10000hr Battery Life Bench Multimeter

[wizard69]

Well that was fast.   It is one of  the great things about this forum, I find out about things that I never knew existed!   Now the question is who makes that meter?   So far I've found:

• http://www.all-sun.com
• https://overseas.sanwa-meter.co.jp/products/digital_multimeters/ps8a.html

These meters are surprisingly hard to find as a google search for "solar powered meters" turns up a bunch of crap about meters for solar power installers.   Honestly Dave I'm surprised you are not offering the Sanwa on your site, it looks like a very green solution.   Not a bench top but rather an ultra compact but the PS8A is very interesting to say the least.

The funny thing here is that in reading the various manuals they have objections to cleaning with Alcohol.   I wouldn't even care normally but lately everything is being cleaned with IPA.

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.

[MBY]

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

Yep, this is the DMM I mentioned and I think it's a good candidate. Turns out to be easily modded (I've done a mod so the annoying buzzer only beeps in continuity mode, described here https://www.eevblog.com/forum/testgear/vc8145-thread-it_s-here-finally/msg3448652/#msg3448652 )

I've done some preliminary work on it. It have four relays, so it consumes a couple of mA more than Daves GW Instek.

PSU---------------------------------------
Measured with: BK 2709B DMM.
9 Vdc single rail from a 7809. About 13.5V on the unregulated side.
It has a "low bat" segment on the LCD and the case has screw towers and such
that are unpopulated. A battery option is certainly possible, but the "low bat"
segment seems NOT to be connected/used as it never comes on.
Winding down the input voltage dims the backlight and the LCD starts fading pretty soon.
The measurement seems stable and reliable down to 7.55V.
I connected a PentaRef (that haven't been calibrated for 5-6 years) set to 3.9001V and the display (Vdc range) showed a stable 3.9009V down to 7.55V, and then started drifting upwards (as expected).
-------------------------------------------

Current consumption--------------------
Measured with: uCurrent on mA range, a BK 2709B DMM.
About 11-12 mA in the following modes:
mAdc, Vdc, Adc

About 44 mA in the following modes:
Vac, mVdc, Ohm (add 0.9mA on short), Cap, Temp, Out

74 mA in:
Diode/Cont

Back-light: add 9,6 mA
-------------------------------------------

My goal differs a bit from Daves. He wants a primary cell-driven low power DMM that lasts for decades with intermittent use. Li-SOCl2 as Dave suggested is an excellent choice for this and I understand the reasoning. But they are pretty expensive and my own user case is a bit different. So I want a rechargeable Li Ion system anyway (or would have selected alkaline or Li-FeS2 primary cells instead, as the DM8145 takes more power).

I've tested to power the DM8145 from a boost converter (this may be critical. There is a reason mains-driven measure equipment always have good old iron transformers and linear regulation) and nothing bad seems to be happening, but I haven't checked all the ranges and scenarios. But I know that many are gonna puke on my "solution" with buck and boost DCDCs, and rightly so. This is just a, eh, shall we call it "design study"?

The upper part (circled in red) of the schematics is the real PSU of the DM8145. The lower part is my quick and dirty add on for a 3.7V li ion battery (bank). It's just a sketch, I haven't done all the calculations, tests or other "brain task" necessary. The only practical tests I've done is measure the currents and a superficial test to power the DMM from a boost converter. The relay inhibit the boost converter when mains power is available. It may not be needed, but without it the battery sees a light duty as long as power is on (driving a useless DCDC) preventing some cycling. Without it the thing works like an online UPS with seamless switchover. I decided to stuff in a large cap (C5) to make the switchover seamless with the relay. 

The case have several screw towers not in use. I've discovered that one was suitable to mount a microswitch for the battery mod. Attached are some pictures. Yes, the "block" used to push the micro switch is a large SMD cap I glued on the power putton steel arm.

Edit: It's a bit silly that they have chosen 5V relays (Omron, no less!) driven from a 5V rail, when they have 9V available and of course use linear regulation. 9V relays would draw less power. A mod is certainly possible with either reed relays or driven from 9V, to minimize the power consumption.

[MBY]

Well that was fast.   It is one of  the great things about this forum, I find out about things that I never knew existed!   Now the question is who makes that meter?   So far I've found:

• http://www.all-sun.com
• https://overseas.sanwa-meter.co.jp/products/digital_multimeters/ps8a.html

These meters are surprisingly hard to find as a google search for "solar powered meters" turns up a bunch of crap about meters for solar power installers.   Honestly Dave I'm surprised you are not offering the Sanwa on your site, it looks like a very green solution.   Not a bench top but rather an ultra compact but the PS8A is very interesting to say the least.

The funny thing here is that in reading the various manuals they have objections to cleaning with Alcohol.   I wouldn't even care normally but lately everything is being cleaned with IPA.

