NASA Office of Logic Design

NASA Office of Logic Design

A scientific study of the problems of digital engineering for space flight systems,
with a view to their practical solution.

2006 MAPLD International Conference

Ronald Reagan Building and International Trade Center
with a session at the Smithsonian National Air and Space Museum

Washington, D.C.

September 26-28, 2006

From Runway to Orbit: Reflections of a NASA Engineer
Kenneth W. Iliff and Curtis L. Peebles

Chapter VIII: The Approach-and-Landing Tests -- First Flight

The first ALT free flight was scheduled for the morning of August 12, 1977.  Even though Johnson was in charge of the flight, Dryden had set up the control room in much the same way as we had for the X-15 and lifting body programs.  This way, the Dryden engineers could monitor the flight in the same fashion as we had with those earlier programs. More than a thousand reporters and between 60,000 and 70,000 visitors were at Edwards. The Shuttle/SCA mated pair was backed out of the Mate/Demate facility, the 747 started its engines, and the combination began the taxi to the runway. It began the takeoff roll at the scheduled time of 8:00 a.m.  The climb took a few minutes longer than expected, but the Shuttle/ SCA and its attending T-38 chase planes were finally in position. Fred Haise, in the Shuttle, radioed that he was ready, then pushed the separation button.

I don't remember any comments on launch, but from my point of view the Shuttle cleared the 747's tail smartly. (Fig. 1) The Shuttle was many tail heights above the 747 by the time the base of the Shuttle was above the tail of the SCA.  I believed that was going to happen, but I was sure glad to see that it really did happen.

The Space Shuttle prototype Enterprise flies free of NASA's 747 Shuttle Carrier Aircraft (SCA) during one of five free flights carried out at the Dryden Flight Research Facility, Edwards, California in 1977 as part of the Shuttle program's Approach and Landing Tests (ALT). The tests were conducted to verify orbiter aerodynamics and handling characteristics in preparation for orbital flights with the Space Shuttle Columbia. A tail cone over the main engine area of Enterprise smoothed out turbulent airflow during flight. It was removed on the two last free flights to accurately check approach and landing characteristics.
Figure 1.  The first separation of the Enterprise
from the SCA with the tailcone on.  (E77 8608)
(high res, 4.5 megabytes)

Haise and Fullerton were heading north in the Shuttle. They did little inputs to see that they were getting responses in pitch, roll, and yaw. Pilots will do that whether they are asked to or not. It's how they know that everything is hooked up right, the computers and the hydraulic system are doing what they should, and the vehicle is responding in each of the three axes due to the stick and the rudder inputs. Fred then did a 180-degree turn to line up with runway 17. Then, as had been done on the M2-F2, he did a practice flare at altitude to see that the vehicle was going to arrest its sink rate much as the simulator and the wind tunnels had predicted that it would. Then Fred came onto final approach and did his final energy management with the speed brakes, part of which was planned, I believe, and did a nice touchdown and rollout of close to two miles on the lakebed.

Getting ready for the first flight was exciting.  It was certainly very visual to watch it separate and to do its 180-degree turn, then set up for the landing and complete the flare. Looking back, it was anticlimactic in a way, because the L/D max was so high that the glide angle (about 17 degrees) never got too steep, so it didn't evoke the feeling that we had with the lifting bodies and the X-15 of the vehicle diving at the ground. Visitors watching the vehicle's flight had been impressed, because it looked steep to them, but they hadn't seen the really steep ones that I had observed in the early 1960s. The first flight was a success. Although there were a number of gripes from the various disciplines, and some things had to be worked on quickly before the next flight, there was nothing major as I recall.

There were no real stability-and-control or performance pulses done on the first flight. But the pilot was still maneuvering the airplane, and those inputs and resulting vehicle responses could be analyzed to make spot comparisons for the lift and drag, and these were as predicted. So we got a little bit of a feel that things were about as advertised from the lift-and-drag point of view, and the same was true from the stability-and-control point of view.  It may not have been as predicted, but it was close enough that the pilot was probably not going to have great difficulty doing the lift-and-drag maneuvers and the stability-and-control maneuvers on the second flight.

We had the strip charts in the control room from the first flight, and we also had telemetered data from the Shuttle to Dryden. The primary things that the stability-and-control and performance engineers were interested in were on the aerodynamic coefficient identification package (ACIP). We also had other information on the operational instrumentation (OI) system, and the backup-flight control system (BFCS). Our first task, in terms of analysis of the data, was to merge those three data streams and account for any time skews from sensor to sensor on a given system and then the time skews between the clocks on the three different systems, so we got some practice doing that.

