Case 5,656: L.J. Comrie and the origins of the scientific computing service ltd.

was playing some kind of gainc with the uirplane. I wauld make a gain change ;md then I would see how 1 could adapt to Lhat gain to make thc airplane bchave reasonably wcll. I was surprised lo [sec1 how well I could adapt. " ' Six more flights followed within llic next 10 days, with all tour pilols participating. Counting the first two flights on 7 April, they tlew 57 high-speed approaches. The Enter[>rise would liavc taken over a year to do the same, sincc it was limited to one approach per flight, and cacti flight rcquired lmirling onto the 741. On 18 April, Manke took off, intending to add lo thc total of high-speed approaches. On the last of his six approaches that day, he roared to the runway at 265 knots with 100 milliseconds of delay addcd to Lhc F-8's usual response timc. Prom thc ground camcra film of that approach, it looked like Manke pulled the nose up a little too high on the ensuing takeoff and compensated with a quick and clearly excessive downward pulse, causing the F-8 to land nearly nose first. It took over five pulscs to sellle on a good dcparlure attitude and bcgin to gain altitude. Thc film from the tail camera is quite dramatic, with the runway lilling the scrccn on the first oscillation. What hnppcncd was that the control system workcd against him, lengthening the rccavery. He reported lrouhlc in controlling pitch the entire flight. On the fiflli approach, hc coni-meoted, " 111 just got it in my mind, I was going to drive that thing into the ground. " Mankc used a scrics of small-pitch inputs to get there, " probably a classic PI0 down to the ground.'' As lie rotatcd afler the high-speed touch-and-go approach, Ihe landing gear struts suddenly cnmprcssed, musing the pitch rate gyms to respond in such a way that the software disengegcd the control augmentation system and pol it in direcl mode. The dircct mode had degraded handling qndlities relative to the control augmcnta-tion modc, making it iiiorc difficult to recovcr. Manke first turned off the dclay, then manually engagcd the pilch stability :nigmenta-lion system lo help in Iinally damping thc oscillations.' As a relieved Manke pullcd away, Kricr deadpanncd from thc control cenler, " John, I don't think WB got the tbfa on that. Wc'd like LO have you run that one …