NASA research pilot Gordon Fullerton checked out how the PCA software worked in the multi-engine simulator at NASA Ames before fight-testing PCA in an MD-11.You're flying a large transport plane carrying hundreds of passengers and instantly you are unable to control the airplane - your controls system has gone out. As a pilot or a passenger, you hope that this scenario never presents itself, but if it did, what if there was a way to safely land the airplane by using throttles only? With a system known a Propulsion Controlled Aircraft (PCA) not only is the concept a possibility, but it is a reality. By using a specially designed software system a successful flight test program at NASA Dryden Flight Research Center was accomplished.
The project's humble beginning came from a rough sketch on a TWA napkin that Dryden Engineer, Bill Burcham, drew on a flight to St. Louis for a McDonnell Douglas Aerospace (MDA) meeting. He shared his idea with his traveling companion, Dryden F-15 Project Manager, James Stewart. He thought it was a great idea and within five minutes had outlined a test program. They shared it with the people at MDA the next day and they thought it might work too.
From Sketch to Sim
The intent of the Dryden program was to develop a computer-assisted engine control system that lets a plane land safely with only engine power if its normal control surfaces such as elevators, rudders, or ailerons are disabled.
A simple sketch on a TWA napkin by NASA Dryden engineer Frank W. "Bill" Burcham led to development and validation of the Propulsion-Controlled Aircraft concept.
Over the next few months Burcham met with several individuals to look into the feasibility of developing a PCA program for use on the F-15 aircraft at Dryden. He met with F-15 simulation engineer, Tom Wolf, to see if the simulator could be configured to fly with throttles alone. He called a day later and said it was ready to try. By advancing one throttle and retarding the other, the F-15 rolled nicely and if both throttles were advanced slightly, the airplane climbed. Burcham crashed the first few times, but with practice did better.
With the concept underway they asked Gordon Fullerton, a Dryden pilot and two time veteran on space shuttle flights, to take a look at the F-15 simulation. They learned that it was possible to keep enough control to maintain a rough altitude or heading without difficulty, but the precise control required to land was just not there. Based on those findings they went ahead with work on the computer assist concept that had been sketched on the napkin.
"After several incidents involving commercial aircraft where hydraulic failures caused aircraft to lose part or all of their flight controls, we started work on developing this automatic engines-only control system. Within a few months, I was pretty sure we could make it work, but I wasn't sure we would get a chance to fly it," said Bill Burcham, Chief of Dryden's Propulsion and Performance Branch.
MD-11 PCA's first landing at Edwards Air Force Base.
As the months progressed, Dryden controls engineer, Glenn Gilyard, put together a simple control system where the pilot used the control stick to make inputs and the computer took the stick commands, plus feedback from the airplane flight path and bank angles and calculated the throttle positions. This system worked pretty well and nice landings in the simulator were made on the first try, but these were all made by those who had practice. So the idea came up to bring in a "novice" to the simulator to see how well it really worked. Bill Dana, a NASA test pilot who had flown the X-15, was called in to give it a go and to their gratification he was able to make a perfect landing on the first try.
Early on James Stewart invited pilot Al Haynes to come to Dryden and to fly the PCA simulator. Haynes was impressed with what the computer assist program could do. As the program progressed, they switched from the use of a control stick to using thumbwheels to make the course adjustments for the throttles. Changes to the NASA F-15's digital flight control system included a cockpit panel with two thumb-wheel controls, one for pitch (nose up and down) and the other for banking (turn) commands. The system converted the pilot's thumb-wheel inputs into engine throttle commands.
Debut on the F-15
With the help of Jim Urnes at MDA, they developed and tested the flight software and managed to squeeze PCA into the already existing F-15 research program. By 1992 the PCA software was working well in the simulation and seemed ready to fly and in early 1993 the airplane became available. With all of the initial success the first PCA flight was disappointing. Roll control was very sluggish and PCA did not handle turbulence well.
Fortunately, the MDA PCA engineer, Ed Wells, had designed almost every part of the software to be adjustable from the cockpit. MDA was responsible for the engineering analyses, integration of the software into the F-15's flight control system and also supported the test flights. The PCA system automatically programmed the engines to turn the aircraft, climb, descend, and eventually land safely by varying the speed of the engines one at a time or together.
F-15 PCA (Propulsion Controlled Aircraft) simulation cockpit.
On April 21, 1993, Gordon Fullerton landed the F-15 twice using the PCA system without using any of the flight controls. Over the course of the program, six other pilots flew the PCA equipped airplane and they were all impressed. The plane flew at various altitudes and in unusual attitudes that might be experienced after a major flight control failure. And in every test, PCA recovered the airplane successfully.
"Having the capability to control and land an aircraft without the use of the flight control system, nor the movement of the aircraft control surfaces, is a real breakthrough in technology for improving the safety of future aircraft, both civil and military," said Dr. James Stewart, PCA Project Manager.
The Transport Test
On August 29, 1995, Gordon Fullerton once again made a successful landing using only engine power for control, but for this flight the PCA system was installed on a McDonnell-Douglas MD-11 transport aircraft. The success of the program was the result of a partnership between NASA and McDonnell Douglas Aerospace, St. Louis, MO, with Pratt & Whitney together with Honeywell designing the software used in the aircraft's control computer. NASA Ames Research Center, Mountain View, CA, assisted in the program by performing simulations.
For the MD-11, the PCA system uses standard autopilot controls already present in the cockpit, together with programming in the aircraft's flight control computers. The aircraft demonstrated software used in the flight control computer that essentially landed the MD-11 without a need for the pilot to manipulate the flight controls and without the use of conventional, hydraulic controls.
The PCA concept is simple - for pitch control, the program increases thrust to climb and reduces thrust to descend. To turn right, the autopilot increases the left engine thrust while decreasing the right engine thrust. Since thrust response is slow, and the control forces are relatively small, a pilot would require extensive practice and intense concentration to do this task manually. Using computer-controlled thrust greatly improves flight precision and reduces pilot workload.
Aircraft manufacturers have decided that a PCA system will be valuable for use in the design of future airplanes. When they incorporate this for future design, it will eliminate the need for a less capable hydraulics-dependent backup flight control system.
"Now that the technology is proven, I hope to see it incorporated into future aircraft designs," Burcham stated. "I also hope it never has to be used."