Good Specsmanship: How well do you understand RH specifications?

This week we discuss best practices in relative humidity specifications.

Check Device Specs + Know what to look for

It's a topic close to home, because of Vaisala's history as creator of the world's most robust capacitive thin-film polymer sensor: HUMICAP. Now, we (usually) try not to use this blog for bragging about our awesome products, BUT, just as all humidity sensors are not created equal, neither are all product specifications. There are some real points of contention on this in our industry.

There are values concerning humidity calibration that are crucial to understanding the accuracy and operation of a humidity sensor. The truth is that these values are often misrepresented or non-existent on humidity instrument specifications. So, this week, at risk of tooting our own horn, we are going to highlight the elements of humidity instrument specs that you need to understand to match the device to your operating environment, regulatory constraints, and application requirements.

Drift... a sad reality of RH Measurement

It's an immutable law of RH measurement. Sensors drift for the simple reason that they are "air breathers." Unlike temperature sensors, the internal structure of the humidity sensor must be in direct contact with the environment, which is constantly changing. Not only is the environment in flux, but it contains airborne contaminants that degrade the sensor over time. This is why, even if the calibration process were perfect (and sadly, it isn't), once exposed to the real world, the measurement accuracy inevitably decreases.

Terms to Know

Look for information on these features of the device or its calibration processes:

Repeatability: Closeness of the agreement between the results of successive measurements of the same measurand carried out under the same conditions
Non-linearity: The closeness to which a calibration curve approximates a specified straight line, and linearity error is the absolute maximum deviation between the calibration curve and the specified straight line.
Hysteresis: The property of a device whereby it gives different output values in relation to its input values depending on the directional sequence in which the input values have been applied. Hysteresis error is the maximum deviation between the two calibration curves of the measured variable as obtained by an upscale going transverse and a downing transverse over the full range. More simply stated, hysteresis is the tendency of measuring devices to not return completely to their original state after a change has been measured. When measuring relative humidity, it can be a major source of error. Unfortunately, too few data sheets include hysteresis as a factor in their accuracy values. Hysteresis unmentioned in a spec sheet can be considered product data misrepresentation.
Calibration Uncertainty: Uncertainty is the parameter associated with the result of a measurement, which characterizes the dispersion of the values that could reasonably be attributed to the measurand. It comprises many components. the uncertainty may be, for example, a standard deviation having a stated level of confidence.

In humidity calibration, one factor affecting uncertainty is the fact all humidity chambers have an associated uncertainty, a major source of which is temperature non-uniformity. This must be factored into a measuring device's accuracy specification. Before humidity calibration, Vaisala performs a high-accuracy temperature calibration on every data logger. Each device's measured temperature is then able to compensate for chamber non-uniformity during RH calibration — greatly reducing this source of error. Another element to control for error is that with a Vaisala DL logger the temperature sensor is placed right beside the RH sensor. This proximity allows both sensors to "read" the same environment, eliminating discrepancies between their measurements.

The Art & Science of Discerning Specsmanship

Product specifications, often one of the only ways decision makers can select a suitable system, must be explicit, easy-to-understand, and straightforward. All of the known influences and sources of error — calibration uncertainty, temperature effect, measurement resolution, and hysteresis — should be included in the accuracy value stated on any data sheet. If these values are not mentioned on a product data sheet, have they been included in that product's stated accuracy?

A Working Solution to Better Spec Evaluation

Product specifications are a key element in the decision-making process of device procurement. Good specs are explicit, easy-to-understand, and straightforward. All known influences and sources of error — calibration uncertainty, temperature effect, measurement resolution, and hysteresis — should be mentioned if not stated value on any data sheet. If these values are not mentioned on a product data sheet, how can you be certain that they been included in that product's stated accuracy?

So, here's what you do if you aren't satisfied with the spec sheet that comes with your device: ask for calibration uncertainty statement. If the accuracy values are not visible on the data sheet, check the calibration certificate. A professionally made calibration uncertainty calculation includes not only the uncertainty of the reference instrument or also factors related to the calibration process and unit under test. If the calibration is performed by an accredited laboratory, the uncertainty estimations are always inspected by qualified technical specialists. The quality of calculations varies strongly among other (standard) calibration providers.