Sec. 1-7J MECHANICAL INPUT TRANSDUCERS 145 



decomposed by electrolysis into gaseous H 2 and 2 . The current / 

 during the time t will decompose a mass of water of 



m water = Alt 



If a humid gas flows continuously through the tube with a constant- 

 mass flow velocity v m = m mixture /2 on m gas /tf, the system will come to 

 an equilibrium when the rate of water absorbed by the P 2 5 layer is 

 equal to that electrolytically decomposed. The current is then 



*£*„ (1) 



where V w and V a are the partial volumina of water and gas. The 

 constant Jc is 



2 X 96,500 PT water 



18.016 X TF gas 



where IF water and W g&8 are the gram-molecular weights of water and 

 of the gas. In the case of water in air, Eq. (1) will take the form 



I = 0.0066 X cv m 



where c ( = F tt ,/F air ) is the concentration by volume of water in air, 

 expressed in parts per million (ppm). If the volume flow velocity is 

 100 cc/min, the current will be 13.2 ^A/ppm by volume (at a tem- 

 perature of 25°C and a pressure of 1 atm). 



The instrument is particularly useful for very small concentrations 

 of water in the gas, between 1 and 1,000 ppm; commercial instru- 

 ments (e.g., Beckman Hygrometer, Fullerton, Calif.) have in general 

 ranges from 0-10 to 0-1,000 ppm. The lower limit is about 1 ppm. 



The accuracy of the system is between ±3 to ±5 per cent of the 

 full scale reading. In the lower ranges of humidity, the water in the 

 flowing gas will not come into equilibrium with that in the P 2 5 

 layer. The "efficiency" of the absorption depends upon the flow rate 

 and the humidity in the gas. Correction curves for freon 12 are pub- 

 lished by Taylor. 1 Because of the incomplete absorption, empirical 

 calibration is frequently required. 



The dynamic response of the instrument is moderately high. If the 

 humidity is suddenly increased, the instrument will reach 63 per cent 

 of the final indication within 1 min. The time lag for a decrease of 

 humidity is about 2 min. 



The pressure of the gas mixture and its mass flow velocity must be 

 kept constant. If the volume flow velocity is kept constant, as it is in 



1 E. S. Taylor, Refrig. Eng., 64, 41 (July, 1956). 



