Diving Into Calibration Curves and How They Work for You

When calibrating a unit, it is best to visualize the changes done as a Calibration Curve. A Calibration Curve is a graph that shows the relationship between the output of a measuring instrument and the actual value of a quantity being measured. It's used to convert an instrument's output readings into the actual measurement value, which helps to improve measurement accuracy. Ref 1, 2 This means that by using known gas quantities, we can adjust the drift of the sensors and make the unit as accurate as possible.

Of course, with sensors drifting due to time/age, or outside stimuli, it means they will need regular adjustment to ensure they stay on target for as long as possible. Once the adjustment can no longer be made or the readings are inaccurate regardless of the adjustment, then it’s time to get a new sensor.

As shown in the picture, the left graph shows the ideal calibration curve. The predicted values match the actual value which shows that the analyzer giving the predicted values is accurate. On the right, you can see the same graph but after some drift with the sensors. Ref 3 Our calibrations aim to restore the unit as closely as possible to the ideal graph. Each adjustment we make will have a slight effect on the other values no explicitly adjusted to. This part of the process involves setting a “zero” point and a “span” point. Ref 4 By setting these points, we can give the unit the information it needs to adjust around those points, similar to a line on a graph. The more points we can adjust to, the more information the unit has to make the smaller corrections.

Each unit is used to a different extent and in different ways giving them each a sort of “finger print” which can cause the calibration process to differ slightly between them due to issues arising on one unit that may not be a problem on another. That’s why it’s best not to risk it and have your unit annually checked with the professionals!