Why use a bridge circuit?

July 14, 2011
Bridge Circuit for Sensor with Differential Amplifier

Bridge Circuit for Sensor with Differential Amplifier

Many sensors today change in resistance value to reflect the condition change they test for. Often the change in resistance is small compared to the base resistance value of the sensor device. In the usual voltage divider configuration the voltage developed across the resistor may only change in the third or fourth significant figure meaning that the top several bits of your ADC are really going to waste. If there is AC pickup (60Hz) or 120Hz from a badly filtered power source or some other noise source that is also going to be exaggerated across the total device and may swamp the signal you want to measure. First you must shield all wiring you can and connect your shields to a common ground point. You don’t want the shield current to share the same paths to ground as your signal.

The objective is to just measure the change to the resistor so that’s where the bridge circuit comes in. A bridge is really just two voltage dividers where is one is made with matched precision resistors (R1 and R2) to get precisely 1/2 of the supply voltage to the bridge. The other side consists of the sensor resistor at the top of a divider (R5) and a resistor at the bottom (R3 + R4) whose value is the maximum of the range of the sensor resistance value. This makes sure the voltage on the + side of the following differential amplifier is always greater than the 1/2 V coming to the minus side. The amplification factor you set in the differential amp depends on the range of the maximum swing of the voltage going into the differential amp and the voltage range of the following ADC.

The sensors may not have a consistent range of values so I’ve inserted potentiometer R4 in the circuit that can be adjusted to the maximum resistance of the particular sensor. Note that the value of R3 + R4 must be greater than the maximum resistance of the sensor R5 to keep the + input of the differential amplifier more positive than the – input. This is because I’m using single rail supplies for everything to keep things simple. Note that the supply for the analog side should be separate from the digital side and should be well filtered. On PC boards the analog and digital circuitry should have separate grounds (and ground planes) that get connected to each other and the cable shields at one point and one point only.

In the schematic above I’ve chosen a x10 multiplier in the differential amplifier so the formula for the resistance of the sensor is calculated as follows:

Vs = 0.5*V + 0.1*Vda  (Vda=differential amplifier output voltage)

Vk = V – Vs  (Vk is the voltage across R3+R4)

Ik = Vk/(R3+R4)  (R3+R4 must be accurately measured before hand so you may want to sacrifice a bit of resolution and use a known value precision resistor in place of the R3+R4 combination but the value must be larger than the maximum value of any sensors that will be used.)

Rs = Vm/Ik  (the resistance of the sensor is thus calculated)