1、_ SAE Technical Standards Board Rules provide that: “This report is published by SAE to advance the state of technical and engineering sciences. The use of this report is entirely voluntary, and its applicability and suitability for any particular use, including any patent infringement arising there
2、from, is the sole responsibility of the user.” SAE reviews each technical report at least every five years at which time it may be revised, reaffirmed, stabilized, or cancelled. SAE invites your written comments and suggestions. Copyright 2014 SAE International All rights reserved. No part of this p
3、ublication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of SAE. TO PLACE A DOCUMENT ORDER: Tel: 877-606-7323 (inside USA and Canada) Tel: +1 724-776-497
4、0 (outside USA) Fax: 724-776-0790 Email: CustomerServicesae.org SAE WEB ADDRESS: http:/www.sae.org SAE values your input. To provide feedback on this Technical Report, please visit http:/www.sae.org/technical/standards/AIR46B AEROSPACE INFORMATION REPORT AIR46 REV. B Issued 1956-03 Revised 1996-11 R
5、eaffirmed 2014-05 The Preparation and Use of Chromel-Alumel Thermocouples for Aircraft Gas Turbine Engines RATIONALE AIR46B has been reaffirmed to comply with the SAE five-year review policy. FOREWORDThis document, originally prepared in 1956, was reviewed by industry and committee membersbetween 19
6、83 and 1988. Those who reviewed it generally agreed that the document, with minorchanges, would continue to be a useful industry reference. Changes were limited to updating the list ofreferences as well as to exhibit temperatures in both F and C.Changes made in 1996 were format/editorial only.TABLE
7、OF CONTENTSFOREWORD .11. SCOPE 41.1 Purpose42. REFERENCES.43. PROCURING AND CALIBRATING TYPE K (CHROMEL AND ALUMEL)THERMOELEMENTS.53.1 Specification of Thermocouple Wire.53.2 Methods of Making Thermocouple Junctions53.3 Calibration of Thermocouple Wire 64. CIRCUITRY FOR HIGHEST ACCURACY74.1 Lead Wir
8、es and Terminals74.2 The Reference or Cold Junction.84.3 Automatic Cold Junction Compensation .94.4 The Selector Switch9TABLE OF CONTENTS (Continued)5. INDICATING AND MEASURING INSTRUMENTS .95.1 Analog Millivoltmeters.105.2 Digital Millivoltmeters105.3 Servo-driven Indicators.105.4 Potentiometers .1
9、06. CORRECTIONS APPLICABLE TO GAS TURBINE THERMOCOUPLES 116.1 Conduction Correction126.2 Heat Gain by Convection136.3 Radiation Correction.136.4 Impact Effect 146.5 Rate of Response.157. LABORATORY EVALUATION OF THERMOCOUPLE PERFORMANCE 167.1 Correction for Radiation and Conduction177.2 Determinatio
10、n of Recovery Factor 197.3 Determination of Characteristic Time .208. LOCATING THE MEASURING JUNCTION219. PARALLEL THERMOCOUPLE NETWORKS.229.1 Common-Terminal Systems .229.2 Ladder-Type Systems 249.3 Electrical Analog for Parallel-Thermocouple Networks .269.4 Connections Between Thermocouples and Ha
11、rness 2610. POSSIBLE EFFECTS OF FABRICATION ON TURBOJET THERMOCOUPLEPERFORMANCE2710.1 Mechanical Working .2710.2 Moisture, Fuel, and Sulfur in the Magnesia.2810.3 Terminal Connectors for Thermocouples3011. TESTING GAS TURBINE THERMOCOUPLES3111.1 For Initial Calibration and Stability After Use.3111.2
12、 With Commercially Available Equipment 346$(,17(51$7,21$/ $,5% 3DJHRI_TABLE OF CONTENTS (Continued)12. REFERENCE DATA ON CHROMEL AND ALUMEL AND ON THETEMPERATURE SCALE3512.1 Composition .3512.2 Processing .3512.3 Physical Properties.3612.4 Stability in Service 3712.5 Thermocouple Lead Wire .3812.6 T
13、he International Practical Temperature Scale.38APPENDIX A DEFINITIONS OF SYMBOLS.536$(,17(51$7,21$/ $,5% 3DJHRI_1. SCOPE:This SAE Aerospace Information Report (AIR) reviews the precautions that must be taken and thecorrections which must be evaluated and applied if the experimental error in measurin
14、g thetemperature of a hot gas stream with a thermocouple is to be kept to a practicable minimum. Discussions will focus on Type K thermocouples. These are defined in NBS Monograph 125 asnickel-chromium alloy versus nickel-aluminum alloy thermocouples.1.1 Purpose:The purpose of this document is to as
15、sist those concerned with measuring gas temperatures inturbines and those who wish to utilize gas temperature for actuating engine controls. In theseapplications, the problem of temperature measurement may be considered to consist of twodistinct parts; namely, a determination of the actual temperatu
16、re of the measuring junction and areliable estimation of the difference in temperature between this junction and the gas. The formerrequires that numerous precautions be observed in the construction, installation, and use of thethermocouple; the latter requires proper application of appropriate corr
