1、AEROSPACE INFORMATION REPORTAIR1608REV.AIssued 1989-07Reaffirmed 2008-02Revised 2002-04Estimation of Total Error in AltimetryINTRODUCTIONIn the field of altimetry, and more generally in the field of air data, it has been customary to specify accuracies in the form of tolerances on specific functions
2、. For example, the specification for an altimeter or air data computer might say that for a particular test condition, the instrument would have a tolerance of 100 ft for scale error, 50 ft for hysteresis, 50 ft for friction, 20 ft for temperature, etc. Because scale error is the most obvious, and u
3、sually the largest of the errors, mistaken assumptions have often been made that scale error alone is a sufficient measure of the accuracy of an instrument, or even of a whole system.Some of those in the field have advocated that tolerances should be lumped; that is that test procedures should be de
4、vised so that for a particular test point, the equipment should be tested so that it would be exposed to all of the pertinent sources of error, and a single numerical limit be set on the net of all of them. This concept has been mooted for many years, but has never found full acceptance.There remain
5、s a need for some means to know the width of the total band of error or uncertainty so that it can be stated that a particular aircraft will be within so many feet of being at its assigned altitude, and that there is good confidence (say, 3 or 99.7%) in the statement.The need for this information is
6、 to be able to establish standards of vertical separation between aircraft and standards of equipment performance in order to operate safely with those vertical separation standards.SAE Technical Standards Board Rules provide that: “This report is published by SAE to advance the state of technical a
7、nd engineering sciences. The use of this report is entirely voluntary, and its applicability and suitability for any particular use, including any patent infringement arising therefrom, is the sole responsibility of the user.” SAE reviews each technical report at least every five years at which time
8、 it may be reaffirmed, revised, or cancelled. SAE invites your written comments and suggestions. Copyright 2008 SAE International All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, phot
9、ocopying, recording, or otherwise, without the prior written permission of SAE. TO PLACE A DOCUMENT ORDER: Tel: 877-606-7323 (inside USA and Canada) Tel: 724-776-4970 (outside USA) Fax: 724-776-0790 Email: CustomerServicesae.org SAE WEB ADDRESS: http:/www.sae.org SAE AIR1608 Revision A- 2 -INTRODUCT
10、ION (Continued)In the 1950s and 60s, air carriers flying the North Atlantic were anxious for economic reasons to fly at 1000 ft, rather than 2000 ft separations. Studies were made by the Air Transport Association (ATA), and four of the resulting reports were eventually used as the bases for SAE docu
11、ments (References 11, 13, 14, and 15). The International Civil Aviation Organization (ICAO) has sponsored on-going studies of the problem by its Panel on Vertical Separation of Aircraft, later renamed Review of the General Concept of Separation Panel (RGCSP). In 1982, the Federal Aviation Administra
12、tion (FAA) asked the Radio Technical Commission for Aeronautics (RTCA) to form a Special Committee (SC-150) to develop a Minimum System Performance Standard (MSPS) to prepare for reduction of vertical separation to 1000 ft above Flight Level (FL) 290.This Report suggests methods of estimating total
13、error for four flight regimes, as originally considered in Area Navigation (RNAV) studies. More specifically, it suggests starting with the limits of probable error (i.e., tolerances) of contributing types of error and arriving at overall limits of probable error or uncertainty. It also provides for
14、 starting with known or measured contributing errors and arriving at an overall error, and for the situation where the available information consists of a mixture of tolerances and known errors.The Traffic Alert and Collision Avoidance System (TCAS) program does not yet call for specific efforts to
15、improve the accuracy of altimetry. “A Field Study of Mode C Altimetry Accuracy in the General Aviation Fleet“ (Reference 22) shows that it is presently adequate up to 3900 ft. At high altitudes however, non-servoed pressure altimeters are less accurate than at low, and at high speeds airframe static
16、 pressure (position) errors become more significant. In the investigation of reports of apparent malfunction or false warning of TCAS equipment at high altitudes and/or speeds, where the accuracy of altimetry is suspect, the procedures of this AIR and of RTCA SC-150s “Minimum System Performance Stan
17、dards for Vertical Separation above Flight Level 290“ (Reference 23) should be useful.Altitudes in this report are in terms of feet. Altitudes may some day be stated in meters or other units, but that day is further into the future than the adoption of other SI units. References 17 and 18 contain ba
