Thermohygrometer calibration


Starting to get Involved
We use some thermohygrometers to monitor environmental conditions in a controlled environment (medical device manufacturing). We appear to have established some somewhat arbitrary limits a number of years ago. I've never been worried about reaching a limit without someone complaining well beforehand.

When it comes to calibrating equipment of this type and purpose it seems many manufacturers do a single point or two point calibration. I think this is reasonable because the range of operation is very small (50 to 85°F) but our quality engineers insist we should be using a 3 point calibration. The cost to do this exceeds the value of the equipment so we are in the position of replacing the equipment instead of calibrating it.

I am curious what other people in the industry are doing.

We have about 15 thermohygrometers. We also have an environmental chamber that holds temperature very constant and has minimal quick variation in RH. Our strategy is to send one unit out for external calibration and then perform internal calibrations on the other units. We are not FDA regulated, though, which may make a difference here. I would think that with the wide range you do have that an internal calibration would still provide a low enough uncertainty to qualify. You may also be able to shop around for a lower calibration cost (we're charged $150 for 3-point on both T and RH plus $50 if an adjustment is needed).

While buying a new unit is tempting, you need to be cautious--an auditor may ask you to prove the mothballed unit's readings were valid at the end.

Some other thoughts:
1. We have glass thermometers that we calibrate at two points using an ice bath and boiling water.
2. The electronic sensors for T and RH exhibit good linearity. You may be able to do an initial performance qualification of linearity using three points and then do two point external calibrations after that.
3. You may be able to test and show that there is no effect to your product over a much wider range than what is comfortable for your employees. If so, you may be able to justify there is no risk if you do not monitor the T/RH in those processes, then eliminate the requirement or indicate it is for employee comfort only.

One other note: if your units use paper charts be aware that the chart information may be wrong. On a unit where the pen would show an obvious step change for a 1 %RH change, I once saw a chart where there was no step change and there was a 1%RH difference in the reading on opposite sides of the (circular) chart.