Measuring low resistances with Arduino

What have you looked at so far?

Haven't tried anything in circuit yet. The only thing that came to mind was a voltage divider circuit, but with a bit of googling I came to the conclusion that the Arduino doesn't have enough resolution to measure such low resistances with a voltage divider. Maybe some kind of amplification would be possible, but I don't have the knowledge to do such a thing.

Any help would be appreciated, thanks.

Hi!

Bipolar transistors BJT's are current controlled devices, Changing the amount of current flowing through base to emitter alters the amount of a separate current flowing through collector-emitter, great for amplifying!  http://www.allaboutcircuits.com/textbook/semiconductors/chpt-4/bipolar-junction-transistors-bjt/

The change in resistance of a small current would manifest as a change in current from a different more abundant source and therefore easier to measure. Sometimes it's necessary to chain multiple transistors in what's known as a Darlington configuration to further boost that through more transistors.

Thanks for the info. I'll look into that.

I also considered the possibile usage of an opamp (a component I have never used before). I watched a couple of Youtube videos about them and this is what I came up with. The component circled in red is the resistor which I'm trying to measure. The wire marked with blue is what I would use to read the signal to the Arduino. The assumption is that this would give me a signal voltage of 0.74 - 2.95 V  to the Arduino depending on the resistance (0.6 - 1.6 ohm)

Am I even close with this one? Would the opamp be able to convert a voltage of 0.022 - 0.087 V to 0.74 -  2.95 V smooth(ish)ly?

Hi,

I did consider this approach too, boosting the voltage with an op-amp is also a potential.. There are often many ways to skin a cat!!

Have you tried changing the 220 ohm resistor for a lower value? That would initially increase the range of the voltage maybe even bring it into range to be measured by an analog to digital converter built into your microcontroller (a higher sample range on the adc would be beneficial. If you can choose a 10bit option over an 8bit this would be better as would 12bit. Im not certain what range the Arduino has?).

Changing the 220ohm resistor for a lower value seems a good idea to start with, this will increase the "baseline" voltage, choosing right the baseline can be increased from 0.022v to 0.74v(or within reasonable parameters.) This leaves the differential range the same so there would still only be a very slight change when the variable resistors value changes.

What is the nature of the variable resistance? is it an actual variable resistor or something else? maybe there's a bit of trickery we can do without needing op-amps or transistors..

The arduino UNO already has an analogue comparator that can be used for this, use one end connected to a set point (Perhaps from a PWM output with an RC to emulate an 8bit DAC or simply a resistor divider, then the comparator will trip once the threshold is reached, knowing the value of the DAC will indicate if the resistor is above or below the threshold to a roughly 1% accuracy. selecting the right resistors to produce say 0 - 1V output and then the 1R + some other value to limit the current can be used for the other input

No external op amp needed in this case

alternate and probably easier is to use a resistor divider with the target resistor in question as part of this. Set the VREF to the ADC to be the 1.1V reference, not the 5V reference (Default). this will give you far better resolution at the low voltage end.

the arduino has a 10bit ADC so 0 - 1023 value which equates to roughly 1mV per count (Assuming the 1.1V VREF)

so a 100R in series with 1R and 5V across the two will result in 5v/101R = 50mA, this across the1R gives 50mV

a 0.5R will give about 25mV

1.5R will give 75mV

the mV equate to the rough count given from the ADC

Yes your working in the low end of the ADC but given your not interested in accurate measurement of the resistor, just a rough go/no go test, this should be more than enough

if you want a better resolution then you could drop the resistor to say 50ohms which will give roughly 100mA through the divider and so 1R = 100mV (Roughly), 150mV = 1R5, 50mV = 0R5

Not bad for just the addition of a single resistor

The arduino UNO already has an analogue comparator that can be used for this, use one end connected to a set point (Perhaps from a PWM output with an RC to emulate an 8bit DAC or simply a resistor divider, then the comparator will trip once the threshold is reached, knowing the value of the DAC will indicate if the resistor is above or below the threshold to a roughly 1% accuracy. selecting the right resistors to produce say 0 - 1V output and then the 1R + some other value to limit the current can be used for the other input

No external op amp needed in this case

alternate and probably easier is to use a resistor divider with the target resistor in question as part of this. Set the VREF to the ADC to be the 1.1V reference, not the 5V reference (Default). this will give you far better resolution at the low voltage end.

the arduino has a 10bit ADC so 0 - 1023 value which equates to roughly 1mV per count (Assuming the 1.1V VREF)

so a 100R in series with 1R and 5V across the two will result in 5v/101R = 50mA, this across the1R gives 50mV

a 0.5R will give about 25mV

1.5R will give 75mV

the mV equate to the rough count given from the ADC

Yes your working in the low end of the ADC but given your not interested in accurate measurement of the resistor, just a rough go/no go test, this should be more than enough

if you want a better resolution then you could drop the resistor to say 50ohms which will give roughly 100mA through the divider and so 1R = 100mV (Roughly), 150mV = 1R5, 50mV = 0R5

Not bad for just the addition of a single resistor

Thanks, Peter!

I totally forgot about the 1.1V VREF of the Arduino. I have a much better feeling of this working now.

Hi!

Just tried it with a 100 ohm resistor and set the Arduino's analog reference to 1.1V, and it worked fine! I used a 12 volt source, though. I just read the analog pin and for my purposes, when the analogRead value is larger than 53, the resistor is at the correct value.

Thank you so much, guys!

With a 12V source, you risk feeding 12V into the analogue pin of the Aarduino when the resistor under test is not connected, this is not good

A 5.1V zener across the input and ground should protect it though, you probably are having the internal protection diodes ctivating at the moment but they may not last with 70mA passing through it

Either way though, good news that you solved it

Peter

Thanks, I'm going to look into zener diodes.

I'm going to have a relay or a mosfet so that I'm not reading the voltage divider at the same time as I am powering the heating element. The resistance of the element will be under 2 ohms at all times, so I shouldn't get anywhere near 5 volts out of the voltage divider provided that my Arduino code doesn't fail