Question 3. What are the important operational problems with infrared thermometers 

 and what possible solutions are there? 



Calibration. It was agreed that blackbody calibration would be best. There should 

 be preflight, in-flight and postflight calibration checks for close tolerance work. The water 

 in calibration containers should be kept as near to the prevailing surface temperature as 

 possible. There was recommended a simple blackbody calibrator described as a metal 

 container (rectangular gallon gasoline can) with a black- surface cone immersed in water 

 within it, the temperature of which can be determined by an immersion thermometer, thus 

 giving the temperature of the blackbody cone very closely; an automatic stirrer within the 

 can would be helpful. A factor that would cause error here is the fact that sea-water 

 emissivity is not exactly that of a blackbody; the probable effect of this must be determined. 



If a water bath is used for calibration it is important that the water in the container 

 be strongly agitated at the time of the calibration check. If the water is not agitated, cali- 

 bration error of up to 1.0°C. can result. 



It is realized that these solutions cannot apply simply to fixed-mounted instruments 

 in flight but they can work easily for the tripod- mounted IRT. 



Power supply. With the IT family of instruments there is a limited tolerance to 

 AC frequency variations. If the input voltage is more than 0.4 cycle off, the instrument 

 should be recalibrated to that frequency. An apparent temperature difference of 1.5°C has 

 been shown to result from a variation of 1.2 cycles. Since many power sources presently 

 used with the IRT are inexpensive inverters with no frequency control, the DC input must 

 be closely watched since the AC output frequency is a function of the DC input voltage. 

 However, frequency controlled inverters which eliminate this problem can be purchased 

 for approximately $500 (such as one produced by Accurate Instrument Company). The new 

 model IT-3 will have tolerance to operate between 58.7 and 61.3 cycles without significant 

 error. Nevertheless, it is suggested that each instrument ensemble include a frequency 

 meter and a voltmeter on the power input. It was also suggested that Barnes Engineering 

 consider designing IRT equipment for airborne use utilizing line frequencies varying be- 

 tween 380 and 420 c.p.s. 



Detector cavity problems. For use in aircraft, with the considerable amount of 

 acoustic vibration experienced, the IRT detector cavity must be tightly sealed. If not, the 

 air within the detector cavity can be set in motion and the record is useless. The late pro- 

 duction runs of the IT-1 series and the early runs of the IT-2 series had unsealed cavities 

 and as a result gave poor performance in aircraft; all models now produced are sealed. 

 Further improvement could be made by evacuating the detector cavity. 



The sensor head must be kept inside the aircraft and properly shielded and pro- 

 tected from the aircraft slipstream. Rapid air motion over the sensor head will affect 

 the temperature of the optics and possibly the detector cavity itself. Also the speed of 

 the chopper blade can be affected. The best mounting arrangement appears to be such that 

 the opening for the sensor should allow for a gentle outflow of air from the aircraft; this 



-18- 



