5 dollar voltmeter with 5 digit (0.1mv) resolution

[splin]

But anyway, soldier Paulie seems to be lost somewhere in the wilderness and I still don't know what the eBay cheap ammeters impedance is  ;-) !!

47.5 milli-Ohms including leads as measured on my 34401A in 4W mode. I bought 4 from this sellor (100% positive feedback) and had no problems:

http://www.aliexpress.com/item/0-3-0000A-Five-digit-0-56-Digital-Ammeter-Four-LinesCurrent-Panel-Meter-Display-Color-Yellow/990444797.html#feedback

The other seller, RD Factory, oddly will sell one of those for $4.94 but if you want more you lose the free shipping! Their feedback score is also slightly less at 99.3% but probably doesn't mean much.

I'm delighted with these meters which seem to be reasonably accurate:

Meter 1    HP34401A
0.0000      0A
0.5000      0.500042A
1.0000      1.000106A
2.0000      1.99602A
3.0000      2.99742A

Meter 2
0.0000      0A
0.5000      0.500152A
1.0000      0.999871A
2.0000      1.99904A
3.0000      2.99815A

However,  my meter hasn't been calibrated for some time so that doesn't mean much either!

[EDIT: corrected Meter 1 1.0000A measurment]

[ym58]

47.5 milli-Ohms including leads as measured on my 34401A in 4W mode. I bought 4 from this sellor (100% positive feedback) and had no problems:
http://www.aliexpress.com/item/0-3-0000A-Five-digit-0-56-Digital-Ammeter-Four-LinesCurrent-Panel-Meter-Display-Color-Yellow/990444797.html#feedback

50mR, that's good enough !

I was even wondering if this model from the same seller would not be a better deal ... coz what I REALLY need is 1mA resolution (way enough to control Li-Ion charging currents), so I think that a 4-digit 0-3.000A ammeter can really fulfill my needs.
On top of that, it's only 3.50€ a piece with free shipping plus a dual V+A display !



Have you also tried that one ?

[splin]

Sorry - good try! But that dual meter is incredibly good value and perfect for a bench PSU. If I hadn't already got several of the 5 digit meters I'd definitely try them out. Please let us know how they perform if you do go for it.

[ym58]

Sorry - good try! But that dual meter is incredibly good value and perfect for a bench PSU. If I hadn't already got several of the 5 digit meters I'd definitely try them out. Please let us know how they perform if you do go for it.

yes, maybe I'll even try one to replace the cheap voltmeter that I installed on my "old-time" customized ATX-PSU several years ago ...



a bit "crowded" inside but so powerful a PSU indeed !

(edit 08:48) just ordered 4 pieces for 14EUR !!!

[ym58]

I am back at EEV with good news : I managed to install two of these 3.50EUR Chinese volt-ammeters on both my PSUs : the vintage ATX and the (trickier) dual 30V/3A SL-1731SB !

A bit tedious to get all things done (especially the dual PSU) but I guess it will worth it !
Nota: the big pot on the dual PSU is a voltage multiturn that I installed a few months back.

[Behemot]

I'll resurrect the thread if you won't mind. It has been suggested here https://www.eevblog.com/forum/projects/i-keep-on-blowing-up-panel-meters!/ there may be some problems running multiple meters from the same power supply. Since these are the meters I'd like to get (the 4wire versions) and we even have a schematic…

So my intention is to run 5 meters to measure output voltage of a power supply (between its 5 rails and GND). And than five more meters to measure voltage accross precise current shunts (so not against ground). I would like to use single power supply to power all the meters (plus some fans), insulated from the measured circuit. Can this work, or do I have to use 10 different insulated power supplies? That would be somewhat, eh, impractical if you see my point.

Now I may or may not connect GND of both the measured PSU and the PSU for the meters, however, as most SMPS are ground referenced, they will actually share a common earth ground anyway (unless I select class II PSU which is not ground referenced). Does this complicate things further?

[Kalvin]

Why don't you use those cheap isolated dc-dc-converters to isolate the meters from each other. A small, isolated 1W 5V/12V or 12v/12V dc-dc-converter will cost around $2 - $4 a piece. To play safe, you will need N isolated dc-dc-converters for N volt meters.  Just go to eBay and enter the magic keywords: isolated 1W dc-dc converter. If the 1W is too little, there are 2W and 3W models available with a bit higher price tag.

