DMM for low R measurement

[JB]

Does anyone know a DMM (preferably PXI) to measure resistances in the order of 90mΩ ?

I'm currently using the NI PXI-4071 but its minimal range of 100Ω produces sometimes inaccurate results.

Many thanks in advance for any advice.

[dmsilev]

I'm more familiar with benchtop instruments than PXI, but in the context of benchtops, the first thing I'd consider would be 7 1/2 digit multimeters, for instance a Keithley 2001 or 2010, or similar units from e.g. Keysight. Lowest resistance range is 10-20 Ohms, with accuracy (in 4-wire mode) below 1 mOhm.

If that's not good enough for your application, and if your budget allows, consider an ac resistance bridge. A Lakeshore 372 has full-scale resistance ranges down to 2 mOhm, and with care the noise level can be in the hundreds or even tens of nanoOhms range for milliOhm signals. If you can find a Linear Resarch LR-700 on the used market, that will have comparable performance. 

[Stephan_D]

Hey JB,

could you provide use more information why the NI PXI-4071 dosn't seems to be accurate enough?

According to the manual, the PXI-4071 provides a 100 Ohm range with 10 uOhm resolution. Which resoulution do you need for your application?

Regards, Stephan

[JB]

Thank you for these interesting instruments.

A PXI solution would be my first choice because it would require less changes. 

[JB]

The resolution is not the problem.

The DMM is used to measure the resistances of the three coils of a motor. A matrix is used to make the connections. Sometimes, the measurement is inaccurate (eg 87mΩ instead of 90).

• Because of the quite high minimal range of the DMM (100Ω vs 90mΩ coil resistance), a 7½ digits measurement is essential. Whith this configuration, the DMM needs about 1.1 second for one 4W R measurement. This long time doesn't allow to average several measurements because this would slow down the cadence of the assembly line. 
• The measurement current of the PXI-4071 is only 1mA.

Therefore I'm looking for a faster DMM with a higher current and more suitable for low resistances.

Obviously, my preference would be to keep the PXI-4071 and I'm very interested in any recommendations which would help to improve the measurement.

[Wayne.C · Accepted solution]

JB,

I don't use any PXI but would suggest you look at the Programmable Supplies/SMU's for PXI.  Should allow you to precisely source current and measure the resulting voltage drop.  Some of these instruments have a built in configuration to do resistance readings.

[crossrulz]

---

Wayne.C wrote:

I don't use any PXI but would suggest you look at the Programmable Supplies/SMU's for PXI.  Should allow you to precisely source current and measure the resulting voltage drop.  Some of these instruments have a built in configuration to do resistance readings.

---

If we are dealing with mOhms, this is actually a very good way to go.  To take it a step farther, I have used an SMU to source a current and then use a DMM to measure the voltage across the pins of the UUT.  It is like doing your own 4 wire resistance measurement, but having full control of the current you source.  R = V/I.

[JB]

Thank you very much for your advice.

Do you have any experience with this setup combining a SMU and a DMM ?

Unless I'm wrong, a PXI-4130 would be able to measure the current and the voltage; making the DMM optional.

I guess the measurement (PXI-4130=voltage+current or DMM=voltage and PXI-4110 or 4130=current) would be much faster than a 4W R measurement with the DMM.

[crossrulz]

I have done that with a Keithley source meter.  I found the DMM was able to measure the voltage better than the sourcemeter.  Probably since the sourcemeter had to switch the MUX to measure the current and then the voltage.

[JB]

Too late to edit my previous message.

After reading its datasheet, it appears that the 0.1mV resolution of the PXI-4130 is insufficient to get an accurate measurement for a 90mΩ resistance. Therefore it is much better to keep the DMM to measure the voltage.

I guess that this R=U/I measurement (U by DMM and I by SMU) is much faster than a R measurement with the DMM. Is this correct ?

Could anyone provide a approximate duration ?

~1.1s for the 4W R measurement with the DMM (7½ digits, offset compensation)