FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1962
1962 - 0947.PDF
LIGHT International, 14 June 1962 Letters The Editor of" Flight International" is not necessarily in agreement with the views expressed by correspondents in these columns. Names and addresses of writers, not for publication in detail, must in all cases accompany letters. Flight-test Instrumentation SIR,—I feel that I must comment on your report (May 24) of the recent Instrumentation Symposium at Cranfield. Quite understandably, you concentrate on the more imaginative fields such as digital recording, flight computers, etc. I must say, however, that your two sentences on the Bristol Type 188 hardly do it justice, furthermore, the second sentence is somewhat misleading. As I described in my paper, the Type 188 instrumentation includes both magnetic tape recording and radio telemetry. Regarding the use of galvanometer recorders in the aircraft, this stems from three main considerations:— (i) The non-availability of packaged tape recording equipment of suitable size at the time that the Type 188 installation was designed. (ii) Notwithstanding Mr Mathison's comparative table of tape recording systems, no convincing evidence was available at the time of the Type 188 installation design to show that the required accuracy of ± 1 per cent (not to be confused with resolution) could be obtained for continuous recording without the use of frequently injected calibration levels. (iii) There is a considerable amount of capital tied up in semi automatic trace-reading and data-processing equipment both in the Royal Aircraft Establishment and in the aircraft industry. This could not be discarded unless the replacement system offered a considerable and proven advantage. I would also like to point out that the time-multiplexing technique employed in the tape and- telemetry systems of the aircraft were designed to permit the introduction of digital equipment, such as the Fresco A-D converter, which utilizes data compression, or the Vosco A-D converter with pure binary output. Both these converters are produced by Bristol Aircraft Ltd. May T suggest that Bristol Aircraft Ltd in general and the Type 188 aircraft instrumentation in particular are not so far behind the times as your brief mention suggests. Bristol L. G. MAY Head of Instrumentation Laboratories, Bristol Aircraft Accident Recorders SIR,—At a time when regulations are being introduced to require the fitting of accident data recording equipment to all civil and military aircraft, and numbers of people on the operational side are being brought into contact for the first time with problems of the relatively new art of flight record ing, it is extremely important to avoid unnecessary confusion of what is a fairly complex picture. Such confusion can arise from the publication of descrip tions of recording equipment which are unbalanced and where only a few of the important factors are presented. On pages 842-843 of your issue of May 24, an accident data recorder employing magnetizable wire as a medium is briefly described. The purpose of an accident data recorder is to preserve after an accident a precise, adequate and reliable record of what took place. Therefore the factors of equal and simul taneous importance are the preservation of the record, the reliability of the recorder, a high standard of accuracy and an adequate number of channels. The accuracy requirement is difficult to achieve using the latest types of plastic magnetic tape, which has superseded wire for the far less stringent requirement of speech recording. No recording method is ideal for an accident data recorder, and all have strengths and weaknesses. Your brief descrip tion concentrates only upon the advantages of the use of wire, and to balance the picture I would like to conduct a short comparison. 945 Firstly, the magnetized steel wire will tolerate a temperature of 550°C and a high-temperature plastic-based tape only 150°C. However, the international standards for fireproofing of the recording medium require exposure to temperatures of 1,100CC and therefore both wire and plastic tape require that the recorder should be thermally insulated. Wire thus requires a thin layer of insulation and tape a thicker (but far from prohibitive) layer. Both media resist sea water equally well and there therefore remains the question of the effect of mechanical damage due to impact. Recording wire is hard, brittle and springy and is "scramble" wound on a spool. Such recorders have been used for a number of years during test flying, as a pilot's notebook to record speech. Several have been in aircraft which have crashed and in many instances the brittle wire has been broken, has sprung off the spool and has become a tangled mass from which it has been difficult to recover even intelligible speech. If magnetic wire is broken, it is extremely difficult to repair and there is inevitably a loss of several seconds of record at the point of the break. As the most vulnerable part of the recording will be that which has just left the recording head, there is an increased likelihood that the lost information will coincide with the most critical period. Plastic-based tape, on the other hand, is inert and flexible and there is negligible damage to the record even if the tape spool is very seriously distorted. If it is broken, it is repaired by ordinary tape-splicing techniques with the loss of less than a tenth of a second of record. Under the circumstances, we feel that magnetic tape as opposed to wire will offer better chances of surviving a crash, provided there is an appropriate thickness of heat insulation. The next comparison is on the score of reliability. It is stated that an amplitude-modulated multiplexed system provides a simple recorder with a correspondingly high standard of reliability in operation. This is perfectly true if the recorder alone is considered, but it is the reliability of the complete system from the original operating or engineering parameter to the recording medium that must be taken into account, and the sum of the total system complexity must include the transducers. Accepted and well tried systems of recording data on magnetic tape require negligible power from the signal source, so that use can be made of many of the existing transducers in the aircraft. The wire recorder des cribed requires special transducers for each input, so the total system complexity in the latter case is likely to be higher and no lower, with the same analogy concerning reliability. The final factors are accuracy and channel capacity. The accuracy of the wire recorder is stated as 6 per cent SMD if a single transducer is required or H per cent with a four- sweep transducer. The technique of the use of a multi-sweep transducer to improve accuracy where the recording system itself is the weakest link in the chain is well proved, but what is not always obvious is that a 50-channel recorder with a low accuracy is only a ten-channel recorder at the increased accuracy, as each section of the sweep requires an individual recording channel plus one other channel to show in which segment the information lies. It is also very unusual to see in this context the expression "SMD," standing for "statistical mean deviation." This is a well established statistical device for reading a reasonably predictable trend through a large amount of error of random amplitude and rate. It is reasonably reliable if there is a predictable pattern, or if the mean component frequencies of the error are very much higher than the maximum expected rate of change of the parameter in question. By its very nature, an accident involves abnormal behaviour and, very probably, abnormally high rates of change. I do not see, therefore, that for this purpose the use of statistical mean deviation as a device for the improvement of the apparent error can be applied with acceptable reliability. It should be noted that all international specifications, be they for civil or military aircraft, are in terms of absolute error, as opposed to average error, which predicates the error at any instant, and this in itself eliminates any averaging techniques of variations of level whose frequency is not high compared with a maximum possible rate of change of the information signal. Byfleet, Surrey ROYSTON INSTRUMENTS LTD K. G. Dobson, Managing Director
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
