Calibration Curve

Sometimes we need to characterize the behavior of a sensor for its measurements to be useful. This is especially true in the world of strain gages that output a millivolt signal proportional to the voltage input based on the amount of stress they are placed under. For our example, we are going to look at creating a calibration curve for a 2 mV/V-100 PSI pressure transmitter.

First, we will apply pressure and record the output millivolt readings. We will also monitor the excitation voltage throughout the process since knowing the millivolt per volt value relies on the voltage part of the equation. We will enter this data into an excel spreadsheet.

To calculate the mV/V readings we will concoct a small formula in excel to subtract the mV reading at zero PSI from the pressurized reading and divide it by the excitation voltage. This gives us our mV/V at each of the pressure test points.

The reported value for full-scale mV/V is simply the value we calculated for the full-scale pressure reading. To obtain the zero value divide the mV/V value by the excitation voltage.

Insert a line graph. The default trendline is “linear”. When we add the trend line to the graph, the “format trend line” dialog will appear to the right of the screen in excel. Near the bottom check the box that says “display equation on chart.” The formula will appear on the graph in the form of a y=mx+b format. Since we have the mV/V plotted on the “x” axis we can plug in our reading divided by the excitation voltage to calculate the pressure reading.

Most modern data acquisition systems allow us to plug these formulas in and use them to interpret data from sensors. By building calibration curves for inexpensive sensors, we can use them to make measurements that might have been reserved for more expensive scalable sensors in the past.