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.

The meter Dave posted is an Uni-T UT90x (UT90C perhaps). Don't know more about it, haven't seen it IRL, haven't used it. But Uni-T devices usually is on the range from crappy to almost decent and sometimes really decent. (They are very "hit and miss". Most of my most useless POS equipment has been Uni-T, but also some of my all time favorites, such as my old trusty Uni-T UT-60C. My experience is that I can't take my opinions and knowledge about a particular model and extrapolate to another. It's all about what I call "soft parameters" never stated in a datasheet. The feeling of using, the feel of buttons, etc.)

[retiredcaps]

Martin reviewed the solar Uni-T and killed it.

[MBY]

Lol! He didn't even try to kill it! I expected the usual studious HV-testing and high energy events to kill a el cheapo meter, but it turns out it could not cope with mains on Freq range (or ohms range), a pretty basic test. This is why I said that Uni-Ts are hit and miss. They are basically all over the place with input protection, ease of use, functions and build quality. You never know from one model how another gonna preform. Edit: Sorry, we are a bit off topic now. Back to battery powered bench meters. Uni-T have some really old "handheld meter in a bench box"-meters that may be very suitable for battery power. If they don't already have it.

[Dwaine]

This is the meter for Dave to mode with a 10000hr Battery .  It already runs off batteries...

https://www.ebay.com/itm/ANENG-AN888S-Digital-19999-Counts-High-Precision-True-RMS-Multimeter-Ohm-N-D/333768482621?_trkparms=aid%3D1110006%26algo%3DHOMESPLICE.SIM%26ao%3D1%26asc%3D20201210111314%26meid%3D24aa7e46a44741a2945f999f7f26c78c%26pid%3D101195%26rk%3D2%26rkt%3D12%26mehot%3Dco%26sd%3D220336911971%26itm%3D333768482621%26pmt%3D1%26noa%3D0%26pg%3D2047675%26algv%3DSimplAMLv5PairwiseWebWithDarwoV3BBEV2b%26brand%3DUnbranded&_trksid=p2047675.c101195.m1851

[EEVblog]

This is the meter for Dave to mode with a 10000hr Battery .  It already runs off batteries...

https://www.ebay.com/itm/ANENG-AN888S-Digital-19999-Counts-High-Precision-True-RMS-Multimeter-Ohm-N-D/333768482621?_trkparms=aid%3D1110006%26algo%3DHOMESPLICE.SIM%26ao%3D1%26asc%3D20201210111314%26meid%3D24aa7e46a44741a2945f999f7f26c78c%26pid%3D101195%26rk%3D2%26rkt%3D12%26mehot%3Dco%26sd%3D220336911971%26itm%3D333768482621%26pmt%3D1%26noa%3D0%26pg%3D2047675%26algv%3DSimplAMLv5PairwiseWebWithDarwoV3BBEV2b%26brand%3DUnbranded&_trksid=p2047675.c101195.m1851

That runs off two 18650's and is rechargable.
May be possibe to nuke the bluetooth speaker crap and just user the meter though.

[Fungus]

This is the meter for Dave to mode with a 10000hr Battery .  It already runs off batteries...

https://www.ebay.com/itm/ANENG-AN888S-Digital-19999-Counts-High-Precision-True-RMS-Multimeter-Ohm-N-D/333768482621?_trkparms=aid%3D1110006%26algo%3DHOMESPLICE.SIM%26ao%3D1%26asc%3D20201210111314%26meid%3D24aa7e46a44741a2945f999f7f26c78c%26pid%3D101195%26rk%3D2%26rkt%3D12%26mehot%3Dco%26sd%3D220336911971%26itm%3D333768482621%26pmt%3D1%26noa%3D0%26pg%3D2047675%26algv%3DSimplAMLv5PairwiseWebWithDarwoV3BBEV2b%26brand%3DUnbranded&_trksid=p2047675.c101195.m1851

That runs off two 18650's and is rechargable.
May be possibe to nuke the bluetooth speaker crap and just user the meter though.

Nope. That one uses the trendy inverse LCD display so it will suck a lot of power when it's on.

(I'm sure you can put in enough batteries to make it last 10000 hours but it's not an optimal choice)

[Kleinstein]

Inverse LCD or not does not make any difference on the energy consumption: it is just a quitestion on the orientation of the polarizer. With some displays the polarizer is a separate filter sheet and can even be fliped to change between normal and negative display. For convenience the direction of the polarizers is often at some 45 deg. So that flipping the polarizer turns the orientaion by 90 deg, so one can use the same parts in both ways.

@MBY:
For the battery mod with VC8145, one could consider to also change the L7809 regulator. These take quite some idle current (some 5 mA) and with a LDO one would use a lower votlage for the boost converter (e.g. 9.5 V instead of 12 V). This would reduce the power needs quite a bit.

