XDT User’s Manual Appendices
Appendix F: Sensor/SpanCheck™ Theory of Operation
The Sensor is constructed as a capacitor whose dielectric consists of porous Aluminum Oxide as well as the gas that has entered in the pores of the Aluminum Oxide. The plates (electrodes) of this capacitor are an aluminum substrate and a porous gold layer deposited on top of the Alumi- num Oxide, the porous gold electrode allows transfer of gases into or out of the Aluminum Oxide pores.
The capacitance due to the Aluminum Oxide is always constant, while the capacitance due to the gas varies according to the gas content and pressure. Since the dielectric constant of water is orders of magnitude larger than that of any gases being measured, the quantity of water vapor present in the pores changes the capacitance of the sensor to a much greater extent than any other system variable. For the same reason (the extremely large dielectric constant of the water mole- cule), any capacitance variations arising from the Aluminum Oxide, such as changes due to tem- perature, are insignificant in relation to the capacitance due to the water content. Thus the sensor capacitance varies greatly in proportion to the water content in the surrounding gas. The sensor is designed such that the relationship of the quantity of water and the resultant capacitance has an "S" shaped curve. At the extremely dry end the curve is asymptotic to the capacitance due to the Aluminum Oxide and the gas, while at the very wet end the curve is asymptotic to the capacitance due to water molecules packed extremely tightly (saturated) in the pores of the Aluminum Oxide. This upper end is an excellent indication of the total pore volume; while the lower end of the curve is an indication of the distance between, & area of the capacitor plates (electrodes), this is the intrinsic capacitance of the sensor. Therefore knowing these two points, a sensor can be cali- brated with compensation for small manufacturing deviations.
As the sensor is used in real world
|
|
| Xentaur HTF™ |
| Sensor: Dewpoint vs. Capacitance Response Curve |
|
| applications, it is exposed to various | |||||||||||||
| 200 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| elements and stress; like any other sen- |
|
|
| Saturated Capacitance Asymtote of New Sensor |
|
|
|
|
|
|
|
|
| sor it may over time drift from calibra- | ||||||||
|
|
| Saturated Capacitance Asymtote of Aged Sensor |
|
|
|
|
|
|
|
|
| |||||||||
| 160 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| tion. However one must note that the |
|
|
|
|
|
|
|
|
| rCurve | nsor | Curve |
|
|
|
|
| |||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |||||||
|
|
|
|
|
|
|
| Senso |
|
|
|
|
|
|
| intrinsic capacitance will not change | |||||
(nF) |
|
|
|
|
|
|
|
| Aged | Se |
|
|
|
|
|
|
| ||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| ||||||
120 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| ||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| ||
Capacitance |
|
|
|
|
| New |
|
|
|
|
|
|
|
|
|
|
|
|
| with use because it is based simply on | |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| ||
80 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| the dielectric constant of Al2O3 and the | |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| distance between, & area of the plates | ||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| ||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
| 40 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| (electrodes), thus the original factory |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| measured values will always be valid. | |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| ||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| What may change is the pore volume, | ||||
|
|
|
|
|
|
|
|
| Dry Capacitance Asymtote of New & Aged Sensor |
|
|
| |||||||||
| 0 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| due to clogging with contaminants, |
|
|
|
|
| 0 |
| 20 | ||||||||||||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
| Dewpoint Reference ˚C |
|
|
|
|
| residual oxidation, metal migration, etc. | |||||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Thus to
instrument needs only to “know” the capacitance at the wet end of the curve. As discussed previ- ously the capacitance at the wet end approaches asymptotically a saturated pore capacitance, therefore if the sensor is saturated (the exact water content will not be important since the curve is asymptotic), then the instrument can measure the capacitance and
45