Picture of Digital multimeter shield for Arduino

This instructable describes a shield, which converts "Arduino" board in a digital multimeter (DMM).

The shield can be inserted on "Arduino" UNO und Duemilanove boards. It can work in three modes:

  • standalone - the measurement data can be seen at the character or graphical LCM
  • connected - the measurement data can be read on the PC screen using the "Arduino" IDE "Serial monitor"
  • combined - the data can be observed on both devices

The second mode does not require the presence of LCM, what makes the shield very cheap.

The "Arduino" based DMM has the following functions:

  • voltmeter with 3 ranges : 0-10V; 0-30V; 0-100V
  • amperemeter - it has a range 0-500mA
  • ohmmeter with 2 ranhes : 0-1KOhm, 0-250KOhm
  • diode, LED, connectivity checker
  • LED functionality tester
  • NPN BJT Beta meter
Remove these adsRemove these ads by Signing Up

Step 1: DMM "Arduino" v/s Standard digital multimeter

The following movie shows how the "Arduino" based DMM works in comparison with non-professional standard DMM

jfrontone5 days ago
Thankyou for the answer and I will do my best. Adding in some LED driver channels to test a line of LEDS and using the current generator for single. The max 407 is actually doing well. Wish you well in your work. Wish I had your job. If time will post it all with my efforts.
jfrontone6 days ago
Hello, I am currently building this project and wondering if you have any updates or as you discussed going further into the problem. I think cost should be last in our thoughts of making this better. I am wondering if the Max 407 off of the constant current generator can replace the LM358? Are there alternative ways to produce the trimming current? You were the only reference for anything like this and appreciate the post. I am putting this in a project box and going to make it more precise to use at work. I need to have 3 decimal points. Anyway please continue to work on this as I think this is the better project out of all of them for anyone in electronics for refresher and learning more about engineering. Well hope You consider to keep going.
Milen (author)  jfrontone6 days ago


I would like to wish you success in the building of the project.

It should be not a problem to use MAX407 as current generator. To have better results, you should use Arduino Due - its ADC has 12 bit resolution. The bad thing is that it uses 3.3V supply - some resistor values might need to be changed. Good thing of using the Arduino Due - it has 10bit DAC, which you could use for creating of some voltage reference (instead a Zenner diode) or to adjust fine the trimming current - the voltage of the DAC could be applied at the V2C converter input.

In the current time I do not have enough free time to do something additional in this project, but you can try to improve the design by yourself (it is not perfect). The only new thing I did is: I included additional menu item - random number generator - it produces for me a set of random numbers, which I can use to play in lottery (6 of 49) :-) . This feature is not published - it is specific for each country. If you improve the design I would suggest you to describe you modifications in a different instructable.

In all cases as this design is now - it should not be used at places where precise measurements are needed - for example at work. Additional work must be done also in the direction of enlarging the input resistance. As it is in the moment in some cases can be a reason for false measurements (if the output resistance of voltage the source you want to measure is relatively high).

Good luck in the playing....

jfrontone6 days ago
Oh and is there an update to this for further investigation on my part to make better.
jfrontone6 days ago
I am building this DMM. I noticed that a max 407 was used in the constant generator and wondering if that can replace the LM358. Also trying to find a generator for trimming. Is there an easier route to produce the current? The generator at school has limits. I might just have to build the generator. I had used your design last quarter to adapt this meter to a C8051F320/1. Everything worked except the obvious, Too much code and just needed to add a variable to finish. I am using the Arduino now for simplicity. Also debounce of the switch worked with a small delay and a small capacitor across the switch. Anyway thank you for being the only reference for this project. Having fun.
serkanshallow6 months ago

Hello. Thank you sharing your work. I have a question regarding the transistors. I dont wanna use smd transistor so can you advice me any?

Milen (author)  serkanshallow6 months ago

I suppose that you want to replace bss123
I think that this : http://ww1.microchip.com/downloads/en/DeviceDoc/VN2210%20E082013.pdf



can be used
yamil11111 months ago

Hi Milen, im doing this project, and im wondering how i measure the current?, i mean, no how works the amperimeter or the VIC, im saying how i conect to the circuit that i want to measure? the way that connects a regular amperimeter (in series to the circuit) or the way that i connect a voltimeter (in paralel to the device measure)?

Milen (author)  yamil11111 months ago
Hi Yamil111,

I have tried to explain this on step 2.... never mind,
the idea is the following - if you use the DMM device connected to you PC, then you can measure only currents which are sourced by devices having the same ground - in this case you connect only one terminal of the device (the red one), the current loop is closed through the common ground. In the case you want to measure a current flowing in some network, with unknown (different form you PC ground potential), it is better to supply the whole assembling (arduino+the shield) by a battery pack or isolated DC/DC converter.
So you use the whole assembling as standard DMM. You have to be aware, that this device does not measure negative currents - it means - the red input connector shall be always connected to the more positive node of the measured branch (the node, from which the current flows out). The black one (the local ground of the assembling supplied by battery) shall be connected to the node with less voltage potential (the node where current flows in). In this way it is possible to measure currents flowing in branches, which potential could be kilo-volts. In all cases, you have to be aware of the possible high voltages and to work carefully.
yamil111 Milen11 months ago

Thank you Milen, i understand what you're saying. Sorry, i miss up what you explain in step 2. But now, you clarify my doubt. Thank you, and excellent project.