[Kleinstein]

It depends on the actual circuit in the meters. It may work to have several voltmeters that measure relative to a common ground powered from the same source. But still you have to check the circuit or carefully try out (with extra resistors for protection first).

For the current measurement on floating shunts you will need a separate supply for each meter. The DC/DC converters or transformers (e.g. 2 outputs from one transformer is common).

[Behemot]

Those converters look nice, there are double converters in single package for few bucks so I'll need just about five pieces. Didn't even know it came in so many variants so cheap…I've just seen the larger uninsulated converters. Haven't been watching this particular market, thanks for the suggestion.

Just seems I'll have to shield it properly so it does not induce extra unnecessary EMI as it's going to be lots of switching stuff in the usuall tens of kHz band. I think I'll make use of some old PC case to house the whole thing so I can also use some ditched PSU casing for the DC-DC unit. This way I'll also use the hole in the back of the case 

Guess I can use thin coaxial cables to run supply for the meters which should radiate less noise on the way, right? To be extra paranoid maybe even with two conductors+shielding so I can use the first layer as return path and the outter one earth-grounded as EMI shielding. Will be much more cabling than using common supply line, but well…

ADD// found the name - triaxial cable

[001]

Funny tale begins 

Looks like these china panel meters became certified 'laboratory' bench units (0.1%):

Any opinions?


P.S.: Can I move decimal point on these beauties without custom firmware? (from 33.000V to 330.00mA, 330V, etc)

[Behemot]

I think it's more than adequate for like schools etc. If the assembly company guarantees accuracy and some input protection, what's really the problem?

[001]

The reference voltage is not a problem?
And what about cruel 0.1% at 5 digits? 

[Behemot]

They may be using different reference etc., what's the problem? As stated, guarantee the accuracy. It's their problem how they do it. It may very well be just a display having nothing in common with these voltmeters, whaddaya know…

[edavid]

Looks like these china panel meters became certified 'laboratory' bench units (0.1%):

Why do you say that?

P.S.: Can I move decimal point on these beauties without custom firmware? (from 33.000V to 330.00mA, 330V, etc)

No.

[001]

Sorry for my english
But attached photo is https://vk.cc/66tqXs
Unit manual guaranted at least  0.1% tolerance

[edavid]

Sorry for my english
But attached photo is https://vk.cc/66tqXs
Unit manual guaranted at least  0.1% tolerance

That is a 4.5 digit meter, so it is obviously not using the particular China DPM module that we are discussing.

[sperb]

Sorry for resurrecting this topic, but I purchased a meter similar to this one and decided to make a custom software for it.

The functionality is similar to the one existing on the original module: 0-35.000V and 0-4.4000V.

Reflashing the unit will allow you to change the gains, ranges, decimal point position and perform a recalibration.

Feel free to use, make changes and update as needed.

The software was created following the explanation existing here: https://embedded-lab.com/blog/category/stm8/ using the STVD and Cosmic C compiller.

Best regards.

[update]

As requested, follow (attached) some pictures from the unit I have.

Unit was bought from Ebay. link: https://www.ebay.com/itm/184581349615

[Vovk_Z]

My goal here might be to calibrate one of these so accuracy can match resolution or as close as possible at room temperature. Unfortunately I no longer have access to a precision lab voltmeter so the plan is to send a home built voltage reference to a friend who does and compare with readings from one of these 4.3000v meters.

What about basing it on something like lm4040a which has 0.1.% initial accuracy spec? Any opinions?

They are typically calibrated at a factory. All precision 5-digit meters I have bought were well-calibrated. But this calibration flows away with age as in any other meter.
0.1% or 0.05% voltage reference is fine not only for those panel meters but for most DMMs too.

[Kleinstein]

The 35 V range and likely also the 4.4 V range would likely need external dividers. So the gain would be set by the reference and the xtra divider. Usually this is done with a software scale factor and not by using 0.1% accurate parts. So even a 2% accurate reference like LM399 is OK for a 6 digit DMM.