18、ckgrounds on this question.1. SCOPE:AIR1608 ESTIMATION OF TOTAL ERROR IN ALTIMETRY proposes a method of estimating overall error of altimetry in order to provide a basis for safe vertical separation of aircraft.2. REFERENCES:1. “Terrain Clearance and the Vertical Separation of Aircraft“, Circular 26
19、-AN/23 Second Ed., International Civil Aviation Organization, Montreal, 1956.2. “The Measurement of Pressure Altitude on Aircraft“, Gracey, Technical Note 4127, National Advisory Committee for Aeronautics, Langley Field, October 1957.3. “Panel on Vertical Separation of Aircraft - - Second Meeting“,
20、DOC 7835-AN/863, International Civil Aviation Organization, Montreal, June 1958.SAE AIR1608 Revision A- 3 -2. (Continued):4. “Altimetry“, Special Committee 70, Paper 215-58/DO-88, Radio Technical Committee for Aeronautics, Washington, November 1958. (Reprinted April 1978.)5. “Survey of Altitude Meas
21、uring Methods for the Vertical Separation of Aircraft“, Gracey, Technical Note D-738, National Aeronautics and Space Administration, Langley Field, March 1961.6. “Analysis of the Effect of Altimeter System Accuracy on Collision Probability“, Gracey, Technical Note D-1627, National Aeronautics and Sp
22、ace Adminstration, Langley Station, March 1963.7. “Report on Vertical Separation Study, NAT Region“, DOC GEN/1951, International Air Transport Association, Montreal, March 1964.8. “Survey of the Errors of Pressure Measuring Instruments in Relation to Air Traffic Separation Standards“, Anderson, Tech
23、nical Report 65262, Royal Aircraft Establishment, Farnborough, December 1965. (AD 478915)9. “Specifying the Calibration of Static Pressure Systems for the Safe Use of 1000 Foot Vertical Separation Standard in North Atlantic Jet Traffic“, Reich and Anderson, Technical Report 66156, Royal Aircraft Est
24、ablishment, Farnborough, May 1966.10. “Performance Report on Static Air Source on Air Carrier Turbojet Aircraft“, Air Transport Association of America, Washington, October 1966.11. “Design and Installation of Pitot-Static Systems for Transport Aircraft“, Stratton, Aerospace Recommended Practice 920,
25、 Society of Automotive Engineers, Warrendale, October 1968.12. “Recommendation on Design of Pitot-Static Systems for Transport Aircraft“, Anderson, Recommendation 34-10-3, Air Transport Association of America, Washington, June 1970.13. “Flight Test Procedures for Static Pressure Systems Installed on
26、 Subsonic Transport Aircraft“, Stratton, Aerospace Recommended Practice 921, Society of Automotive Engineers, Warrendale, July 1971.14. “Maintenance of Pitot-Static Systems of Transport Aircraft“, Anderson, Aerospace Information Report 975, Society of Automotive Engineers, Warrendale, July 1972.15.
27、“Barometry for Altimeter Calibration“, Anderson, Aerospace Information Report 1075, Society of Automotive Engineers, Warrendale, March 1974.16. “Implications of Altimetry System Errors for Collision Avoidance Systems“, Mundra, Technical Report MTR-7232, Mitre Corporation, McLean, May 1977.17. “Probl
28、ems in Worldwide Standardization of the Units of Height Measurement“, Gilsinn, Report FAA-EM-78-2, Federal Aviation Adminstration, Washington, February 1978. (AD A051150)18. “SI Units of Measurement in Aviation (Together with Non-Linear Units of Altitude)“, Anderson, Engineering Report F-1856, Unite
29、d Airlines, San Francisco, September 1978. (No text; a collection of references.)SAE AIR1608 Revision A- 4 -2. (Continued):19. “Estimation of Total Error in Altimetry“, Anderson, Engineering Report F-1870, United Airlines, San Francisco, March 1979. (Draft of Society of Automotive Engineers Aerospac
30、e Information Report 1608.)20. “Measurement of Aircraft Speed and Altitude“, Gracey, Reference Publication 1046, National Aeronautics and Space Administration, Hampton, May 1980. Also published by John Wiley, New York, May 1982.21. “Height Indication by Pressure Altimeters“, Anderson, Engineering Re
31、port F-1588, United Airlines, San Francisco, July 1980.22. “A Field Study of Mode C Altimetry Accuracy in the General Aviation Fleet“, Cohen, Report MTR-86W231, Mitre Corporation, McLean, March 1987.23. “Minimum System Performance Standards for Vertical Separation Above Flight Level 290“, Special Co
32、mmittee 150, Radio Technical Commission for Aeronautics, Washington, Fifth Draft, January 1988. More drafts expected before publication as an RTCA DO- _document.3. GROUPING OF ERRORS:Cursory discussions of accuracy of altimetry are usually concerned with the errors peculiar to one aircraft and its e
33、quipment, or in the case of collision probability studies, two aircraft. While this is not invalid, neither is it complete. There are additional sources of error that are common to all aircraft in a locality, others that are common to a type of aircraft, and still others related to the manner of ope