17、ections for such effects asimpact and heat exchange by conduction, convection, and radiation.2. REFERENCES:2.1 W. F. Roeser and H. T. Wensel, Methods of Testing Thermocouples and ThermocoupleMaterials. J. Research Nat. Bur. Standards 14, 247 (1935).2.2 W. H. McAdams, Heat Transmission, 3rd ed. McGra
18、w-Hill, New York (1954).2.3 E. F. Fiock and A. I. Dahl, Temperature Measurements in High-Velocity Streams of Hot Gas. Proc. Iowa Thermodynamics Symposium, p. 190. State Univ. Iowa, Iowa City (April 1953). AlsoNBS Report 3164 (April 1953).2.4 A. I. Dahl and E. F. Fiock, Response Characteristics of Te
19、mperature-Sensing Elements for Usein the Control of Jet Engines. J. Research Nat. Bur. Standards 45, 292 (1950).2.5 M. W. Carbon, H. J. Kutsch and G. A. Hawkins, The Response of Thermocouples to RapidGas-Temperature Changes. Trans. Am. Soc. Mech. Engrs. 72, 655 (1950).2.6 A. I. Dahl and E. F. Fiock,
20、 Use of Parallel Thermocouples in Turbojet Engines. USAF TechnicalReport No. 6546 (July 1951).2.7 A. I. Dahl and E. F. Fiock, Circuitry Errors of Ladder-Type Thermocouple-Harness Assemblies. WADC Tech. Report 53-4 (Dec. 1952).2.8 A. I. Dahl, Stability of Base-Metal Thermocouples in Air from 800 to 2
21、200 F. J. Research Nat.Bur. Standards 24, 205 (1940).2.9 International Practical Temperature Scale of 1968, National Bureau of Standards.6$(,17(51$7,21$/ $,5% 3DJHRI_2.10 National Bureau of Standards Monograph 125.2.11 “Manual on the Use of Thermocouples in Temperature Measurement“ ASTM Special Tech
22、nicalPublication 470B, July 1981.2.12 “Temperature, Its Measurement and Control in Science and Industry“ ISA, Vol. 3 (1962) andVol. 4 (1972).3. PROCURING AND CALIBRATING TYPE K (CHROMEL ALUMEL) THERMOELEMENTS:3.1 Specification of Thermocouple Wire:A reasonable thermoelectric specification for Type K
23、 thermocouple wire to be used in makingengine thermocouples is as follows: The temperature-emf relation for thermocouples made fromthe wire shall conform to that given in Tables 7.3.2 and A7.1.1 of National Bureau of StandardsMonograph 125.3.2 Methods of Making Thermocouple Junctions:The methods for
24、 forming the measuring junction are as follows: autogenous welding; joining thethermoelements by means of a third material such as solder, brazing spelter or welding rod; andthe use of a third element through which the thermocouple circuit is completed, as when thethermoelements are individually pee
25、ned or welded to a solid conductor or dipped into a moltenone.The most widely used method is autogenous welding, in which the thermoelements are fusedtogether by a torch or by electrical means without using any other metal to form the joint. Thenoble metals may be welded with a torch in the absence
26、of flux. For base metals, it isadvantageous to use a flux to minimize oxidation. Care should always be exercised to limit theheating to the very ends of the thermoelements and to avoid heating them too long or too hot. Oxy-acetylene, oxy-gas or air-gas flames with small, sharply defined inner cones
27、are suitable formaking thermocouple junctions. Borax is often used as a flux, and after the weld has been made,all traces of flux should be removed. Immersion in boiling water for several minutes is effectivefor most fluxes.Three methods of electric welding are in common use. An arc between two carb
28、on electrodesmay be used to fuse the ends of the thermoelements together, or the thermoelements may beused as one of the electrodes with a carbon rod as the second. Still another method utilizesmercury covered with oil as one electrode and the thermoelements as the other, so that the latterare arc w
29、elded together when brought into contact with the mercury. In all electrical methods,current and voltage must be adjusted to meet the specific needs. Resistance welding is welladapted to making butt-welded junctions. The ends of the thermoelements are held in firmcontact and a surge of current is pa
30、ssed to effect fusion.6$(,17(51$7,21$/ $,5% 3DJHRI_3.2 (Continued):In measuring the surface temperature of a piece of metal, it is often advantageous to attach thethermoelements separately to the metal. This may be done by spot welding or peening the wiresto the metal whose temperature is to be meas
31、ured. To avoid errors, the points of attachmentshould be close enough together that there is no significant difference in temperature betweenthem.Twisting the thermoelements at the measuring junction is not recommended, despite the fact thatsuch a junction may have greater mechanical strength than i
32、ts untwisted counterpart. Twistingleaves the exact location of the first electrical contact between the thermoelements in doubt; itleaves strains in the wires which are particularly difficult if not impossible to remove from theAlumel; and a twisted junction responds less rapidly than its untwisted
