At present EMC clearance of an aircraft is performed by experimental testing alone: there is no alternative source of information concerning the finished platform's susceptibility characteristics. This cannot be said to be an entirely satisfactory state of affairs for several reasons. Also, the severity of EMC problems is likely to get worse with the increasing use of imperfectly conducting materials for aircraft construction. It would therefore be of considerable practical benefit if some form of trustworthy, independent supplementary procedure existed for determining susceptibility characteristics. This overall aim of the work described here is to assess the usefulness of computational techniques in this regard: in particular, the prediction of currents induced upon cables located on aircraft-like structures. The configuration studied is based on the low-level swept current (LLSC) test used in aircraft clearance

The increasing use of carbon fibre composite (CFC) material for aircraft skins raises inter-related concerns regarding the operation and effects of airborne antennas operating at HF (2-30 MHz). Establishing the credibility of computed results is particularly important in a traditionally experimental subject such as EMC. To this end it is therefore useful to consider a test problem in which modelling CFC material is the dominant feature. In formulating such a problem it is, however, necessary to avoid certain complicating factors (particularly at HF) associated with the numerical methods available. In this article a particular problem was formulated so as to circumvent as many of these complications as possible, and to permit additional corroboration of the computed results both by straightforward experimental measurement and by analytical calculation. A problem which satisfies these requirements is that of determining the magnetic field screening of an open-ended CFC cylinder.

At radio frequencies, electromagnetic compatibility (EMC) clearance of aircraft by practical measurement is expensive and becomes a more lengthy process as further electrical and electronic equipment is installed. A means of producing a reduced set of experimental tests is therefore attractive. Complementing such testing by numerical modelling is potentially a useful way of effecting this provided that sufficient confidence is first established and an indication of the likely magnitude of any differences between calculation and measurement is known and the causes understood. The establishment of confidence in numerical techniques would also be valuable at the design stage by virtue of the physical insight which can be generated by numerical experiments. This use in support of synthesizing a new design is important because restrospective treatment of EMC problems has always been expensive. The article outlines some comparisons between calculation and measurement made in the initial stages of a programme aimed at assessing the applicability of numerical modelling as a supplement to measurement for aircraft EMC clearance purposes

This paper examines the consequences for calculations of near-field (antenna reactance) and far-field (antenna resistance and radiation patterns) dependent parameters of an installed `long-wire' antenna of removing the wings from a mathematical model of a medium-sized passenger aircraft. In the context of calculating link budgets, it is considered that the absence of the wings is not likely to perturb these properties to an extent which is of great practical importance and that neglect of this part of the airframe may be a useful way of making the solution of problems involving larger aircraft with a similar antenna installation both tractable and affordable

A moment-method calculation is described of the maximum mutual coupling (or minimum isolation) between two loop antennas mounted on the tail-cone of a helicopter at the low-frequency end of the aircraft HF band. The wire-mesh mathematical model of the aircraft is constructed with a high sampling density in an attempt to produce estimates of the required near-field dependent parameters with sufficient precision to calculate the maximum mutual coupling with acceptable accuracy. The calculation is a fairly demanding test of the procedure adopted for the construction of the mathematical model, because it depends upon the accuracy with which both near-field and far-field parameters are determined. This mathematical model does not incorporate any empirically derived information, and the calculation was of necessity performed in advance of an aircraft being available for experimental purposes. The difficulty of assessing the credibility of calculated results under such circumstances is considered. NEC was used as the solution code for the wire-mesh model. An independent corroboration of the calculation was made possible by the siting and configuration of the proposed antennas. This allowed a simple equivalent circuit model of the arrangement to be constructed, values of the circuit elements to be estimated, and hence an independent approximate determination of the maximum coupling to be made. The degree of consistency between the results obtained by these two methods is considered to enhance the confidence which can be placed in the results obtained using moment-method mathematical models of this type when applied to near-field aircraft HF antenna problems.

