Go Down

Topic: Ideas for a precision ohmmeter (Read 6719 times) previous topic - next topic

brucethehoon

I've been throwing around ideas for an Arduino based precision Ohmmeter lately.  The simple voltage divider method is cheap and easy, but I'd like something to rival my good Fluke desktop meter.

If I cob up a good constant current source and use a good 12 bit ADC, it seems like I would get some pretty accurate reads. 

While not trying to make a big amazing LCR meter, I feel like we should be able to out do most expensive multimeters.

Any suggestions from folks that have gone down the cc/adc route before?

thanks!

cmiyc

Precision and accuracy aren't the same thing...  Precision is a matter of digits.  Accuracy is the quality that those digits represent.

How accurate your meter is depends on more than just the ADC and measurement method you use.  You need to calibrate out the error introduced from the traces, wires, and leads between your ADC pins and your point of test.
Capacitor Expert By Day, Enginerd by night.  ||  Personal Blog: www.baldengineer.com  || Electronics Tutorials for Beginners:  www.addohms.com

brucethehoon

My apologies.   My initial post was going to focus on the benefits of using the 12bit ADC to get the enhanced precision over the 10bit built-in ADC.  I then didn't bother to go back and visit the methodology and calibration sides.

I intend to have a zeroing routine that will zero first at the closest point to the ADC, then at the point between where the leads connect to the board.   After that, a touch-zero can help.

Aside from trace width and other board design concerns, do you have any ideas about overall methodology?  I've constructed several ohmmeters before that I considered reasonably accurate (they agreed with my Fluke 8846 to 4 digits), but just wanted to milk all the performance I could out of a small, cheap Arduino based solution that would make it easy to output readings any way I like, and be a fun little project.


robtillaart

Never build an ohm meter, but I would at add a temperature sensor in it. That would give the ability to correct for temperature drift.
Also measuring thermistors and other temperature dependant resistors would be possible.

With an peltier element and temp sensor you could create a reference temperature of e.g. 25C before measure the resistor.

Rob Tillaart

Nederlandse sectie - http://arduino.cc/forum/index.php/board,77.0.html -
(Please do not PM for private consultancy)

terryking228

Hi,
One of the challenges is low-resistance measurement which require 4-wire connections. Have you looked at that?

A quick DIY I have used for larger motor windings etc is simply use a bench supply to put 1 amp or 10 amps through the device under test, and measure the voltage across it.  Not highly accurate but OK for quick measurements without the cool equipment you are talking about...
Regards, Terry King terry@yourduino.com  - Check great prices, devices and Arduino-related boards at http://YourDuino.com
HOW-TO: http://ArduinoInfo.Info

ec212

Im sure this will help you out:

https://www.youtube.com/watch?v=0vMvCVyOp9g

The key is " R=V/I
 
If "I" (current) is = 1Amp, the resistance will be equal to the voltage.

AndreyS

Im sure this will help you out:

https://www.youtube.com/watch?v=0vMvCVyOp9g

The key is " R=V/I
 
If "I" (current) is = 1Amp, the resistance will be equal to the voltage.
Is it safe to use this method to find a short on the motherboard if I limit voltage 1.2V and current 1A?
I have vcore shorted to the ground somewhere.

DVDdoug

Quote
Is it safe to use this method to find a short on the motherboard if I limit the voltage 1.2V and current 1A?I have vcore shorted to the ground somewhere.
I wouldn't recommend 1 A amp with a motherboard or any regular low-power circuit board, at least not on a routine basis.   It's one thing to test a wire, motor, or transformer with 1 Amp or more, but you could burn a trace off  a board, or fry a component with reverse voltage, etc.

However, I might be OK occasionally if the board is dead and you've got nothing to loose...   I've used a high current power supply to burn-off a power-to-ground short a couple of times, but most of the times I've tried it, that doesn't work.    And even if that does clear the short, you don't always see where it was and you may leave a very-small gap that can get shorted again.


I've got a 4-wire meter at work and although I've never measured it's current, I'm sure it no more than a standard multi-meter and perfectly safe for any components.

AndreyS

#8
Dec 12, 2015, 06:49 pm Last Edit: Dec 12, 2015, 06:58 pm by AndreyS
I wouldn't recommend 1 A amp with a motherboard or any regular low-power circuit board, at least not on a routine basis.   It's one thing to test a wire, motor, or transformer with 1 Amp or more, but you could burn a trace off  a board, or fry a component with reverse voltage, etc.

However, I might be OK occasionally if the board is dead and you've got nothing to loose...   I've used a high current power supply to burn-off a power-to-ground short a couple of times, but most of the times I've tried it, that doesn't work.    And even if that does clear the short, you don't always see where it was and you may leave a very-small gap that can get shorted again.


I've got a 4-wire meter at work and although I've never measured it's current, I'm sure it no more than a standard multi-meter and perfectly safe for any components.
I asking because this guy does this
I feel like if I check circuits which are already short, the current will flow through least resistance anyway, and i feel like I'm not gonna do worse.

bisbrmn


Go Up