This article is a memoir of Harry Schmidt, a Pratt & Whitney test pilot at Edwards AFB during the mid-1950s. Schmidt recalls that the J57 was critical to P&W’s long-term success, so the testing program was vital both to the United States Air Force (USAF) as well as to P&W. Early in the testing program, the engine would stall even in a shallow bank; however, after a couple of months, the team was able to complete hard turns without a stall. To solve the high-speed stall problem, P&W engineers tweaked the engine turbine vanes to provide greater stall margin. In the fall of 1955, the team received its F-101, a twin-engine fighter with two J57s and a top speed of about Mach 1.5. Because the team was operating the fastest aircraft in the world, they figured they could also set an unofficial world’s altitude record by achieving max Mach at about 40,000’ and then climb trading airspeed for altitude.


The high-performance MIG-15 over Korea in the early 1950s spurred development of a new US fighter. The USAF wanted supersonic speeds and 50,000 foot altitude capability. North American designed their Mach 1.2 F-100 utilizing Pratt & Whitney's new 10,000-pound thrust engine, the J57, to meet the USAF desires. In those early days of high performance jets, with an industry having only rudimentary knowledge of supersonic aerodynamics, no computers to help design either airframes or engines, and a wing sweepback design copied from a German jet, it was no surprise that the airframe had very troubling flight characteristics (highlighted when NA's chief test pilot George Welch was killed during a high-speed dive, the first of several crashes), and the engine encountered severe compressor stalls caused by inlet duct curvature. Both NA and P&W expanded their flight test operations to rectify their individual problems.


I had just completed a 4-year stint as a USAF fighter pilot flying F-94s after having won a big air-to-air rocket firing competition a month before leaving. The win made me a favorite of the CO, Maj. Chappie James, later a 4-star general. When P&W needed a fighter pilot to test their new F-100 at Edwards AFB, Chappie recommended me. Although I had numerous civilian jobs offered, the thought of flying at Edwards was an overwhelming challenge. I applied. Thanks to a degree from RPI and Chappie, P&W offered me the job.

When I arrived at Edwards I found a very small P&W staff operating that facility consisting of the manager, Jim Peed, supported by four engineers. Since I was a new P&W employee with no flight test experience, and I perceived that the small test engineering group at Edwards was also relatively young and new to flight testing, I was humbled that P&W would entrust testing their important new J57 engine to us. While I was the weak link and awed with the responsibility, I wondered if we could we meet the challenges. The J57 was critical to P&W's long-term success, so our testing program was vital both to the USAF as well as to P&W.

I had serious reservations – the F-100 was known to be a difficult aircraft to fly, maybe even dangerous under certain conditions. For unknown reasons, the few military pilots at EAFB who had flown the aircraft before were largely uncommunicative with me, then a civilian, about abnormal flight characteristics, and flight manuals weren’t available for several years more. Everybody knew about a low speed instability problem on takeoff - all pilots on their first takeoff experienced severe up and down changes to pitch attitude in quick succession, something they called it the “JC” maneuver. You only had about 15 seconds from lift-off in which to figure out how to control this unusual flight problem before crashing. And, of course, I knew about the compressor stalls that were associated with aircraft maneuvering or throttle movements - aggressive maneuvers distorted the airflow through curving inlet ducts into the engine causing the high-speed compressor stalls. We had to duplicate these maneuvers, simulating air combat, in the real aircraft (the F-100 and later the F-101). I was not warned about the unusually high airspeed necessary to complete a landing, caused by the lack of trailing-edge flaps (although it had leadingedge slats), or a few other unusual flight characteristics

Our primary flight testing maneuver was the wind-up turn. That is, to start turning, at first rather gently, but slowly increasing the rate of turn by flying ever-tighter circles up to the maximum G load of the aircraft (about 7 Gs). You can envision driving your car at high speeds in increasingly tight circles to determine when the tires lose traction and the car skids - this is what we did in the air. Early in the testing program the engine would stall even in a shallow bank, but after a couple months we were able to complete hard turns without a stall.

In addition to this testing being hard on the engine it was also hard on the pilot, since as the “G” load is increased blood is drained from the pilot's head until it is possible that the pilot would “black-out” (essentially losing vision and consciousness) from lack of blood supply. To counter this problem we wore G suits (similar to a very tight corset, further inflated by air pressure) fitted to the lower body keeping blood from draining into the legs, retarding blacking-out.

To solve the high speed stall problem, P&W engineers tweaked the engine turbine vanes to provide greater stall margin. After P&W experts in Hartford had reviewed the data from each test flight, their engineers would further modify the stall margin as appropriate, and we would fly again to perform the same maneuver to evaluate the change to compressor stall boundaries. It was important that the wind-up turns be performed in exactly the same manner each time, for without consistency of flight profile, the stall results could not be trusted to be comparable.