Considers an example of a moment-method modelling procedure which has been applied previously to other generic aircraft and antenna types. The overall aim of these studies is to provide a `recipe' for the prediction of airborne HF antenna system characteristics; the application of which will produce estimates of near as well as far-field dependent parameters which are credible and sufficiently accurate for engineering purposes. The procedure simulates the airframe as an electrically very fine wire-mesh (of mesh side length of the order of 1500 wavelength at 2 MHz) and uses NEC as the solution code. An important aspect of the studies described is that no empirically derived information of any kind is incorporated into the mathematical models constructed. Thus, true predictions are produced

A mathematical modelling procedure for the prediction of near-field parameters for aircraft HF antennas is applied to the example of a variable geometry airframe. The model is constructed in such a way that it can be reconfigured as a real aircraft of this type might change its geometry and achieves this without the need to alter the discretization. This engenders some uncertainty concerning the quality of the numerical solutions obtained with the model. This is because the moveable surfaces clear the rest of the airframe by electrically very small distances. A further potential difficulty is that the excitation of the model is via a notch antenna; the area of which is of the same order as that of the wire mesh simulating the airframe. No empirically derived information is used in the construction of the model and NEC is used as the solution code. Calculations of the antenna terminal impedance throughout the band are made using this model. An assessment of the credibility of these calculations is attempted by independent theoretical means using a simple equivalent circuit model, and by semi-quantitative physical arguments.

Moment methods are necessarily used for calculation of the characteristics of HF antennas mounted upon aircraft. Whilst it is possible to obtain reasonable estimates of parameters which depend upon the far-field using a relatively coarse discretization of the airframe, this will not generally suffice for calculation of those which are determined by the near-field. Finer discretization is necessary if the latter parameters are to be calculated with acceptable accuracy. The author attempts to give an indication of discretization judged to be adequate in this regard by examining the prediction of some near-field dependent parameters for three generic aircraft types: a helicopter, a fixed-wing aircraft and a variable geometry aircraft. In the examples described, no empirically derived information has been incorporated into the models. Thus the results represent true predictions rather than `postdictions'

Two means of driving mathematical models are considered which appear to offer greater realism than the simple pulse source. These are the magnetic current annular ring (magnetic frill) source and the current slope discontinuity source. It is shown that the former is conveniently considered as a toroidal transformer. This approach enables the driving electric field to be derived by inspection from the magnetic vector potential produced by the magnetizing current flowing in the transformer primary. It is indicated that, for practical computation, this source is equivalent to the pulse source. The current slope discontinuity source is portrayed as a portion of a biconical capacitor. Examination of both sources leads to the conclusion that for point matching calculations, neither is free of difficulty: implementation necessitates an integration process which tends to remove the configurational details and hence the benefit of realism. For this type of calculation, neither appears to have any clear advantage over the pulse source

An HF loop antenna installation on a helicopter is described. The azimuthal radiation patterns at 10 frequencies throughout the HF band are obtained by three methods: mathematical modelling, measurements with a physical scale model, and measurements using the real aircraft in flight. The predictions from the mathematical model were made in advance of any measured results being obtained and no empirically derived information was incorporated into the model. The salient problems connected with each technique are discussed. The results obtained in the study are presented and compared. In the context of HF, the level of agreement between the three methods is considered to be good. The predictions of the mathematical model exhibit a fair degree of consistency with the results of the other two methods

The author gives an indication of the accuracy attainable with RF numerical methods in the prediction of those quantities which depend upon the near-field of radiating structures. A wire-mesh RF mathematical model of a medium size airliner is described. Using this model, and with NEC as the solution code, predictions are presented of the terminal reactance of an installed `long-wire' HF antenna. These represent true predictions rather than `post-dictions' because the RF mathematical model does not incorporate any empirically derived information. The difficulties involved in obtaining the corroborative measurements are examined and direct comparison with the predictions is made. The particular antenna configuration chosen for this study is considered to constitute a stringent test of the predictions. The agreement between calculation and subsequent measurement is fair

The author discusses the fundamental limitations of electrically small radiators and their significance to electrically small antennas. Some of the parameters are derived. Systems incorporating electrically small radiators are constrained by fundamental limitations to be either inefficient or narrowband. These conditions are apparent in the case of aircraft HF (2-30 MHz) antenna systems, particularly for small aircraft at the low end of the band. The author discusses the HF antenna problems for aircraft, in particular the need to distinguish between the airframe and the antenna. The types of antenna used, loops and notches, are discussed. Examples of measurements for a BAC 1-11 airliner and a helicopter are given.

It is of practical relevance to be able to calculate antenna properties which depend upon the near-field as well as those which depend upon the far-field. A study performed by the writer, of the estimation of some of these quantities for two helicopters using numerical electromagnetics code (NEC) is outlined. The two helicopters studied were both fitted with loop (ie. magnetic dipole) antennas. The first was fitted with two antennas mounted upon diametrically opposite sides of the tail cone, and the second fitted with one antenna mounted on left-hand side of the rear cabin area. An attempt has been made to compare the results of these computations with experimental measurements (radiation patterns) performed upon both real and scale model aircraft, and with independent calculations of some near-field quantities