'd just find a meter who's display connect with ribbon cable. Extend the cable length and 3d print a new case for the thing.

Why not?   This especially if it is easily battery portable.   It might not be the main meter on a bench but there are plenty of cases where you don't need high performance and endless resolution.   

What we are really talking about here is a bench meter that is easier to use than a handheld and I'm pretty confident that we have all used a handheld meter on the test bench at one time or another.

[wizard69]

This is one of Daves better videos as it inspires and with this thread highlights meters I never knew existed.

Actually I'm hoping it inspires one or more manufactures to take some of the ideas expressed by Dave and others in this thread to heart.   For one long battery life in a desktop multimeter should be easy with today's technologies.   Even if it isn't 10'000 hours, a meter that could remain on for years at a time would be fantastic.   The idea expressed above for solar power is pretty good too if the goal is to charge or supplement the contained battery.   It is actually surprising to me that manufactures haven't gone this route with handhelds or in cases like this where the device is a "portable" bench top design.

[Fungus]

I recently scored this on eBay France:

Looks like the guy had a bunch of them and I got the last one.



It's in perfect condition. I had a look inside and it's just a Fluke 27FM in bench format.

The footprint is quite big so I've been thinking about chopping the back off. It seems a shame to do it, but... it would make a really small "bench" meter.

The main issue with handhelds (or at least, the ones I own) isn't lack of digits, it's that they have auto-power off and start to beep at me after a few minutes sitting idle. A bench meter will sit there all day and be happy.

The Fluke 37 has a 9V DC power socket so it would be easy to run off an adapter but it also goes for 1000+ hours on a 9V battery. There's a big storage space in the back for leads/tools so it would be an easy mod to put in a huge battery, eg. 6xAA would be about 5,000 hours use, 6xC cells could be the mythical 10,000.

The only downside I see is using a dial to select ranges. Most bench meters use push buttons.

[EEVblog]

I recently scored this on eBay France:
Looks like the guy had a bunch of them and I got the last one.

Nice! 

[EEVblog]

The idea expressed above for solar power is pretty good too if the goal is to charge or supplement the contained battery.   It is actually surprising to me that manufactures haven't gone this route with handhelds or in cases like this where the device is a "portable" bench top design.

[AVGresponding]

HP 3468/78 have LCD displays, would be interesting to know how much power they use...

The only LCD bench meters I have are my Micronta 22-195 which uses 4x C cells anyway, and which runs pretty much forever on them, a Black Star 3225 which ditto, a Black Star 3210MP, which is basically a newer 3225 and mains powered, and should therefore be easy to mod, and a Keithley 197, which I believe may be available with a battery option (mine doesn't have it).


EDIT: Forgot the Solartron 7150 I have, but there's absolutely no space in the case for batteries on that, it's properly stuffed, though I suppose if you ripped out the trafo you might have enough room.

[Kleinstein]

The old 3468 / 3478   (5.5 digit)  use quite a lot of part. They are not at all build to be powered from battery.  They include and old 5 V powered CPU and thus take a lot of power - expect more like 10 W.  3478 specs are < 25 W.

Battery power with a high resolution meter would be more like a thing for rechargable battery and than maybe some 1-3 W and 10 - 30 h  for a singel charge, unless one uses a really large battery. It would still makte sense sometimes because of isolation.

[AVGresponding]

The old 3468 / 3478   (5.5 digit)  use quite a lot of part. They are not at all build to be powered from battery.  They include and old 5 V powered CPU and thus take a lot of power - expect more like 10 W.  3478 specs are < 25 W.

Battery power with a high resolution meter would be more like a thing for rechargable battery and than maybe some 1-3 W and 10 - 30 h  for a singel charge, unless one uses a really large battery. It would still makte sense sometimes because of isolation.

The inputs on these instruments are fully isolated anyway? Not sure what you're getting at there.

My Keithley 197 appears to use around 3W, but that's measured using an inline mains power meter. For a better reading I'll have to open it up and get into the DC side, I'd expect to see maybe 2W at that point, but of course that's still a bit much for the purposes of this thread.

The Solartron says 23VA on the back, so clearly way too power hungry...

[thm_w]

HP 3468/78 have LCD displays, would be interesting to know how much power they use...

The only LCD bench meters I have are my Micronta 22-195 which uses 4x C cells anyway, and which runs pretty much forever on them, a Black Star 3225 which ditto, a Black Star 3210MP, which is basically a newer 3225 and mains powered, and should therefore be easy to mod, and a Keithley 197, which I believe may be available with a battery option (mine doesn't have it).

3478 uses 7.3W