AFAIK the original low cost panel meter use an MCP3421 ADC with internal reference only. The internal ref. my be of comparable quality as an LM4040. 

[Canonius]

Hello,
i think either your project is not ready, or you uploaded something wrong. Because for example this variables
   value2 = value2_loc;                                             // update display variable
   value = value_loc;                                                // update display variable
   dp = dp2;
will never be applied to the 7 segment display. At least this routine is missing.

Please, can you share the final source code with us? I would really appreciate it because i need to replace the analog meters from my old power supply.
Can you also tell me which option bits should be set?
I am not familiar with the STM8 and i have to use the IAR IDE because i didnt get the damn licence file for the cosmic compiler up till now.

Thanks in advance!

Sorry for resurrecting this topic, but I purchased a meter similar to this one and decided to make a custom software for it.

The functionality is similar to the one existing on the original module: 0-35.000V and 0-4.4000V.

Reflashing the unit will allow you to change the gains, ranges, decimal point position and perform a recalibration.

Feel free to use, make changes and update as needed.

The software was created following the explanation existing here: https://embedded-lab.com/blog/category/stm8/ using the STVD and Cosmic C compiller.

Best regards.

[sperb]

Hi Canonius.

The version i shared is the latest I have. And it is supposed to work.

The variables that you are seeing (value2, value and dp) are used by the interrupt routine to update the display. Please check the file stm8s_it.c and you will find those variables being used to update the 7 segment display.

I don't remember if I had to mess with the option bits, but if they are being set, you will find them in the other files of the project. Just  keep digging.

If you want something ready, I recommend using the platform I used (wait for the cosmic compiler). The code I shared should work then. And you can customize as you want after that.  I have no experience with  the IAR IDE, so I don't know how hard will be to convert the project to that platform.

Hope it helps.

[Canonius]

Hi Sperb,
please don't get me wrong. I am really happy that people like you sharing these things!

Now i got the license file for the Cosmic compiler. I will check your project later that day and give here feedback!

Thanks again!

Regards

EDIT:
Very Nice! it worked out of the box!
Thank you very much!

Hi Canonius.

The version i shared is the latest I have. And it is supposed to work.

The variables that you are seeing (value2, value and dp) are used by the interrupt routine to update the display. Please check the file stm8s_it.c and you will find those variables being used to update the 7 segment display.

I don't remember if I had to mess with the option bits, but if they are being set, you will find them in the other files of the project. Just  keep digging.

If you want something ready, I recommend using the platform I used (wait for the cosmic compiler). The code I shared should work then. And you can customize as you want after that.  I have no experience with  the IAR IDE, so I don't know how hard will be to convert the project to that platform.

Hope it helps.

[b_force]

Little kick on this threat, mostly because I am actually looking for a 5 digit meter that sits in the mV range, about 900mV max range (or 999.99 I guess).
Ideally one that can also be adjusted for calibration, since I have a 0.02% voltage reference here.
Most I can find only go up to 200mV, which is a littler awkward, since my range is around 100-400mV or so.

Obviously I could cobble something together with a micro and a proper ADC, but I am more looking for an of the shelf affordable solution.

[Kleinstein]

Many of these low cost versions use a MCP3421 ADC and a µC like STM8. The MCP3421 has a range up to +-2 V and a few internal gain stages to get ranges like 1 V , 500 mV and 250 mV. So the HW should supprot a higher range, just a question of the SW. Similar the calibration / adjustment is a SW question. This may be even supported, just no documentation provided.

[b_force]

Many of these low cost versions use a MCP3421 ADC and a µC like STM8. The MCP3421 has a range up to +-2 V and a few internal gain stages to get ranges like 1 V , 500 mV and 250 mV. So the HW should supprot a higher range, just a question of the SW. Similar the calibration / adjustment is a SW question. This may be even supported, just no documentation provided.

Yeah, that's also what I realized after this topic.

I was more kind of hoping someone would have direct experience, example or like a quick and easy tutorial or something.
Another idea is rolling my own PCB or something.

Post 141 is already in the right direction
I was more looking for a ready made solution.

edit: another option would be adding some external gain, but unfortunately that adds some extra error and accuracy.