AvrDon1 year ago

Hi Milen

Hi can I convert your DMM project to read voltages say from 80v ac and 150v volts ac. I'm wanting to building ac voltage monitor for single or 3 phase that I can read with the w5100 ethernet module was thinking a web page or telnet.



Milen (author)  AvrDon1 year ago
Hi Donnie,

You can rectify the AC voltage to DC and read its value. You have to change the voltage divider resistors (make them higher) and to change their ratio - at 150-160 V AC, after rectification, at A0 you must have 5V. You have to be aware that the AC voltage is measured as RMS - the amplitude is higher and the generated DC voltage is different. You have to calibrate the voltmeter with the proper coefficient. May be you have to add some correction because of the voltage drop over the rectifier diodes. And must be very care full - the banana sockets are not insulated and these voltages are dangerous!
Mr. E1 year ago

A quality project and great presentation ! The design discussion provides great information useful well beyond the scope of this wonderful project !

Have you considered storing additional calibration data in EEPROM to help deal with your non linearity concerns?

Milen (author)  Mr. E1 year ago
I did not considered this, but under desire it can be done. This is nice idea for people who want to go deeper in the problem. I preferred to try to correct mainly the linear behavior errors. Practically with careful device choice you can have only such kind. This is most valid for the transistor choice - if the transistor has low Vsat voltage, for Vds/Vce voltages over Vsat it has more or less linear behavior. The used NTD2955 is not the best choice, but using the PNP improves the linearity. Also suitable PMOS can be chosen so that it works also in the linear regime, and than the non linearity problem disappear. But, for the pure science...your idea is very good. It require more than two measurements and making some kind of look-up table or trying to find the equation, which models the performance...
achand81 year ago

Awesome !

granz1 year ago

To all potential builders: If you are powering this thing from your computer USB port you need to be very very careful what/where you are measuring. The USB port ground is usually earth grounded. This means that you can't go about measuring taking measurements with both leads on other mains-powered devices like you might with a commercial meter. Consider what happens if you try to measure a current in a circuit that is also mains powered, and you connect you ground lead somewhere in the circuit which isn't earth ground (Hint: you have created a short to ground via your PC). Commercial meters are either A) battery powered or B) isolated if they are mains powered. If you want to keep things safe, just power the Arduino from a battery pack.

dieferman1 year ago

Excellent , Thanks For Sharing !!!!

murfmv1 year ago

Milen, very nice job. Easy to follow good idea. How does one go about having a shield printed?

Milen (author)  murfmv1 year ago

Hi Murfmv. I have attached the design data. It is in "Eagle" format.

Download the ZIP file and un -zip it in some folder.

You can download the free lite version of the tool from here :


Install the program, start it and in the File menu>open one of the files (Schematic or Board) - the other will open automatically. Work with the board window. Press the "Ratnest" button - like "X" . Yo have to print separate the top and the bottom layers if you want to use the ink transfer method.

The top layer must be mirrored (in the print menu). Scale factor must be 1.

Solid and black checked. For the top layer - in the layer menu, first make all layer invisible, and then activate the following layers Top, Vias, Pads, Dimension.

For the bottom layer (not mirrored) activate the layers Bottom, Vias,Pads,Dimension.

The files are checked with the Iteadsudio checkset. They can be sent there, and you will have 10 boards for ~ 25 USD. If you want, I can attach also the gerber files.

Milen , this is a very impressive Instructable ! We all know that there are professional meters that can be purchased , but the impressive thing is this Instructable explains so much about what goes on inside a meter.

I feel ( and we are nearly the same age :) , that in our lifetime we have seen such an escalation in technology . I am not an expert in any aspect of electronics , but I am a lifetime student of this fascinating field. My father was enrolled in National Technical School correspondence learning and I was able to look through all his course materials in my early teens . Getting a basic understanding of Vacuum Tube theory , and then moving into transistor theory was so fascinating .

Integrated circuits started to simplify design , but we don't always take the time to appreciate what goes on inside and the engineering it took to create the special function chip.

Radio Shack had a great book out that I purchased on how to use your DMM , and that book helped me a lot . I hope that this Instructable gets a lot of attention as it covers in depth how a DMM works , and shows just how much design goes into a multimeter .

I think this would be a great project for College electronics students to build as it ties together electronics measurement theory and uses the Arduino controller which can be used for so many creative things.

I am going to download this Instructable as it is a keeper ! I will read through it more in depth as I am sure there is some interesting areas that will help me further my understanding.

Truly , I thank you for all the time and effort you put into this Instructable.


