Measuring NTC Thermistor Resistance In Live Circuit

[danner123]

Hi, I know with a generic resistor to measure the resistance accurately it should be disconnected from the circuit and measured by itself. I am trying to measure the resistance of an NTC over time in a live circuit. I want to create a plot of the NTC temperature vs NTC resistance over time.

How do I measure the NTC resistance in a live circuit (powered up)? Can I just measure with a DMM?

Thanks!

 

[KlausST]

Hi,

I see two methods for measuring an R:
1) Measuring V and I and applying Ohm´s law.

2) adding a known resistor and measure V across the known resistor and measuring V across the unknown resistor. Then R_x = R_k * V_x / V_k

***
In a live circuit, you can not know how the excitation signal looks like. It may be DC, AC or pulsed ..it may be current or voltage.
Any added circuitry, like voltmeter, amperemeter, resistor may corrupt the measurement.

Thus none of the above methods is secure.

I reccommend to analyze the excitation signal first.

Klaus

 

[FvM]

Unless multiple NTCs are MUXed, the live circuit is most likely a simple voltage divider. If so, the NTC resistance can be easily derived from a simple voltage measurement.

In the general case, you need "ideal" current and voltage measurements to determine the resistance. That's possible but not simple.

 

[barry]

Unless multiple NTCs are MUXed, the live circuit is most likely a simple voltage divider. If so, the NTC resistance can be easily derived from a simple voltage measurement.

This presumes that you know the exact values of the OTHER components in the circuit. This is no more likely than knowing the value of the NTC in question. Either you trust the data sheet(s), or you don't.


WHAT is the "live circuit"? Is it a constant-current source? A bridge?

And I have to ask the question: Why are you doing this? If you are trying to verify the NTC, then you should REALLY do this before even installing it in the circuit. Maybe you can use a known-good sensor and compare its readings to the ones from your in-circuit device.

 

[andre_luis]

I want to create a plot of the NTC temperature vs NTC resistance over time

You certainly are aware of the self heating impact on the measured value; although in general the NTC resistance do not drift too much due to this, in order "to measure the resistance accurately" as you said, a side effect as such should not be negleted.

 

[c_mitra]

How do I measure the NTC resistance in a live circuit (powered up)...

First you need a circuit diagram and lacking that, a model of the circuit.

Resistance is a property that manifests only under an external potential. So you need to measure the potential across the NTC thermistor first. If that is a DC voltage, that is fine but that for a NTC the applied voltage may change as a function of time.

Now the model comes next: you need to see a branch that has another resistor in series with the NTC. You need to measure the voltage at the same time across this fixed resistor too.

Rest is simple.