Maintaining exactly similar altitudes, airspeeds, and G loads was not a simple task. We carefully modified stall margins since engine efficiency is highest when operating close to the stall line. In other words, fuel consumption (SFC) increased with greater stall margins, so we had to balance increased SFCs vs improved stall margin. Achieving the optimum balance was important since the J57's attractiveness was based upon greater thrust as well as greater efficiency. After a few months we had resolved and optimized the compressor stall issue.

In the fall of 1955 we received our F-101, a twin-engine fighter with two J57s and a top speed of about Mach 1.5. Made by McDonnell, it represented a major step forward in speed and altitude performance beyond the F-100. The engine had even worse compressor stall problems than the F-100 since the design had engine air inlets on each side of the fuselage, with more severe turns in the air duct further distorting airflow to the engine compared to the F-100.

We had no access to prior operational experience with the aircraft, once again no manuals, operating procedures or recommended airspeeds, and no USAF or McDonnell pilots to talk with. We soon knew it had “pitch-up”, and that high AoAs (angles of attack) were responsible for pitch-up – but how high was too high was never answered, and what caused the problem was unknown. After two crashes only a few months later we knew it was dangerous – warning, avoid pitch-up or crash. Initially, however, compressor stalls occurred long before the pitch-up boundary, so we were operating in a safe flight regime. That soon changed. Initially it was not hard to get compressor stalls – just advance the throttles and an engine would stall. Slowly we improved the stall margins and began with serious wind-up turns at all altitudes and airspeeds. We were performing supersonic wind-up turns above 50,000’, substantially above the F-100's capabilities. We had made progress with improving the stall margin, but now encountered the dangerous threat of pitch-up.

While most initially believed pitch-up was caused by the swept-back wings, I personally thought it was the wing that blanked-out flow over the T-tail (elevators) at high AoAs. As our compressor stall situation improved, soon I was living on the dangerous intersection of the compressor stall and pitch-up boundaries, and flew several flights at the request of the USAF just to evaluate the pitch-up boundary. Fortunately I believed I could detect the onset of pitch-up, keeping me safe – many others were not so lucky. Eventually McDonnell engineers (and Lockheed with their F-104) concluded it was a T-tail problem and installed a stick shaker / stick pusher system to theoretically keep pilots out of trouble, and some years later that same system was adopted for civil aircraft with a T-tail design.

In about January 1956 we received a new engine from Hartford. This engine had a “convergent-divergent nozzle” which, in theory, would permit supersonic airflows in the nozzle, substantially increasing thrust and speeds. With only one new engine we went to Mach 1.7+, and then, when the second engine was installed, we went closer to Mach 2.0 (depending upon altitude and temperature). We were testing the fastest aircraft in the world! Only the few in P&W engineering knew how fast we were going.

When USAF Fighter Ops at EAFB learned in March 1956 that the English had just broken our US-held world's speed record, made in an F-100C the year before at 822 mph, the USAF guys were distressed that there wasn’t a US aircraft that could reclaim the record. I told them about our F-101 with our new more powerful CD nozzle engines and offered to fly anytime they desired to regain the record. Instead they wanted to use our aircraft, but did not want a P&W (civilian) pilot. The USAF would use a military pilot. In December 1957 they sent Maj. Adrian Drew to fly our P&W test aircraft setting a new world's speed record of 1207 mph.

We could operate above 55,000’ supersonically, although our maximum speed was somewhere between 35 - 40,000’. Since we were operating the fastest aircraft in the world, we figured we could also set an unofficial world's altitude record by achieving max Mach at about 40,000’ and then climb, trading airspeed for altitude. One day we tried. I followed our plan: Max level speed just below 40,000’, then slowly climb and bleed-off airspeed during the climb. At about 70,000’, looking outside, I saw a sight I was not prepared for – the sky was black, you could see stars during the day, visibility was out to the horizon a jillion miles away, and you could easily see the curvature of the earth. Suddenly a fire warning light came “on” on the left engine, so that engine was shut down. Almost immediately the right engine started a compressor stall, so all thrust was lost from that engine which was throttled back. I headed down quickly, going over the top at about 75,000’ with only one operating engine.

An uneventful landing was made at Edwards. Then the engineers and I went out for a few beers discussing the flight, particularly the beautiful views of earth.

In the space of several months flying the F-101 we had … 1/ successfully completed our compressor stall investigation, 2/ tested new CD nozzle engines making us the fastest aircraft in the world, 3/ flown to 75,000’, making us the highest flying aircraft, 4/ contributed to the knowledge of pitch-up in aircraft designed with sweptback wings and T-tails, helpful for later fighter as well as similarly designed civil aircraft. Altogether, not an insignificant contribution to aeronautical knowledge from a new and previously unproven flight test group.

Then we got new J57/J75 powered aircraft, new test programs, and more challenges.