Milen (author)  Build_it_Bob1 year ago
Hi Build_it_Bob,

Thank you for the nice words. I also went similar way. I started to solder some parts when I was 12. My first working detector receiver I did in 1977. Long history from this time... :-).
I want to mention - the design is not the perfect one, and no very universal.. - it has few week points : the low input resistance, the accuracy, the ranges comparing with standard industrial DMM. But his main purpose is to show a way of thinking - how using few devices a lot of functions can be implemented, how to have relatively accurate device, even you use standard and not trimmed building parts...This instructable is still not finished. Remain few additional things to be done - the ohmmeter trimming...and may be the conclusion.
aaguilar211 year ago
does anyone know some good books for easy to learn basic on electronics my nephew wants to learn n build
Milen (author)  aaguilar211 year ago

How old is he? Does he understand formulas?

May be the best solution should be to buy him some design kit with possibility to experiment with plenty of circuits and to learn the electronics practicing it... this will keep his interest, instead reading some book with theory.

he 14 n i have bought him some kits before n he really good but now he wants to take things apart but i rather he made stuff on his own from parts from old radios n stuff like that ... but i would like him to learn the basics
Milen (author)  aaguilar211 year ago

I did a small research in the internet...it seems that a lot of people recommend more or less the same books:


and a classical example:


There are also some very useful sites like:


I hope that this would give you him a nice start.

and thank yu for yur replys :)
thegoodhen1 year ago

That's nice! I am just thinking about designing and building my own universal multimeter-but mine will be benchtop and higher end. I will use dedicated adc chips, voltage references and relays. But... I have to point out that your design has EXTREMELY low input impedance. I understand that you don't wanna go all the way into megaohms because of noise, but maybe you could add a low noise input buffer?

Milen (author)  thegoodhen1 year ago
You are absolutely right. The input resistance is low. I decided to make a compromise, also with the ranges... only to keep the schematics simpler and cheaper. But I think that for applications, for which this DMM could be used ( Arduino projects, digital projects... etc), the low resistance would not be such a problem. I wish you good luck with you project, and I hope that we all will see after a nice Instructable describing it.


I see. I didn't understand the reasoning behind the low resistance, but now I do-it's to keep the complexity down. I wasn't sure if you are fully aware of this being a flaw, so I wanted to point it out! I would also like to mention that I am not trying to nitpick the flaws of your (I think brilliant) design, but with this in mind... Is the spacing between the rails sufficient? It seems like the 100 volts might easily spark over... Maybe that's just me though.

Milen (author)  thegoodhen1 year ago

As I wrote, before - I did compromise also with the ranges. May be 100V is not the suitable range, but here is another problem - I do not have any insulation of the banana socket - if there is really high voltage, it can be dangerous for the users. I did not want to have any dangerous high voltages on the board.The Zener diode is able to pass over 100 mA. This current multiplied by the input resistance of 10 KOhm, gives a voltage drop over 1000V at the input resistor. It is 1W and for some time will survive. But I do not suggest, that anybody tries that ! :-).

I want also to mention something additional about why I put so small input resistance - I did not want to use a buffer. I do not know how exactly the AD conversion is done in the Atmega chip. There is Sample & Hold circuit, but I do not know is there any buffer inside the chip before the S&H circuit. If no buffer exist, it is possible that the time constant of the S&H circuit capacitor and the big input resistance is enough high comparing with the sample time to introduce huge error. (I had some problems with ADC's without buffers before the S&H circuit, which were sampling false value because of the high output resistance of the previous circuit). OK, here the filtering capacitor of 10 nF helps, but with higher input resistance this capacitor also will create time constant and if the measurement was short in time, it would be also not correct.

msabhi631 year ago

ommg/////.........nice......one more possibility......

rpotts21 year ago
oh! realistically, how would this compare to a harborfreight model? same? better? WAY better?
Milen (author)  rpotts21 year ago

The described DMM has some limitations - The ranges are not the most appropriate.

The reason for that - the ADC has reference voltage 5v - that defines an input range for Vmeas 5V. This DMM can be done relatively precise inside this region, but comparing with even harborfreight model, it has the following disadvantages:

1) the ranges...

2)the accuracy - for the industry DMM special dedicated chips are used, which are trimmed internally, have the required resolution. For the "Arduino" based DMM, the resolution is limited to 10 bit...

3)the input resistance is relatively low...

In other words - the proposed solution is cheap, relatively precise, but with some limitations

rpotts21 year ago
Nice writeup! So you used a combination of opamps and voltage dividers for the ranges?
Milen (author)  rpotts21 year ago


The voltage dividers are used for the voltage measurements...this approach use also the professional voltmeters. For the current measurement using voltage gain amplifies, may be is not the best solution, because the error is also amplifies. There a different set of very precise resistors is used. Their connection is done by mechanical switch, which contacts have practically 0 resistance and do not influence the accuracy. I wanted to use this approach for the current measurements (also for the voltage), I had to use some kind of relays (reed) , what would make the circuit more expensive and heavy. That is the reason I have only one range for the current measurements.