Loaded resistance measurement with NI 9217

[g.hobold]

Hi,

Is is possible to measure the resistance using the 9217 while the resistance is loaded?

I'm trying to perform a temperature measurement using a custom ~0.5 ohm Nichrome resistance wire.

I have available to me a cDAQ 9178 with the NI 9217 RTD measurement module.

Given that I will be dissipating current (up to 50A at voltages lower than 30V), I would like to know whether it would be possible to measure resistance while simultaneously loading the wire.

I am able to measure the resistance of the wire as a function of temperature with reasonable accuracy, but only when it is not loaded.

Thanks!

[Henrik_Volkers]

Yes, it is possible since U=R*I 🙂

You want R, you need to know U and I.  Yes, it's that simple 😄 😄

You have a current source, is it a stable current source , with known output impedance?

If no,  with the addition of an extra current shunt and two voltage measurements (one at the shunt the other at your resistor) your done.

Since the input voltage range of the 9217 is 50mV to COM both resistors (shunt and wire ) need to fall into that 50mV ...   .. sorry, even if it is not impossible, it will be better to use a 2 channel differential voltage input card.

If you really absolutely must have to use the 9217....   tell us, and we maybe find a way, depending on the fact how to cheat the 3/4wire detection and current source 😉  warning soldering involved 

BTW: Accuracy needed?

[g.hobold]

Thanks for the reply.

The measurement must be very precise. I am measuring resistance variations on the order of ~0.0001 ohm.

Unfortunately, while I do have a controllable current source, I don't think the power supply is sufficiently precise to trust its informed current value to calculate current. Additionally, I also don't think I have a sufficiently precise 30V voltmeter. Hence why I am trying to solve this issue using only the NI 9217.

[Henrik_Volkers]

1/5000 ratio isn't that hard.

I would use a 2channel diff input  simuitanious! sampling unit. Speed doesn't seems to be the problem (24bit sigmadelta?) but multiplexed inputs a prone to ghosting.

get a good four terminal current shunt with low tempco

other way: build a bridge circuit.

[Henrik_Volkers]

I have an idea of how you can measure current and voltage with your 9217:

using a 5mOhm current shunt to measure load current (read ohms but with a R0_read/5mOhm =Ix:1mA ratio)

and measure voltage with a voltage divider R1 and R2 where R1_read=Ux(R1/(R1+R2)) / 1mA

Results in Rx =R1_read/R0_read * 5mOhm * (R1+R2)/R1

BUT  unfortunately this assumes that the 1mA current of the 9217 is equal and constant for two channels.  And I'm not sure that this is an valid assumption, since a valid and very common way to do such measurements is to measure the coarse 1mA during RTD measurements.