FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1971
1971 - 0804.PDF
FLIGHT International, 20 May 1971 709 TriStar in the lab The development programme behind the trijet involved about 20,000 hours of wind-tunnel testing at 20 different sites at a cost of $20 million. G ROUND TESTING of the Lockheed TriStar is now 85 per cent complete, while the flight-test programme is running ahead of the revised programme instituted after the Rolls-Royce collapse, and is targeted to achieve FAA certification next April. This was said by Mr G. D. Sim, Lockheed's chief engineer, advanced design and laboratories, and Mr Don Moor, senior experimental test pilot, during a lecture tour of several branches of the Royal Aeronautical Society throughout Britain which they com pleted last week. Brief mention was made of Mr Moor's talk in last week's Flight. Mr Sim gave a well illustrated lecture on the laboratory testing, which has cost $100 million (£42 million) and involves 90 per cent of all test man-hours. An early "brain- drain" man from Britain to the USA, Mr Sim was formerly Vickers Armstrong's wind-tunnel manager at Weybridge before joining Lockheed an 1956; he is now head of Lockheed-California's $44 million, 500-acre Rye Canyon laboratories, which employ 1,300 people, more than half of . them professional engineers and scientists. Four men acted as the connecting links between the Rye Canyon laboratories and the project engineering staff at Burbank, in the respective areas of structure, vehicle systems simulator, sub-systems and avionics, said Mr Sim. Any laboratory test data put forward to the project people had to be answered by a TMA—test memorandum action— within ten days. In this way management was quickly aware of the budgetary effects of laboratory findings. When the Mach 0-85 TriStar was projected Lockheed had no experience of a Mach 0-8-plus commercial jet air craft, and their fastest previous transport was the C-141 Starlifter at M = 0-74. This fact, coupled with the competi tive pressures from the DC-10, the need to prove new tech nology, the customers' demand for performance guarantees and awareness that the financial consequences of failure would be "devastating," decided Lockheed upon a very extensive wind-tunnel testing programme. This involved some 20,000 tunnel hours at 20 widely scattered facilities at a total cost of about $20 million (£8-3 million). Of the total, 18,000hr were completed in two years. Some of the wind-tunnel work was done in Britain by BAC at Weybridge and by the ARA at Bedford, to whom Mr Sim paid a particularly warm tribute. The ARA was employed as a back-up facility to the Cornell Aeronautical Laboratories for tunnel testing over the cruise-speed range, and Lockheed took 80 per cent of the ARA's transonic "' wind-tunnel testing capacity in 1969, said Mr Sim. There was very good correlation of data between Cornell and ARA. Once the propensity of Customs facilities to delay models for up to a week was appreciated and overcome there was no difficulty in flying models 6,000 miles to Bedford for testing—it was as quick as sending them across Los Angeles when the freeways were congested. In the 13ft X 9ft, 4m X 2-7m tunnel at BAC Weybridge some direct comparisons between the TriStar and the VC10, which have similar flap systems, were made. BAC provided VC10 tunnel-to-flight comparison data which provided an independent benchmark for the TriStar and aided in the extrapolation of tunnel test data on its high- lift systems. High Reynolds-number test work was done at Nasa's Ames 12ft, 3-7m pressure tunnel, while Lockheed- Georgia's Stol tunnel was also used. After the configuration development was nominally complete, the lecturer added, testing showed that a revised, larger, fuselage-to-root fillet reduced cruise drag by a very significant 3 per cent at M0-85. Thrust-reverser flow patterns were tested in the wind tunnels, to ensure that operation inadvertently or because of malfunction produced minimal pitch change. Rolls-Royce's problems with Hyfil blades and the need to switch to a heavier titanium fan produced substantial changes in engine gyroscopic effects and, for Lockheed, involved the construction of a test rig to evaluate changes necessary in the pod, pylon and wing assemblies. Beginning 12 months before the official start of the project in April 1968, a flight-simulation programme had been running 24hr a day. Based on an Electronic Asso ciates computer, with an Aylesbury-built Redifon moving- belt visual system, the simulation was entirely for develop ment purposes and not for training, so a simplified cockpit, on a fixed base, was used. Lockheed had experience of simulator versus aeroplane calibrations for the C-5A (also done on this simulator) and thus their confidence in the facility was high, said Mr Sim. Actual flight test data had shown good agreement on overall handling, damping and stick forces but some rotary derivatives were improper and recent attempts to use the simulator to investigate turn- co-ordination phenomena had been unsuccessful. A full-scale fuel-system simulator, comprising a steel copy of the starboard wing, was made and ground-tested to 45,000ft, 13,700m simulated altitude, and to 120°F, 49°C. Operation to +15° pitch was available. The tests resulted in some redesign to eliminate boost-pump surge and a modification of the scavenge pump evolved on the rig reduced unusable fuel by 5001b. Wide-bodies, said Mr Sim, produced new challenges in cabin conditioning. Wooden mock-ups, with smoke and local-velocity measurements, were used and, in the developed configurations, were subjected to stringent sampling by critical juries—male and female—of typical passengers. Small sectional mock-ups were used to develop wall- and floor-heating systems and insulation. One such mock-up, said Mr Sim, would save an estimated $6 million, £2-5 million on a 170-aircraft production run. A complete mock-up of wing and fuselage ducting had completed proof tests and 144,000 pressure cycles—four times the aircraft design life of 36,000 flights or 50,000hr— at operating temperatures without any failures. Testing was now proceeding on engine ducting—much of it built by Shorts in Britain—where the goal was 216,000 cycles. Reliability data Mr Sim spent considerable time describing the vehicle systems simulator which resulted from an early decision to make maximum use of integrated systems simulation, to minimise production-line changes and flight-test problems, to obtain early maintenance and reliability data and to be used in lieu of flight test for some FAA approvals. Costing $15 million, £6-25 million—the sales price of a complete aircraft—the VSS enabled 12 months of operation with the all-flying tailplane, unusual on a big transport, to be logged before the maiden flight. It incorporated primary and secondary flight-control systems, avionics, hydraulics and undercarriage systems on a true space layout. Avionics in the automatic flight-control and autoland systems were integrated with the hardware to provide a total system on the VSS. Sensors for these avionics operate in a simulated "real world" by reason of the fact that the gyros and inertial equipment are placed on a programmed three-axis table. Among the benefits of the VSS had been the early identification of aircraft drawing and fabrication problems continued on page 729
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
