Taylor

Home

 

 

Introduction

 

This previously unpublished account was written by G N S (Gerry) Taylor at the solicitation of Sir Robert Telford, as one of several draft contributions by radar staff solicited in 1992 for a new Marconi History envisaged for the centenary celebrations in 1997.

 

MARCONI 50 CENTIMETRE RADARS FOR AIR TRAFFIC CONTROL

 

In his “A History of The Marconi Company” Bill Baker describes how the Services Equipment Division (SED) decided to enter the air traffic control market via the unlikely route of marketing radars operating in the 50 cm band while the rest of the world were, at that time, using, or thinking of using, the 3 cm and 10 cm wavelengths.

 

This, in a nutshell, sums up the birth of the Company's apparently eccentric entry into the ATC radar market of which the following is a somewhat vague and wordy summary. I have no idea whether it will be useful or not: but here goes anyway!

 

In 1950 SED had their first discussions with the then Ministry of Civil Aviation (MCA later to be called MTCA and now known as the CAA). The most significant outcome to SED was the admission by MCA that they were not at all sure how radar could best be used in air traffic control. An Experimental Unit (ATCEU) had been set up and the Air Ministry had presented MCA with an American MEW radar (similar to two British Type 14 radars mounted back-to-back) but for a number of reasons it was of little use in improving on the pre-war procedural method of ATC.

 

It was realised within SED that if it wanted to be in the ATC market for radars it was essential to become familiar with the science of air traffic control and to understand its problems. The only way to do this was to keep in close touch with MCA and particularly with the ATCEU. Dr. Eric Eastwood (as he then was) had come to the same conclusion and he and Chris Colchester spent a good deal of time looking at ATC problems. Chris Colchester even wrote a book on the subject of ATC. (Editors note – listed in the References)

 

MCA bought a Marconi 10cm Type S14 search radar and an associated Type S13 height-finder and these were installed at Heathrow in 1952. It was quickly established that radar height-finders are not accurate enough in comparison with airborne altimeters. As Bill Baker says “aircraft under ATC are cooperating with the ground controllers so the latter can always ask the pilot what his altitude is".

 

It was also in 1952 that, quite by chance, the Air Ministry offered MCA a wartime Type 11 50cm radar which had been modified to include a prototype Moving Target Indicator (MTI) system. The development work to devise the modifications formed part of the MOS contract which Dr. Eastwood (as he then was) brought with him from the Nelson Labs at Stafford to Baddow in 1948. The rest of the contract included the manufacture of 24 kits to modify wartime T11 Mk4s and turn them into Mk7s.

 

After the MCA had used the Type 11 for a few months they invited the Company to send a team to Heathrow to see it in operation. The MCA were very enthusiastic about the performance of the MTI system which completely removed all ground clutter (unwanted reflections from fixed ground objects such as buildings, hills etc.). This meant, for example, that inbound aircraft could be seen from the pick-up range right down to touchdown on the runway and outbound aircraft could be seen as soon as they were airborne. With the MEW radar, ground clutter obscured aircraft until they were more than 20 miles from the radar. This meant that each outbound aircraft needed to be identified when it emerged from the clutter and this involved the pilot being asked to make a turn while the controller watched his display screen to see which of several aircraft was turning.

 

Another advantage of the Type 11 was that, due to the relatively long wavelength on which it operated, weather clutter (echoes from rain, hail or snow) was only rarely seen. On the other hand microwave radars often lost aircraft in a mass of precipitation clutter on the display. What was a surprise to us was the controllers' enthusiasm for the Type 11 in spite of the much higher resolution of 10cm radars. A clutter-free display was their first priority in 1952 and remained so for many years.

 

The MCA did not like a number of engineering aspects of the Type 11 such as the high level of antenna sidelobes which caused problems when a large aircraft was about 5 miles from the radar. The poor state of the equipment generally was another niggle. This was hardly surprising as it was 10 years old and, being a mobile equipment, seemed to have been stored out of doors for most of that time. The nub of our discussion arose when MCA asked if there was any possibility of the Company turning the basic system design into a modern ATC radar with a new antenna and turning gear. To cut a long story short, this was the birth of Marconi's first production ATC radar, the S232.

 

At the same time as the Type 11 was on trial at Heathrow the Ministry of 8upply amended the Contract by reducing the quantity from 24 sets to 10. As it happened the Works had completed all 24 kits so there were 14 left over. Each kit comprised a new transmitter design, a water delay line and about 6 units designed to fit inside the RAF's large, standard Display Unit Type 16. Negotiations took place with the MOS for the Company to buy the 14 redundant kits.

