Bruce Lumsden joined the Blind Landing Experimental Unit (BLEU) in 1966 straight from the University of Edinburgh to work on the automatic control and guidance of aircraft particularly during the landing phase. Initially, he investigated the application of Kalman filters to guidance systems and developed improved dynamic models of the aerodynamics and Avon engines powering the unique Comet 3B. He successfully developed autopilot control laws to conduct automatic curved approaches to airports simulating the guidance proposed for the new microwave landing system (MLS) using the Comet 3B with its analogue TR48 based versatile autopilot (the Comet’s last trial before withdrawal). He alsoand flight tested new approach and landing control laws incorporating direct lift control using the modified spoilers of the BAC 1-11 (XX105) which had replaced the Comet 3B as the principal BLEU research aircraft.
He then designed control laws to enable the BAC 1-11 to conduct automatic two segment approaches (60/30) to touchdown through the standard SEP5 autopilot as a means of alleviating noise at airports, comparing the results with other approach procedures such as the continuous descent, low drag and ILS glide paths >30. The two segment approaches were successfully tested at Gatwick and Heathrow in the UK and at Zurich to establish any incompatibility with existing procedures and certification. Based on noise measurements undertaken at Zurich and Bedford, the benefits of such approaches were then determined for the range of aircraft operated by the then BEA/BOAC.
Transferring to the HS748 (XW750) at Bedford, Bruce became the project officer for the demonstration of the UK Doppler MLS contender for the ICAO MLS competition. Following the first ever automatic landing using MLS at an international airport at Gatwick, DMLS was trialled and demonstrated at Manchester and in the mountains around Berne in Switzerland where two segment approaches in both azimuth and elevation avoiding high ground were undertaken. Further demonstrations in Tehran (Iran) in the most difficult of circumstances and in Montreal during the ICAO meeting showed the capabilities and robustness of the DMLS. Unfortunately, the vote went against the UK in favour of the US/Australian time referenced scanning beam system.
The UK Dept. of Transport and Industry (DTI) sponsored the Economic Category 3 programme to reduce the costs and complexity of existing fail operational Cat 3 systems, it was established by Bruce’s team at Bedford that manual landings could be achieved down to 250m RVR from a 50ft decision height and with a fail passive autopilot (the SEP6 autopilot with monitored autoland) this could be reduced to 200m. In addition, further reductions were possible with a head up display as a backup measure. The HUD used in the HS 748 was the Marconi Avionics Monocular HUD which proved successful when the symbology was designed and developed for the approach and landing task. The programme included new designs of approach and runway lighting to minimise costs and new techniques of RVR measurement.
Bruce then turned to military all weather operations which were limited to a minimum DH of 200ft at suitably equipped runways and established that further weather minima reductions were possible with transport type aircraft using better procedures and techniques coupled with independent monitoring and pilot aid systems. For fighter aircraft, the existing precision approach path (PAR) radar approaches were modified with the development of minor changes to the HUD symbology and the talk down controller phraseology. The performance achieved reduced the lateral and height scatter on approach significantly and provided the required delivery accuracy to support Cat 2 operations with a 100ft DH. The new systems were tested using the HS748 in fog conditions across a range of RAF airfields in the UK and Germany and these were universally accepted, reducing the workload of both pilots and controllers. The HUD in Tornado ZA326 was also modified and successfully tested at Bedford and RAF airfields, providing the same level of improved performance as the 748. In spite of heavy lobbying, PAR was planned to be replaced by MLS and the planned modification fleet wide was not taken up; PAR is still with us, some 35 years later.
Bruce also investigated the use of night vision goggles for night time approaches in poor visibility, the application of forward looking infra red (FLIR) systems in the aircraft and on the ground, and map based navigation techniques to provide approach and landing guidance at austere sites, the forerunners of enhanced and synthetic vision systems (EVS/SVS).
Away from approach and landing, Bruce’s team devised and flight tested a dual Smiths Industries 1553B data bus system in the HS748 which connected a range of systems and devices including MonoHUD, automatic flight control system, inertial navigation system and guidance systems (ILS, map based guidance). This was the first ever aircraft flight using 1553B to connect flight critical systems.
Part 1 consists of 19 chapters of 500 pages, is full of detail and illustrations and priced at £25 plus £5 P&P in the UK. Prices for Europe and the Rest of the World are available on request. It is available from the author or through the Bedford Aeronautical Heritage Group (BAHG) or the Farnborough Air Sciences Trust (FAST) bookshop.
Part 2 of these memoirs, to be published separately, covers the author’s involvement with rotary wing aircraft, in particular the Sea King and Wessex research aircraft, developing an all-weather recovery system for helicopters operating to small ships at sea. For this work, Bruce received the RAeSoc Aero Marine Award and the IMechE William Sweet Smith Prize in 1998, and a Technical Cooperation Programme (TTCP) achievement award for work on the prediction and validation of ship air wakes using CFD techniques in 2000.
Copyright © 2023 Waverley Aerospace Publications