Rich Maine had developed the software for doing it.  Some things weren't quite as they were specified in terms of official time skews, but we had some techniques that we could use for refinement. It was very important for all of the time histories from these sensors be within a few milliseconds of the time the measurement had been. We then interpolate between those samples so we can line them all up; as if they were all simultaneously sampled. That's a very important thing for us to do, and we learned how to deal with this problem on the Enterprise during the ALT program. We got a feel that the aileron, elevator, and rudders were all about as effective as predicted, which was within 10 or 20 percent.  The control effectiveness wasn't going to cause any big problems with trying to get the maneuvers that we'd requested for subsequent flights. The same was true of the ACIP instrumentation for the lift-and-drag data from the maneuvers that we were going to perform starting with the second flight.

So my memory of the first flight is that it was exciting getting ready for it, the flight itself was exciting, and it was exciting getting our hands on the data the first time and seeing what it had to tell us. It's probably a difficult thing for other people to understand, but the nerdiness of an engineer with a new set of data to analyze is a lot like Christmas to a five-year-old. Everything's exciting -- the package, the box and what's in it, and all the things 'to play with. There's definitely an overload at that time.  It's very pleasant to an engineer who is going to analyze that data when things go well. When things don't go well, it's equally exciting, but if we're on the critical path, it can be overwhelming in terms of trying to understand what the data is telling us so that we can feed that to program management. If it's a serious problem, it either gets fixed or avoided. If it's not a serious problem, we can show that it is a problem that's not going to impact things.

Those were the kinds of things we were looking for with the data from the first flight. We had the normal crew debrief that we have after every flight.  The Johnson Space Center was in charge of those flights from the time the mated pair started taxiing until the Shuttle was back on the ground, but it didn't really feel that way because our control room was very active. Deke Slayton was there. A lot of the Johnson people were there, even though control was being run out of Houston, so even though Dryden was not really controlling the flight, we had all the information that we needed.

At the crew debrief, we were very interested in hearing the reaction of the pilots to what they had just flown, along with responses from various people who were in charge of monitoring other things in terms of vehicle health -- the hydraulic pressures, the APUs, and the problems that occurred. It was quite a long postflight debrief, since it was a first flight, but those are always very interesting. We always had those for all of the airplane programs that I've worked on. We usually have the postflights as soon as we can, while things are still fresh in the pilots' minds.  There could be more extensive debriefrng and questioning later, but listening at the crew debrief is how we capture the feeling of the crew.

We had the debrief, then we went into Lancaster and had a party. It was a different kind of party than we had for the X-15. It was more formal and, I think, less noisy and less enthusiastic than we had for Dryden or Edwards-driven programs such as the X-15 and the lifting bodies. But the atmosphere was the same. It was a celebration for those who were actively working on the program, the chance to let their hair down and get rid of some of the tension.

The next thing was to get back to the data. Initially we just had the strip charts, and over a period of a few days we started to get information from the other systems so that we could merge all of the data and interpret it. Once the time came to do our analysis, we did it exactly the same way as we'd always done it. The way we fed it back into the program was more formal and much different because we didn't just reveal our results or concerns to others working on the program or program management. There was a very formal structure, and within Dryden we had a formalized structure of how the various ALT engineers and technologists worked -- that is, what we did with our data, what kind of review and oversight it
had by our management, and how that was then fed up through the various disciplines and levels of management up to the top of the Shuttle program. That was different from the smaller programs that I'd worked on before, but the analysis and the excitement of the data was exactly the same as they had been with the other vehicles. I did have the feeling that I was a much smaller frog in a much larger pond than I'd been with the M2-F1, the M2-F2, and the X-15 to a lesser extent.  It was the same kind of job, but it was in a bigger world than I'd been in before.

After I had looked at the time histories from the first flight, it was clear why the separation had been so positive. The Shuttle pulled about 1.8 g's as it separated from the 747. With the Shuttle launching at about 1.8 g's, and the 747 probably being a little less than 1 g, they separate at about 25 feet per second squared. A launch at 0 g from the B-52 separates the two by roughly 32 feet per second squared acceleration. The Shuttle was just going up as opposed to going down like the X-15. Once I saw that, separation never concerned me again. At launch on the first ALT flight, Fred Raise had rolled it off to the right as Fitz Fulton, who was flying the SCA, rolled to the left. The roll that Fred did was actually the very first Shuttle maneuver that I could analyze. It wasn't a large roll, but it was a very definite input on the aileron.


Invited History Talk - Fred Haise

2006 MAPLD International Conference: Home Page

Home - NASA Office of Logic Design
Last Revised: February 03, 2010
Digital Engineering Institute
Web Grunt: Richard Katz