[SOLVED] Kelvins double bridge connections

[AzharElectronics]

Hi All,
I have a substance which measures mili ohm (around 40 mili-ohm) resistance and due to its face wear the resistance changes.
Wear out causes change in resistance of few micro ohms (around 70 micro-ohm). And I wanted to measure this change in resistance using some electronics circuit.
Initially I tried using Wheatstone Bridge but I was not able succeed. Later when I did some study online I got to know Kelvins double bridge is used to measure low resistances.
My subject material (substance) can be four wired. Hence I wanted to know how these four wire can be connected in Kelvins double bridge ?
Whether using Kelvins double bridge is suitable or some other technique is available to measure change in micro ohm resistance ? please let me know that other technique if available.
Also If we are using Kelvins double bridge how practically I need to balance the second arm of Kelvins double bridge ?
A Picture diagram will help.
Thanks & Regards,
Azhar

 

[KlausST]

Hi,

when yo do an internet search about ""kelvin double bridge" you find a lot of informations.

***
I´d do it with a usual "kelvin wiring":
Two "force" wires ... to introduce current.
Two sense wires to measure the voltage.

The precision mainly depends on the introduced current and the introduced waveform.
Is AC excitation allowed? If yes, then look for a "Lock In Amplifier". It´s not that complicated as it seems.
What max. current is allowed?

Klaus

 

[AzharElectronics]

Do I need to use specific wires for Kelvin wiring or general 22/24 AWG will solve the purpose, specific wiring like Kelvin Probes, Jigs, Cable Clamp or PCB etc.
Yes AC Excitation is allowed.
Current up to 1 ampere can be designed (lesser upto 500mA will be easy to design).

 

[KlausST]

Hi,

wire gaues is not critical with 4 wire measurement.
But for sure the force wires should not cause a lot of voltage drop @ measurment_current.

To reduce EM influence I recommend to use twisted pair sense wires.

1A at a resistor variation of 70uA causes a voltage variation of just 70uV.
Thermocouple effects at each junction may cause DC offset and thus error with a DC measurement method.
The higher the current the higher the usable voltage.

Klaus

 

[AzharElectronics]

Hi,
Thank You for your reply and all suggestions.
1. How much current do you suggest for design to detect change of 70µΩ ? I do understand current limit depends on my design options but wanted to hear from you what can be ideal current value to design.
2. Can Current Pulse be used to read the small resistance ? (Pulse of 100ms)
3. For Suggested AC Excitation can I use below circuit from Analog Devices for design ?
https://www.analog.com/media/en/technical-documentation/application-notes/AN-683.pdf

 

[KlausST]

Hi,

To clarify: I see no improvement in using a bridge configuration. Simple 4 wire measurement should do the job.

1)
I can not give a value, because it depends on your "substance". We do not know anything about your "substance".
From the signal - or better say the signal-to-noise - view: The higher the current, the better.

2)
Yes. Don´t expect a signal improvement. It´s just to reduce power consumption.

3)
Yes.
Expensive solution. You can build your own lock in amplifier with cheap CMOS analog switches and a comparator.

Klaus