 

At this time the big Vast and Rotor contract which the Company was given by the MOS was gathering pace and absorbing more and more development staff. Fortunately Dr. Eastwood was very anxious to help as much as possible and was fully aware of the importance of getting the S232 off the ground quickly to meet a market which was just opening up. He not only found design engineers but got Scanners Ltd. (later Marconi's Gateshead Works) to redesign their Pivot Mount which was used to support and rotate radar antennas, to make it work upside down.

 

By 1953 the prototype S232 was completed at Rivenhall and the first of hundreds of demonstrations was given to MCA. They immediately placed an order for one for Heathrow and promised an order for another for Gatwick provided that the S232 had adequate horizontal resolution to allow an aircraft in an emergency to be “talked down” to the runway threshold. Trials were carried out at Rivenhall by simulating the Gatwick runway and controlling an aircraft down an imaginary approach path. The results were within the accuracy required for emergency situations such as an ILS failure. In due time we got the order for an S232 at Gatwick.

 

The MCA's equipment for Heathrow, which was the first production model, was installed in 1954. Although the S232 antenna needed to be located close to ground level the water-table at Heathrow was so close to the surface that the normal underground building with its roof at ground level was not practical. A surface building was therefore used. Flight trials showed that this did not provide satisfactory coverage in the vertical plane. The antenna was therefore swapped with the prototype one at Rivenhall which was of a different shape, and this solved the problem.

 

Further sales of about 10 S232s were made in the next four years in both the UK and overseas. MCA bought a third one for Elmdon Airport at Birmingham and the RAE bought two for Farnborough and Bedford airfields.

 

Meanwhile two development projects were put in hand. The first was a new and larger antenna which would reduce the horizontal beamwidth from 4 degrees to 2.1 degrees. This was in the form of a parabolic cylinder, 52.5ft (16m) long and about 12ft (3.7m) high with an offset slotted waveguide radiator. This arrangement was chosen because it was easier to make and allowed the designer more scope in controlling the horizontal radiation pattern. A new turning gear fitted with dual drive motors was designed. When using the same transmitter/receiver as the S232 the radar was known as the S264 and was intended for use at the larger airports for approach control.

 

The other development project was to increase the output power of the transmitter. Work was started on the design of a power amplifier which could be added on to the output of the existing 50kW (peak) transmitter. The amplifier had two stages: the first used a single power tetrode as a straight amplifier which drove an output stage using two of the same tetrodes in parallel. The peak output power was about 200 to 250kW.

 

While work on this was going on Dr. Eastwood had discussions with EEV about the possibility of using a klystron as the power source. This would mean an entirely new transmitter/receiver instead of an add-on to the existing transmitter. The concept had a number of advantages: unlike a magnetron a klystron is designed to be driven by a low-power source, so it will work with a fully coherent system; it would aim to have a peak power output of 500kW or 10 times that of the existing transmitter which would give the S264 80% more range and high cover; the klystron could expect to have an exceptionally long life which would balance its relatively higher cost: klystrons require very little power to drive them so the early stages of the transmitter would be simple: and finally, a suitable klystron already existed which had been developed for TV and had a mean power output of 17kW. In a radar it would only need to produce 1.0kW. (mean).

 

This klystron had four stages with a gain of 40db and was water-cooled. To ensure very stable operation one of the stages was removed, making the gain 30db. The cooling was changed to a forced air system which was simpler, cheaper and quite adequate, even in tropical countries due to the very low duty cycle of pulse radar systems.

 

Having obtained a prototype of the klystron a breadboard of the transmitter was produced at Baddow and installed at Rivenhall alongside the valve power amplifier for comparative trials. It was essential that no serious problems should emerge once the winner was chosen, particularly as this was probably the first time that a power klystron was to be used in a production radar. Based on the trials, the decision was made by Dr. Eastwood to go for the klystron, a choice welcomed by everyone concerned. This was in 1957 (or 1958).

 

It was somewhere around 1958(7) that the International Telecommunication Union (ITU) held a meeting in Stockholm to discuss frequency allocations. Up until that time the UHF bands 4 and 5 in Region 2 (Europe) were allocated to Television as a primary service and to navaids (?) on a secondary basis. Between bands 4 and 5 there was a small band from 585MHz to 610MHz which was allocated to navaids as a primary service and this was the band used by Marconi 50cm radars. At the meeting a large majority voted for this band to be allocated to television and to be reduced by 4 MHz in width so that it would provide three 8MHz TV channels. The UK and Belgium voted to accept the reduction in bandwidth but announced that they wanted the band primarily for radar.

