How to verify metal oxide dew-point sensors in the field

It's possible to make a valid verification check of a Michell Easidew dew-point transmitter in the field. The simplest way to verify the performance of the sensor by using gas with a certified H₂O content.

Cylinders of gas (N₂ is fine) with certified H₂O content are available from most specialty gas producers but it is crucial to buy a cylinder with a certificate and stated uncertainty, as without a known reference the results of your test will be invalid.

Ensure that you specify an H₂O content which is as close to your typical measurement point as possible. Ideally it should be slightly wetter than the typical measurement to reduce test time and gas consumption.

You will also need the following items:

• High-integrity pressure regulator with stainless steel internals (to reduce test duration and prevent ambient moisture ingress)
• Clean stainless-steel tubing and compression fittings for making connections between cylinder and transmitter (to prevent ambient moisture ingress)
• Calibrated pressure gauge or regulator of an appropriate range (dew point is highly pressure dependent)
• Stainless steel needle valve
• A stainless steel Easidew sample block

The test procedure is as follows:

1. Ensure that the Easidew transmitter is exposed to dry gas. This can be for the following timescales, depending on the dew point prior to testing:

   Dew-Point Temperature °Cdp	Minimum Stabilisation Time
   -100	5 days
   -90	5 days
   -80	3 days
   -70	3 days
   -60	24 hours

2. Connect your pressure regulator and the calibrated gauge and valve downstream of this. Also, connect recommended sample tubing required. Configuration should look like the flow diagram below:

   

3. Purge the regulator and tubing with dry gas for the same timescales listed in #1.
4. Quickly connect your regulator, tubing and valve to the calibration cylinder and resume purging.
5. Connect your Easidew to the calibration gas.
6. Set the pressure to 0.5bar(g).
7. Set the flowrate to around 2Nl/min.
8. Leave to stabilize for as long as possible (ideally 6 hours).
   
   

   Note: to limit the effects of temperature on the adsorption/desorption rates of moisture in the regulator and tubing, time this test to begin at the time of day when the temperature is as stable as possible. Ideally conduct the test in a temperature stable environment.

9. The reading from the sensor should be representative of the H2O content listed on the cylinder.

The alternatives to the above are:

• Comparison of your sample gas against another calibrated hygrometer
• Calibrate against a chilled mirror reference hygrometer
• Service exchange
• Returning the sensor to Michell UK for laboratory 'As Found' readings