RAF Radar Home

Radar Type Numbers

Chain Home Radar System

Chain Home Low

Chain Home Extra Low

Ground Controlled Intercept

Post War Planning

Rotor Radar System

Master Radar Station

Linesman Radar System







"Cold War" and Radar links

Contact the Editor

FAQ (Frequently Asked Questions)



The rigid spine which is the positional reference for the antenna modules is lowered hydraulically to the horizontal position during transit. The spine also forms the housing for miscellaneous electronic equipment including frequency synthesizer, beam forming networks and secondary radar system where fitted, the upper part of the assembled planar array being designed to support an SSR antenna if required.


Height finding by radar is based upon the measurement of the angle of elevation of the target: thus 3D radars must ascribe an angle to every echo received. CH, because of its metric wavelengths, naturally produced a pattern of vertical lobes from the interference caused by ground reflections; therefore, perhaps curiously, the very first long range sensor had the desirable 3D capability! But, again because of its long wavelength, it was unable to detect targets at low angles of elevation.

By comparison, the radiation from microwave radars is largely independent of ground effects and can be made to produce a more complete vertical coverage pattern: by various means, this pattern can have a beam structure enabling responses from different elevation angles to be identified. It is at this point that the main design options and problems of 3D microwave radars arise, and over the years different solutions have been adopted. Many have been variants of the general concept of a number of feed horns looking into a common reflector to produce a structure of stacked beams; there is the further option of transmitting and receiving on each beam separately, or transmitting over the full elevation coverage but receiving on individual beams, as is done in Martello.


The late 60's and early 70's constituted a period of intensive air defence study in the UK, NATO and abroad. Many views were put forward but all seemed to agree that future long range sensors should

    (a) be inherently 3D,
    (b) have the best possible ECM resistance and
    (c) be, as far as possible, mobile or reasonably transportable.

This was a clear indication that the days of the '2D + heightfinder' and of the heavy permanently installed 3D radar were numbered.

It was against this background that Martello was conceived in the mid-70's as a design to meet the requirements of countries requiring a long range sensor for air defence. From the start, it was defined as a transportable 3D radar with low antenna side lobes since, whatever other ECCM features are added, this is an essential feature in resistance to jamming.

A planar array would have to be used since there is a limit to the side-lobe level achievable with conventional feeds and reflectors, many arrangements of which had been exploited by the Company over the years. Typical 3D horn-stack radars have 'first side lobes' in azimuth in the range -20 to -25 dB depending on frequency, but Martello was designed to be around -30 dB over the entire operating band, with very low 'far-out' side lobes.


Conventional surveillance radars sweep out a volume of space by rotating a beam, or set of elevation beams, through 3600. This is normally achieved by mechanically rotating the entire antenna. The possibilities of using static phased planar arrays facing in different directions, covering all azimuths by electronic control were not overlooked but it was thought that complexity and expense would be excessive, as indeed has been the experience elsewhere, including the USA.

Thus it was concluded that the solution for Martello still lay in the mechanically rotated planar array which would give the required performance at minimal cost but it remained to decide on the form that the height finding system should take. Many approaches were possible: some employ a change of operating frequency or phase in order to elevate the beam or to form a vertical beam structure, whilst others use discrete frequencies for each beam. All such methods increase the difficulty of achieving frequency agility - another highly desirable characteristic for the avoidance of jamming.

Martello avoids any commitment of the operating frequency for the process of elevation beam forming by employing a unique passive beam forming network (b.f.n.) operating at the second intermediate frequency of the receiving system. Beams are synthesized and the receiving system includes separate signal processing channels for each beam. The radar output thus consists of parallel data from all the elevation receiving beams, height being assessed on every return by a process of monopulse extraction, using the ratio of signal powers in adjacent beams(1,2).

Consider a vertical stack of equally spaced horizontal receiving dipoles, as viewed in figure 5. If the outputs of all the dipoles are combined so that the relative phases of the signals received by each are preserved, the direction of maximum reception is broadside to the stack, figure 5a.

If, however, a phase shift is introduced into the output of every dipole so that each is equally and progressively displaced in phase from its neighbour, signals from a common source will only add in-phase when they arrive at an angle to the array. The phase shifts now compensate exactly for the natural phase relationship of the signals arriving at the separate dipoles, figure 5b.

By introducing suitably controlled amounts of phase shift in each line a beam can be synthesized at any required elevation angle, although the beam shape becomes less well defined at the extremes. In practice, small positive and negative angles are sufficient to provide the total elevation coverage if the planar array is tilted slightly backwards.


Previous page

Top of page

Next Page

Updated 06/11/2001

Constructed by Dick Barrett

Email: editor@ban_spam_radarpages.co.uk

(To e-mail me remove "ban_spam_" from my address)

ęCopyright 2000 - 2002 Dick Barrett

The right of Dick Barrett to be identified as author of this work has been asserted by him in accordance with the Copyright, Designs and Patents Act 1988.