[Fungus]

The only thing I could find on Aliexpress is this:

https://www.aliexpress.com/item/33021584962.html

Not cheap but the paper specs are quite good.


Edit: There's some cheaper ones on there but they're all manual ranging. I assume we don't want that.

[Fungus]

Inverse LCD or not does not make any difference on the energy consumption

It does when the backlight is always on.

I guess you could disconnect the the backlight but then it's not very readable.

For convenience the direction of the polarizers is often at some 45 deg. So that flipping the polarizer turns the orientaion by 90 deg, so one can use the same parts in both ways.

That would need to be tested.

[MBY]

For the battery mod with VC8145, one could consider to also change the L7809 regulator. These take quite some idle current (some 5 mA) and with a LDO one would use a lower votlage for the boost converter (e.g. 9.5 V instead of 12 V). This would reduce the power needs quite a bit.

Yep. Exactly my thinking. As any mods needs to solder things on the PSU PCB anyways, it seems reasonable to replace the 7809 with an LDO! If the instrument behaves good, you even can have the boost DCDC feed the 9V rail directly, but I thought it would be a extra filtering step to have a linear regulator after the boost.

The only thing I could find on Aliexpress is this:

https://www.aliexpress.com/item/33021584962.html

Not cheap but the paper specs are quite good.


Edit: There's some cheaper ones on there but they're all manual ranging. I assume we don't want that.

This is the one I'm talking about and it have at least one long thread on this forum: https://www.eevblog.com/forum/testgear/vc8145-thread-it_s-here-finally/

It's certainly no HP or Fluke, but it's good enough. I've used it for a couple of months, but the model seems almost a decade old (and that's a good thing, not a come-and-go model from this weeks Shenzhen market but something under active production and with a lot of knowledge about it online). You can argue that it's too expensive for a Chinese no-brand (or "any brand") bit of gear, but I think it's at least reasonable. I would say that the closest thing, if you want a good meter, and the one they tried to copy, is Fluke 45 (100,000 counts), but they are different beasts really (and if you can get a Fluke for that price, the DM/VC8145 doesn't stand a chance of course).

But, all in all, with high availability, good specs in general, large mod potential and a lot of empty space, it's a good contender to be a "standard meter for modding". The display is good, the backlight stays on if turned on, it has a good set of features (and some total useless gimmicks such a timer function nobody seem to understand and very limited waveform output) such as RS232 with a speced protocol, delay/peek hold that actually works and seems to be built with decent components, etc. I like it, and I know good from bad (having Flukes, TTIs, Keithleys and HPs, along with a bunch of really shitty gear). So, if it were a vote on "best half decent bench DMM for the masses to do an battery mod on" I gladly vote for the 8145. One thing I hate, and I know a lot of people hate, is it's totally annoying buzzer, but I've fixed mine without compromising away the buzzer when it *should* operate (I've linked to my mod in an earlier post).

But, the downside is, 8145 is probably not what Dave intended as it would preform badly on li thionyl chloride batteries. It simply draw too much. We can swap out relays all day long, but I don't think it's gonna beat Daves DMM that does 1.7 mA at 3 volts...

Edit: Btw, that Aneng 888 must be some kind of sick joke. I wonder how the design meetings were for that one...

[Dwaine]

This is the meter for Dave to mode with a 10000hr Battery .  It already runs off batteries...

https://www.ebay.com/itm/ANENG-AN888S-Digital-19999-Counts-High-Precision-True-RMS-Multimeter-Ohm-N-D/333768482621?_trkparms=aid%3D1110006%26algo%3DHOMESPLICE.SIM%26ao%3D1%26asc%3D20201210111314%26meid%3D24aa7e46a44741a2945f999f7f26c78c%26pid%3D101195%26rk%3D2%26rkt%3D12%26mehot%3Dco%26sd%3D220336911971%26itm%3D333768482621%26pmt%3D1%26noa%3D0%26pg%3D2047675%26algv%3DSimplAMLv5PairwiseWebWithDarwoV3BBEV2b%26brand%3DUnbranded&_trksid=p2047675.c101195.m1851

That runs off two 18650's and is rechargable.
May be possibe to nuke the bluetooth speaker crap and just user the meter though.

Nope. That one uses the trendy inverse LCD display so it will suck a lot of power when it's on.

(I'm sure you can put in enough batteries to make it last 10000 hours but it's not an optimal choice)

But Dave can listen to music when he's in the lab taking measurements.

