Strange behaviour of a Wheatstone bridge

Question

I am trying to do some strain measurements. I am using a strain gauge in a Wheatstone bridge configuration. The connection is shown in the image below:

^{simulate this circuit – Schematic created using CircuitLab}

I am using an amplifier and some circuitry to amplify and read the Vo signal. SG1 is a strain gauge. R3 is set at resistance of 120 ohm. The voltage across SG1 and R2 (or) R3 and R4 is 2.6 V. As all the resistors are of equal value the voltage across each resistor should be 1.3V. When I turn on the Vex the voltage across R3 and R4 is 1.3 V each initially. After sometime I notice that the voltage across R3 goes up to 1.47 and the voltage across R4 is 1.1 roughly. The voltage across R3 was increasing like 1 mV/s. I tried to vary the potentiometer to balance the bridge, but the voltage was fixed across R3 and did not decrease.

Can anyone explain why does this happen and how to get rid of this?

I'm using through hole metal film resistors and soldered my connection on a prototype PCB.

Answer 1

Do not use R3 in a rhoestat configuration. The wiper to track contact resistance will be unstable, causing you large and varying errors at the high current running through it.

It's better to use it in a potentiometer configuration like this, assuming that your sensing circuit is relatively high impedance.

^{simulate this circuit – Schematic created using CircuitLab}

This has various advantages over your circuit

• The R3 wiper is now carrying no, or only low, current, so that changes in the wiper-track contact resistance have a minimal effect
• Choose R4=R5, and make R3 smaller than them, this increases the available resolution at balance
• Choose the R3/4/5 arm to have higher resistances, to reduce your current consumption and heating errors