• If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • Stop wasting time looking for files and revisions. Connect your Gmail, DriveDropbox, and Slack accounts and in less than 2 minutes, Dokkio will automatically organize all your file attachments. Learn more and claim your free account.

View
 

Fur Hat Presentation

Page history last edited by Ian Gillis 3 years, 5 months ago

Fur Hat

Home

Introduction

This article first appeared in the December 1992 edition of News and Views. Thought to have been written by Derek Watkins it has been slightly adapted to deal with the passage of time.

 

“FUR HAT is a big topper”

At the Farnborough Air Show in 1992 the Royal Swedish Air Force paid tribute to Marconi Radar, to whom it presented a plaque commemorating a remarkable achievement — the continuous operation of the digital data handling system installed by Marconi in 1963, with no store failure in all those years!

 

In the early 1960s it was obvious that the time available from the initial detection of a possible air threat to the deployment of appro­priate counter-measures was desperately limited. The  problem for Sweden was particularly acute owing to its geographical position.

 

The Royal Swedish Air Force Board, as part of the STRIL 60 project, recognised that the answer was the full or partial automation of tasks that were then carried out manually, and turned to Marconi, who, in a project known as FUR HAT, set about applying high-speed data processing techniques to the new generation of control centres.

 

The resulting system, which was the most advanced of its kind in the world, was still operational in 1992 and for a number of years afterwards. It acquired information on all aircraft from a number of long-range surveillance radars, other centres and reporting posts. It was also continuously updated with meteorological information and data regarding the status of all defence resources.

 

Operators responsible for the different defence functions were automatically presented with up-to-date data relevant to their particular tasks, thus speeding the deployment of defence measures, whether they be the scrambling and guidance of interceptor aircraft, the selecting and passing of acquisition data to ground-to-air missile and gunnery sites or liaison with the Civil Defence forces.

 

The guidance of the interceptor aircraft to its target was assisted by the use of Marconi-designed computers [TAC]  which continually recalculated the appropriate intercept path and sent instructions to the aircraft via ground-to-air data links. Similarly, instructions to bring the surface-to-air missile tracking radar onto the correct bearing and elevation were passed automatically to missile sites.

 

By this instantaneous correlation, presentation and transfer of information, not only were the air defence controllers relieved of all routine tasks and left free to make decisions on which the defence of the country depended but also every vital second was saved, thus enabling defence armament to be used with maximum efficiency.

 

The main aircraft track data store, whose 29-year non-stop performance is commemorated in the presentation plaque, consisted of 72 synchronised delay lines, each made from a 30 ft coil of wire. Each line stored 1,500 bits of data re-circulated every 3 ms. Hence the store was (72 x 1500) bits and occupied 12 cabinets of equipment. Even though one small chip (little more than 100 kbytes) could later have replaced this capacity, the facilities were so far in advance of their time in the '60s that they were comparable with present day systems.

 

The system was designed with a considerable redundancy and was able to continue in operational service under a variety of fault conditions. Faults could be located, repaired and equipment brought back on-line without the system having to close down. FUR HAT was acknowledged to be a major project even in the days when massive projects abounded.

 

For more than two years about 60 Marconi engineers were employed in Sweden on the installation and commissioning of the display and associated systems, which were housed in a four-storey building inside a mountain. The entrances to the tunnels leading into the mountain were camouflaged by ‘trees’ ingeniously made from wooden poles and netting. One day one of the engineers tied a big red apple onto one of them. The Swedes were not amused and asked that it be removed. The next day only the core forlornly hung there.

 

The commemoration plaque which marked the Swedes’ appreciation of the system that served them so well was made from a spare store unit mounted on an oak panel with the names of many of Marconi’s Fur Hat engineers engraved around the frame. It was flown to Farnborough for the presentation ceremony in what was then the latest Gripen JAS 39 fighter.

 

In later years the Swedes also entrusted Marconi Radar with the Mayflower and DBU260H (TOR) projects, providing still further major capabilities.

 

Photographs of presentation  

 

Those who worked on Fur Hat:

- on the left, Don Beckett, Rex Manville, Geof Akerman (dec'd), Clive Gildersleeves (dec'd), Mike Steeds, ANO (Field Services?), Ted Pegram, Harry Fancy

- on the right starting from the centre: Frank Savill, Alan Matthews, FMV ?, Jock Williamson, ?FMV x2, Don Halstead, FMV x2.

(Editorial Note: FMV is the Swedish equivalent of the MOD Procurement Executive.)

 

Presentation of Delay Line Store:

Alan Matthews, Rex Manville, Mike Steeds, Ted Pegram, Harry Fancy, ANO, ANO (FMV?), Frank Savill.

 

Delay Line Store

Input from Brian Partridge

Re the magnetostrictive delay line, Jock Williamson and myself were responsible for getting this developed at Ferranti. It enabled us to create a parallel store up to 80 bits or more wide and a good number of words per store. This was needed so that we could do computation in real time as data came in. There was a small electronic store between the end and start of the wire store to enable those computations to be made and new data entered. Each wire had a sliding start point so that all individual bits could be aligned. The original store had a compressive transducer to send data in, but Jock and I found this to be sensitive to mechanical shock, so this was changed to a torsional transducer which was less sensitive to shock.

 

This was one of the first parallel stores used for computation. Seems daft now.

 

Input from Alan Matthews

As I remember it the store delay was 3ms divided into 2microsecond bits. That is 1500 bits in circulation. Each of the 150 target aircraft was I believe allocated 10 of these bits and there were 72 duplicated stores so each target would have 720 bits of data. X and Y coordinates would take 24 bits so at 720 bits per target there would be plenty of room for quite a lot of information.


Another feature of the system is that to get at track data the system only had to wait 3 ms for the data to appear from the store.


However, physically it occupied 12 6-foot high cabinets for the 144 units which for 1500 x 144 bits of data would be quite a lot of real estate by modern standards !!!

 

Names on Plaques on Store

 

Fur Hat

Home

Comments (1)

Ian Gillis said

at 4:09 pm on Feb 11, 2016

Page checked

You don't have permission to comment on this page.