 

Several delegates (who were mainly concerned with broadcasting and communications) apparently did not realise that their civil aviation authorities had bought Marconi 50cm radars and were using the band already. Everyone realised that a potential interference problem existed. A decision was made to set up a technical sub-committee to recommend a solution.

 

A few weeks later the sub-committee met in Brussels with SED represented by two “observers” attached to the UK delegation led by the Post Office. After three days of deadlock it was finally agreed that, of the three channels available one would be allocated to high-power radars, one to TV and one shared between low-power radars (S232 and S264) and low-power TV (I believe the latter was defined as having an ERP of one kilowatt or less).

 

The basis on which we were able to accept such a plan was the existence of the Marconi Pulse Recurrence Frequency Discriminator (PRDF). This device was designed to prevent “running rabbits” appearing on a radar display due to a neighbouring radar operating on the same radio frequency. The use of PRFD would allow all the projected high-power S264A radars south of Birmingham (ironically including the two French radars bought from Marconi) to share a common radio frequency. The only stipulation was that every station must operate on a different PRF.

 

Those stations North of Birmingham would be so far away from the French coast that no co-channel interference would occur to their TV. For the same reason the low-power radars in the UK could share channels with French TV without trouble. In the case of Jersey with two S264s they could operate with PRFD in the high-power radar channel. A PRF plan was drawn up at Baddow for all the European 50cm radars.

 

Meanwhile sales of both the S232 and the S264 were steadily increasing. Orders for the S232 usually included a single set of electronics because they were for use at airfields that were not operational 24 hours per day. The exception was the first one at Heathrow. The S264 on the other hand was used at large airports which operate continually and therefore needed dual electronics for extra reliability. Early overseas sales of the S232 were made to 8elgium, Germany and South Africa while the 8wiss bought two S264s.

 

It was already clear before the high-power transmitter design was complete that an expanding market for 50cm. radar was within our grasp. As the only supplier of such equipment it may seem that we had a rosy path all the way, particularly as MTCA (formerly MCA) and the Royal Aircraft Establishment (RAE) were supportive, not only with orders for equipment but in their belief that 50cm was the optimum wavelength for general ATC radars.

 

In practice it was hard going to convince the market that one firm on its own had got it right while the rest were out of step. The proof lay not only in having demonstrable hardware always available at Rivenhall airfield but in ensuring that prospective users talked to the MTCA and heard their views for themselves. Another problem was that countries like New Zealand and the UK which bought relatively large numbers of ATC radars, did so in dribs and drabs. This made it difficult to forecast just when a particular order would be placed .

 

In total New Zealand bought five radars to provide both long range ATC cover as well as approach control. Three sites in the North Island at Auckland, Ohakea and Wellington installed S264As while Christchurch and Dunedin in the South Island had an S264 and an S232 respectively.

 

 

At Wellington the radar site was on a mountain-top some miles from the City and it was a requirement that the station should operate unmanned. A wideband microwave link carried the radar signals to the ATC Centre while control signals for the equipment passed in the opposite direction over a narrow band link. As there was no direct line-of-sight between the radar and the CC, a large, flat metal plate was used to “bounce” the link signals in the right directions. The first NZ radar, which was for Wellington, was delivered and installed in 1960 and, like the majority of the other stations is, I believe, still in operation over 30 years later.

 

The MTCA went out to open tender for long range radars for airways surveillance about 1959. The specification called for each station to comprise what was effectively an S264A with a 10cm parabolic dish antenna mounted on the back. All the electronics were to be duplicated and housed in a building beneath the antenna. Needless to say only our tender had the required credibility and negotiations started with MTCA.  They had added a 10cm system to the S264A in the belief that it would improve the radar cover at low altitudes and perhaps to make it look as though they were not over-favouring Marconi. The outcome was that the dual-frequency radar was too expensive so they ordered standard S264As.

 

MTCA required two complete radars at each site and ultimately bought 10 S264A's for the five sites at Heathrow; Ash, Kent; Ventnor, IOW; St. Anne's, Lancs; and Lowther Hill in South Scotland. Four were bought initially and the remainder were spread over the years until 1968. In addition to the radars, fully duplicated broadband microwave links were supplied by Marconi Communications to bring the radar video signals from the Ash and Ventnor radar sites to the London Air Traffic Control Centre (LATCC) which at that time was located at Heathrow.