34、ration of the aircraft.3.1 Errors Common to all Aircraft in a Given Locality:Aircraft below 18,000 ft and receiving their QNH altimeter setting number from a single source are all equally dependent on the accuracy of that source. If there are errors in the QNH transmitted, all aircraft using it will
35、 be flying too low or too high by the same amount. This may increase the risk of long or short landings or of hitting obstructions on the ground, but should not increase the risk of collision between two aircraft.3.2 Errors Pertaining to Aircraft of a Type:One of the steps on certification of an air
36、craft type is flight calibration of the static system of one or more of the first few aircraft. It has been recommended that at least three aircraft be calibrated (Reference 12). It is usual that the flight test results are a series of data points with some scatter, and that a fair curve, or family
37、of fair curves is drawn through the scatter field. The fair curve then is published in the government-approved Flight Manual as applicable to all aircraft of that type. Subsequent aircraft of the type will thus be subject to whatever errors or uncertainties there were in the original calibration.SAE
38、 AIR1608 Revision A- 5 -3.2 (Continued):Where the airframe builder has not been able to find a static source of negligible error, he should provide the aircraft with an automatic correction system, such as might be included in a central air data computer. There are no rules defining how large an err
39、or is negligible. One airframe builder might consider up to 75 ft of altitude error negligible, while another might choose to neglect 200 ft.Until automatic altitude reporting for traffic control became a requirement, it was feasible to post near the altimeter a correction card. Altitude reporting r
40、egulations, however, require that the altitude reported digitally via the Air Traffic Controller Transponder agree (on a 1013.2 mb/29.92 in Hg basis) with the altitude shown by the altimeter within 125 ft. In effect, this precludes the use of correction cards and requires either that there be an aut
41、omatic correction system or that the error be asserted to be negligible.3.3 Errors Pertaining to an Individual Airframe:In an aircraft with small flush static pressure port fittings set in the skin, there may be minor variations of skin contour near the ports as compared with the aircraft used in th
42、e original flight calibration and type certification. These may be due to manufacturing variations and/or to subsequent damage and could affect calibration.Where Pitot-static probes or hard flush static port plates (Reference 12) are used instead of small flush ports in the skin, surface condition i
43、s more readily controllable, although probes are exposed to surface deterioration, damage, and misalignment which could affect calibration.3.4 Errors Pertaining to an Individual Aircraft Set of Equipment:Any set of equipment, whether a pressure-operated altimeter or an air data computer driving an e
44、lectric altitude indicator, will be subject to scale, temperature, friction and other errors. In computers which apply a static pressure correction, errors of Mach will affect accuracy of altitude output, particularly at high speed. Where the atmospheric pressure is measured in a computer and the al
45、titude is transmitted electrically to an indicator, there may be error in the servo loop; in the absence of gross malfunction, this should not exceed 5 ft and is disregarded in this report.Aircraft equipment will also be subject to the errors of the shop equipment used to calibrate it. In this repor
46、t only the accuracy of the barometer or other absolute pressure standard is considered.3.5 Flight Technical Error:Errors of pilot interpretation of vertical guidance instrumentation, pilot operation of aircraft vertical controls and deviations caused by aircraft response characteristics are often lu
47、mped together as flight technical error. For the purposes of this report, reading error and altitude hold error are considered separately.SAE AIR1608 Revision A- 6 -3.5.1 The least increment of most altimeter readouts is 20 ft. In level flight, reading errors should not exceed this. (Some altimeter
48、displays are said to be susceptible to misreading by 1000 or 10,000 ft, but that problem is considered to be outside the scope of this report.) In ascent and descent the errors are likely to be larger than 20 ft, and lagging. Thus, reading errors would be treated as rectangularly distributed in leve
49、l flight, but as plus offset in descent.The plus offset could be aggravated in the situation where approach and landing are done on a QNH basis. Here the pilot must repeatedly read the altitude above sealevel and mentally subtract from it the field elevation in order to get the height of the airplane above the field. The errors and lags of the subtraction process can be eliminated by operating the altimeter on a QFE basis. Definitions and derivations of QNH and QFE are given in Reference 1. When QFE in mb or in Hg is set on the baro scale of the altime