 

[MBY]

I remembered that I have four D-cells Pb-acid batteries that don't serve any function. I've successfully desulfated three of them, but the fourth is beyond repair. These kind of batteries were once common in backup power for instruments. I know that my Keithley 177 had an option for this (four cells I think) and I've once owned a pretty small CRT-based portable oscilloscope that also used them (two I think). I have a lot of 18650 li ions, but they are kind of useful for other things. So why don't make a quick 'n dirty battery options for the DM8145 using three of those D-cells? That's pretty straight forward and Pb batteries have no problem working in an online setting.

Going to build this thing now. I return with some pictures when finished.

Maybe I should start another thread? This differs a lot from Dave's idea with a 10,000h battery and is more like a 10h UPS...

[Fungus]

My Fluke 27 is still running the same crappy 9V battery that came installed when I bought it on eBay about 6 years ago.

(maybe I should check it for leaks...?)

[robert.rozee]

what about a fluke 77 or similar re-housed in a bench case? with a '77, the selector switch could be desoldered and fitted into a 3-d printed housing, and the LCD could be remoted on a daughterboard wired back to the main meter PCB. the main meter PCB could then sit flat inside the bottom of the case under a battery shelf, with 4 wires going up to new front-panel terminals.

the end result would look similar to a fluke 37, although possibly a little smaller. one advantage of using a 77 as the basis is that they seem to turn up here and there with broken cases and LCD mountings, but electrically intact (ignoring the 1k fusible resistor blown).


cheers,
rob   :-)

[Fungus]

what about a fluke 77 or similar re-housed in a bench case? with a '77, the selector switch could be desoldered and fitted into a 3-d printed housing

I'd definitely prefer push buttons over a rotary switch in a "bench" meter.

[robert.rozee]

push button operation could be achieved with bipolar relays - the schematic of the fluke 77 is readily available.

serial output could even be added by snooping on the 4-bit bus that runs between the ADC and display driver.

the question is: would a 3200 count display be sufficient for such a project?


cheers,
rob   :-)

[TheDefpom]

As I posted in the comments of the video, maybe look into the Fluke 45, it has a space for a internal battery already ! AND I am pretty sure I saw you had one already Dave.

[Wytnucls]

Other contenders are the Fluke 8050A and Keithley 176.
Both with 20,000 count and LCD displays.

The Fluke lasts for 10 hrs with its NiCad battery pack option (6W)
The Keithley lasts for 1,000 hrs with 6 internal C cells.

[bdunham7]

I have a few Fluke 8050A and I think they draw way too much current for a project like this.  My only battery-capable one was ruined by leakage and you can't just put batteries in the ones that didn't come with them--no charging circuit.  But I think they draw 50-100mA.

What is the lowest current draw anyone has measured on their DMM?  My Harbor Freight D830-based freebie draws about 1 mA on most ranges. Resistance draws 1.2mA, 200R and diode draw 3.3mA, all with the test leads open circuit.  It uses a 9V battery and the battery low indicator comes on at 7.5V, but the meter works down to 6.0V or even a bit lower.  Has anyone measured a meter to draw less? 

[Fungus]

What is the lowest current draw anyone has measured on their DMM?  My Harbor Freight D830-based freebie draws about 1 mA on most ranges. Resistance draws 1.2mA, 200R and diode draw 3.3mA, all with the test leads open circuit.  It uses a 9V battery and the battery low indicator comes on at 7.5V, but the meter works down to 6.0V or even a bit lower.  Has anyone measured a meter to draw less?

With nothing connected my Fluke 37 draws about 720 uA in all the AC ranges and 325 uA in the non-AC ranges.

[AVGresponding]

My Micronta 22-195 uses just under 900uA for DC ranges, and just over 1100uA for AC.
To complicate matters, the 4 C cell battery is centre tapped, so a dual rail 3V battery.
The h_FE function that Dave so hates on DMMs uses 2200uA unloaded!
The ohms ranges use up to an additional 1250uA, also continuity.
Diode test uses 1450uA.

I'm going to pull my Keithley 197 apart this weekend if I remember, and see if I can make any useful current measurements.

[bdunham7]

With nothing connected my Fluke 37 draws about 720 uA in all the AC ranges and 325 uA in the non-AC ranges.

That sounds like the winner.  If you can pack 6 D-cells in there, you've achieved the 10K hour goal.  And they wouldn't need to be alkaline.

[Fungus]

With nothing connected my Fluke 37 draws about 720 uA in all the AC ranges and 325 uA in the non-AC ranges.

That sounds like the winner.  If you can pack 6 D-cells in there, you've achieved the 10K hour goal.  And they wouldn't need to be alkaline.

6x zinc-carbon C-cells would be enough for 10,000 hours if you don't do much AC measurement. With alkalines you'd be looking at 25,000 hours.

https://en.wikipedia.org/wiki/C_battery#Properties

With D-cells you'd be headed for 50,000 hours but I'm not sure they'd fit.

Edit: Yes, I just tried it and they'd fit easily!