33、counterpart to suddenchanges in temperature.3.3 Calibration of Thermocouple Wire:All calibrations must be related directly or indirectly to the International Practical TemperatureScale (see 11.6). In general, the fixed points defined therein are used only for primarycalibrations and the devices so c
34、alibrated are then used for calibrating others for practicalapplication. More specifically, type K thermocouples are normally calibrated by direct comparisonwith a standard platinum-platinum rhodium couple. As a result of experience and commonpractice over the years, reference values of emf as a fun
35、ction of temperature have come intogeneral usage. National Bureau of Standards Monograph 125 Reference Tables forThermocouples contain identical reference tables for several combinations of thermoelements,shown in Table 1.TABLE 1 - Thermocouple MaterialsTypeB Platinum - 30% rhodium versus platinum -
36、 6% rhodiumE Nickel-chromium alloy versus copper-nickel alloyJ Iron versus a copper-nickel alloyK Nickel-chromium alloy versus nickel-aluminum alloyR Platinum - 13% rhodium versus platinumS Platinum - 10% rhodium versus platinumT Copper versus copper-nickel alloyFor any given combination of thermoel
37、ements, the reference table serves as the goal towardwhich performance is directed, and from which unavoidable differences are determined andapplied as corrections. Manufacturers and vendors of type K thermoelements need to know thethermoelectric characteristics of each alloy individually, in order
38、that batches of wire may bematched to perform as nearly in accordance with the reference table as the purchaser mayspecify. This sort of calibration is usually done by determining the emf of individual samplesagainst pure platinum. One alloy being positive and the other negative to Pt, the sum of em
39、fsagainst Pt is identical with the emf of the type K couple.6$(,17(51$7,21$/ $,5% 3DJHRI_3.3 (Continued):Those who use type K in making complete thermocouples, and those who use the finishedthermocouples are interested only in the emf of the particular samples of type K employed. Inother words, the
40、calibration which the users need is not of the individual thermoelements againstPt, but rather of the type K directly.A calibration of the latter type is made at NBS by welding the junction of a standard Pt-PtRhcouple to that of the Chromel-Alumel couple being tested; by placing this common junction
41、 nearthe center of a Chromel tube furnace which is heated by passing current through the furnace tubeitself; and by taking simultaneous readings of the emfs of the two couples over the range oftemperature which is of interest. Both are long enough that their reference junctions may be keptat the ice
42、 point. Immersion is sufficient to insure against loss of heat by conduction from thejunction. By using the comparison method, problems of heat transfer to and from the junction byradiation and convection do not arise.It is normally convenient to express and use the results of such a calibration in
43、the form of a tableor chart of differences from the reference curve. While samples of the thermoelements currentlyused in the manufacture of gas turbine thermocouples are calibrated in this or similar fashionoriginally, and are spot checked at one or more stages in the manufacture of the couples,dev
44、iations of the actual production thermocouples from the standard curve are not measured andhence no corrections can be applied for such differences as may exist from the reference curvefor Chromel-Alumel. Assuming that the calibration of the type K thermocouple is no farther fromthe reference curve
45、than the thermoelements were originally, then the maximum errors that canarise from not applying corrections are therefore given by the tolerances in the specificationsunder which the thermoelements were purchased originally; namely 2.2 C (4 F) up to 277 C(530 F) and 0.42% (3/4%) of the temperature
46、in C (F) above 277 C (530 F).The calibration of thermocouple wire by direct comparison with a standard Pt-PtRh thermocouplein a tube furnace, other methods for calibrating thermocouples and testing thermocouplematerials, and the precautions which must be observed in order to attain various degrees o
47、faccuracy are given in detail in a paper by Roeser and Wensel (2.1). Such calibrations areconducted at the National Bureau of Standards for the public on a fee basis.4. CIRCUITRY FOR HIGHEST ACCURACY:4.1 Lead Wires and Terminals:Where highest accuracy is sought, it is preferable that the thermoelements extend without abreak from t
