FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1961
1961 - 1633.PDF
FLIGHT, 9 November 1961 737 Outer Shield Total Temperature Probes F. D. WERNER, BS, MS, PhD, R. V. DELEO, BS. MS,* and B. ROGAL, BScfEng, AMiEEt Klow Restriction Deiclng Heaters Sensing Element Airplane Skin Figs I and 2. Models 101 (left) and 102 total temperature probes TOTAL temperature measurement is much superior to skintemperature measurement. Temperature is essential totrue airspeed measurement and various jet-engine control,fire control, and bombing computations. In general, either meas- urement would do equally well if of sufficient accuracy. ForMach numbers above 0.5, temperature of the ambient air cannot be accurately measured by a "flush bulb" (which measures aircraftskin temperature), by reason of the facts that (a) the recovery factor is uncertain and (b) an additional error creeps in because the Machnumber at the bulb location may generally differ from the flight Mach number. Using typical installations for flush bulbs at re-spective Mach numbers of 0.6, 0.8 and 1.0, corresponding tempera- ture uncertainties can be readily calculated as —2.16, 4.86 and6.01cC. This problem can be substantially eliminated by the use of total temperature sensors. Extensive programmes have been conducted in the UnitedStates to determine the best methods of measuring total temperature, and three principal types of total-temperature sensors have beendeveloped as a result. One type, not de-iced, has been widely used * Dr F. D. Werner is the president, and Mr R. V. DeLeo manager,of the Aeronautical Research Department of Rosemount Engineering Company, Minneapolis, Minn, USA.t Mr B. Rogal is managing director of Research & Engineering Controls Ltd, Bognor Regis, Sussex, England. MICA SHEET DETAIL A V 100 TIMES SIZE / ~~- Fig 3. Construction of sensing element Fig 4. Speed of response lO-Or S SO h- 2o u 2 10 0-5 2O 5O IOO 2OO 5OO l.OOO 2,000 TOTAL PRESSURE BEHIND NORMAL SHOCK (mm Hg) < 4 _. for Service installations and has an excellent reliability record.Pictured in Fig 1, it is identified as the Model 101 series. A second variety (Fig 2) permits application of de-icing power whilestill permitting accurate total temperature measurements and is known as the Model 102 series. A third type, specially intendedfor flight-test purposes, is similar to that shown in Fig 2, except that the de-icing power is omitted and the element is an openwire for very fast response and high recovery factor. It is also included among the 102 series. Fig 3 shows the construction of the hermetically sealed element,in which the platinum wire is wound on a platinum tube, covered with another tube, and sealed with gold solder. Platinum tubes areused to ensure the uniform expansion of the platinum wire and supporting structure as temperature varies, thereby avoiding strainin the wire and promoting calibration stability. Returning for a moment to Fig 2, the construction involveswhat might be termed internal boundary-layer control, in the following way: The increased pressure within the probe housingcauses the internal boundary-layer air to move outward through the holes in the shell. Thus, heating of the shell for de-icing causesminimal disturbance of the temperature of the core of the airflow. This central flow is caused to turn 90 into the lateral conduit inwhich the sensing element is located. This turn eliminates water droplets by inertial separation. Except for the de-icing characteristic, performance data obtainedapply almost identically to both the 101 and 102 series. These results are summarized in Figs 4, 5 and 6. More detailed informa-tion is available from publications listed in the bibliography. The resistance v. temperature relationship is expressed preciselyin the Callendar-Van Dusen equation for pure platinum. A slight modification of the calibration curve by use of trimming resistorshas permitted many total-temperature sensors to be manufactured with interchangeabilityofO.LC over the range from - 50to -f 150°C. Fig 4 shows the speed of response, in terms of the time requiredfor 63.2 per cent of the response to a step function of total tem- perature. The time-constant depends primarily upon total pressurebehind the normal shock-wave. For the open-wire 102 series sensors mentioned, the response is more than 100 times faster(approximately 0.02 sec). Radiation errors are negligible, being typically less than 0.1 percent of absolute temperature up to Mach 2.5 and an altitude of 80,000ft. This applies to all three jsensors by virtue of similarityof internal aerodynamic design, which includes radiation shields. Fig 5 shows the recovery error, which is defined as the per centerror (per cent of absolute temperature). Fig 5 applies to the 101 and de-iced 102 series, but open-wire 102 series have recoveryerrors about one-fourth as large. The relation between recovery error v and recovery factor r, as well as the definitions of v and r. Fig 5. Recovery error 0.50 g 0.20 § < Q10 Q; U U < Q0£ r^ o.ci \ \ \ ^1 \ \ \ \ \ \ •-, \ \ \\ \\ \N\ \ \ , \ V X "\ -- Vy\\\ \;/ / \ % - 1 1 I- £>0% CO 4="-= t -iFlDEf \ CEL \ •• MIT 0.5 1.0 MACH NUMBER 50
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