 

During this period sales of the S264 and S264A increased while the number of S232s sold declined. The market showed a preference for the larger antenna with its narrower horizontal aperture and smaller “blip” size on the displays. This was confirmed by several ATC authorities who converted their S232s into S264s simply by changing the antennas.

 

MTCA asked the Company to produce another version of the antenna with increased cover at high angles of elevation. These were needed for control of aircraft in the Terminal Area (TMA) which usually surrounds a major airport and which can include several other airports as well. In addition, there are often ”overflyers” which are not landing but simply flying across the TMA from one airway to another.

 

A modified aerial was produced by altering the shape of the top part of the reflector to reduce the angle of the “cone of silence” at the cost of a small reduction in maximum range. As with the standard antenna the new antenna could be used with either the low-power or high-power transmitters. The radars became S264H and S264AH respectively. The new antenna was first used at Heathrow in an S264AH which was sited on top of the British Airways car park at a height of about 200ft. For many years there were seven surveillance radars at Heathrow Airport of which five were Marconi S264 series.

 

This period saw the development of a “stretched” version of the antenna which increased the horizontal aperture to 67.5 ft. (20.6 m) and reduced the “blip-width” on the display to about 1.8 degrees. The additional gain of the antenna increased the maximum range of the S264A to about 150 miles.

 

By 1969 50cm radars had been on the market for about 16 years. During that time thermionic valves had been replaced by transistors. New signal processing units were developed for these radars and the original water delay line was replaced by a quartz polygon. In addition work was started on a new type of digital signal processor which not only employed integrated circuits but also promised to revolutionise the whole processing technique.

 

Sales of the low-power S264 declined over this period but the high-power S264A continued to sell overseas.

 

In 1972(?) we were informed by the CAA (formerly MTCA) that the BBC had told them that they needed to use all three channels (35, 36 and 37 between Bands 4 and 5) in the future for TV. CAA had objected but after a hard fight they had agreed to vacate channels 36 and 37 by 1990, ie. in 18 years time. The CAA would therefore not be buying any more 50cm radars but would, in the future, use the 23cm  band.

 

The news was alarming to say the least. Twenty years is not a long time in the forward planning of ATC systems so with the drying-up of CAA business we could not expect a lot more orders from the ITU's Region 2 (Europe). Fortunately Region 1 (Far East and Pacific) was not affected by the Brussels agreement. In Region 3 (N and 8 America) the whole of Bands 4 and 5 are given over to TV and both navaids and radar are denied frequencies. And so it turned out: orders from Europe all but ceased but from the Far East they hung on for a few more years

 

There was another factor: over the previous few years digital techniques had been applied to radar signal processing with considerable success. Radars operating on the higher frequency bands had the most to gain from these techniques because it was difficult to provide the required stability for good clutter rejection with analogue circuits. Now the advantages of using the 50cm band, outlined above were no longer so apparent, while the disadvantages of the large antenna and the need for careful siting were more obvious.

 

Over these years I was no longer involved with 50cm radars so I do not have accurate figures for sales over this period. From memory, there were sales to Iran (two radars) and to Malaysia (one S264A with the “stretched” antenna). I think one or more S264s were bought back from the Swiss, refurbished and resold. If so, they are not included in the totals below.

 

From memory, total sales over 25 years were, I believe, around 60 to 65 radars. Thinking harder, I can remember where and sometimes when 55 of them were sold. This leaves some 5, or perhaps 10, unaccounted for. No doubt there is a full record of all 50cm radar sales and dates held in MRSL today.

 

Although the sales may have tailed off with the end of the 70's, it was not quite the end of the story. The S232s and the S264s (all sizes) were big and their size matched their durability. Most ATC radars survive for around 10 to 12 years and are then replaced. A lot of the 50cm radars lasted for 20 years before TV took the band away (now the ill-gotten Channel 5). As mentioned above most of the New Zealand radars have exceeded 30 years active life. A good record.

 

Postscript

In 1980 I was told that the BBC had told the CAA that they now had no use for the three TV channels between Bands 4 and 5. Alas this was far too late as the CAA were well advanced with plans to replace 50cm radars with others working in the 23cm band. Clearly this story, which needs to be confirmed, made no difference to the demise of the band for radar. Ironically, a recent attempt to get Channel 5 TV going within this band was a failure.

 

 

 

Taylor

Home