wheatstone bridge vs 4-wire measurement for straingauge measurements

[ravekanth]

I had thins in a different forum, maybe I will have more luck here 🙂

This may be a very basic question, which I do not seem to find a concrete answer to(r maybe I do not grasp it all that well)
Hopefully I have better luck asking it directly.

I have to do strain measurements using strain gauges and I have an NI9205 AI module. I do understand how a wheatstone bridge works, but is it really necessary over say using a 4 -wire measurement system where I do nto use any other resistances?

As an added question, what would I need to go for when measuring any resistance when I have something like the NI9205 module?

wheatstone bridge or 4-wire measurements.

Hope this dilemna is clear enough( and this is the right place to post this)

[Henrik_Volkers]

The usual 2 or 4 terminal resistance measurements is fine to measure the resistance but to measure small relative changes you need a very high resolution over the hole range. The bridge allows you to adapt the available range (resolution an accucary) to the range of interest . You don't want to measure 360+x Ohm , you are interested in the +x.  And x is RELATIVE and RELATIVE small in a strain gauge 😉

The bridge is also a way of resisitance measurement and has some benefits for a strain gauge. You measure imbalances in the bridge by comparing resistors (in a  strain gauge) , this (mostly can) eliminate(s) other effects like temperature drifts (if all or at least two of the resistors share the same temperature and self heating current)  and thermo and galvanic  EMF  in an AC brigde. The most precise resistance measurements are still done with a (four terminal) bridge but the realization ist more complex to compare the 'unknown' resistor with a Quantum-Hall-Element 🙂 .

With your 9205 you can measure voltages and since U=R*I  you need a known current to measure resistance.  A REF200 together with an external power supply can provide a quite accurate 100µA to 400µA (see datasheet).

Or you use a voltage (or current) source  and measure the current with a known stable resistor in series to your unknown.

The question where to use a bridge or not  can be answered if you have a actual usecase , the bridge can always be  more accurate but is not as flexible as a current and voltage measurement.

If you have a straingauge and a 9205, go with a bridge.

[ravekanth]

Thanks that does make things much clearer.

I have an electrical power supply(Rigdol DP1116A) which I do use as a current source and the source seems to be stable enough. 
My understanding of input module is that it being a 16bit device with a programmable rage of +-200mV. I should have steps of 6.1µV.
Does that mean I should be able to observe changes of at least 6.1µV or higher. That was the actual intention behind asking if the bridge configuration was necessary. 

Please do corrrect me if I am wrong or there are other factors which I need to take care of as well. 

[Henrik_Volkers]

First have a look into the spec, there is a part for the uncertaincy calculation.

Do you have a single strain gauge element (quarter bridge) ?

[ravekanth]

Yes. I have a single strain gauge element.
The accuracy for the NI9205 is +-.157mV