Profiling works in solving airplane problems–usually

Profiling is, in this discussion, a procedure to narrow down the possible causes of a problem based on its location in the airplane and timing in the order of events in a flight, and then with evaluation of the potential causes, select appropriate solutions to try. The case I’m going to describe occurred with the Cessna T-37 twin jet trainer, when it had been in service about a year in its mission to provide primary training for Air Force pilots, with an inclination towards fighter flying, even though graduates–with additional training–were just as likely to fly bombers, transports or VIP airplanes.

T-37 jet trainer
The T-37 was a fine airplane, but not so much after a flameout.

The airplane had been assigned to just a few bases, but the students were introduced, among others, to one particular type of approach to a landing, the fighter approach. The fighter approach didn’t depend so much on the characteristics of the landing field, as it was consistent in its application: make an approach aligned straight down the runway at a given altitude and relatively high speed in the direction for landing, then on crossing the near threshold power down and execute a tight, somewhat high-g 360 so as to be at landing speed right at the approach end of the runway for touchdown just at the completion of that turning maneuver.

The problem was that “flameout” of one or both engines began to occur when this acrobatic type of approach was used. Amazingly there had been no accidents, I guess because the nature of the maneuver itself assured the ability to complete the landing almost as intended. And then, it was found, the engine, or engines, could be restarted and operate normally. But even if not so life threatening in this situation, suddenly occurring flameouts when the engines had just been working well were still worrisome.

Meanwhile, back at the plant, I was the Chief Technical Engineer and was supposed to correct such potentially dangerous problems. I used profiling logic, by asking what things are necessary for an engine to continue running. Two things were immediately obvious to me–air and fuel. The age of the airplanes never entered my mind, since even the first delivered were just babies compared to the sort of 20-year life we expected the model to have (and some of them actually were in service for over 50 years).

So the things I wanted the field people to investigate were the air inlet and ducts, and the fuel system components of those airplanes that had exhibited the problem and those that hadn’t. The air inlet system was seemingly more problematic because the units hadn’t been used long enough to expect wear to have occurred, but damage might have, and then the acrobatic nature of the fighter approach itself might have sometimes caused unusual pitch and yaw angles that even a pretty well maintained air inlet couldn’t cope with.

J69 engine
The weak link in the J69 engine turned out to be rubber bushings.

The field folks couldn’t find any differences in the inlet and fuel systems of the “problem” airplanes, and the instructors reported that no unusual flight attitudes had been experienced during approaches when the flameouts occurred. Again, back at the plant, for confirmation we tried to simulate–in flight–excessive yaw and pitch attitudes and related airflow angles that might be induced in an excessively done fighter approach, but didn’t find any indication that intake flow, or engine operation, were adversely affected. So, unhelpfully, we had eliminated, by profiling, what I thought were the two most likely airplane “systems” that could contribute to the problem.

But the persistent field folks continued with their own more general profiling and found that, quite unexpectedly, airplanes with more service time really were more likely to experience the flameouts, so they did more investigation inside the engine compartments of those planes. What they found was that those stiff and hard–and large–rubber bushings we selected for use in the engine mounts were no longer hard and stiff in the older airplanes. It was reasoned that, of course, the temperature inside the engine compartments was fairly high, so over time the bushings softened. What happened then was that with high g’s and soft bushings the engines moved within their mounts, but the power controls were anchored to the airframe. Thus the engine movement essentially caused the controls to reach shutdown position on the engine and, so to speak, made the engines literally turn themselves off!

One simple and expedient solution was obvious, and implemented–even harder rubber bushings and periodic replacement. To my knowledge, that flameout problem never occurred again with credit due to the continued good problem profiling.

5 Comments

  • Harry:

    A good story, as always. It immediately reminded me of the very recent problem with engine “flexing” in the F-35. Unfortunately, that fix involves a lot more than some higher-durometer bushings…

    Always a pleasure to read your stuff, Harry.

  • The response to Paula about a blog was mine. I forgot to identify myself. And Tom Yarsley is undoubtedly right that the solution to the F-35 flex problem will not be as simple as ours was on the T-37.

  • Thank you for an informative article. Well stated, and explained. Describes the complexities of actually finding, then diagnosing, and finally, fixing a problem the engineers didn’t foresee. At the same time, it demonstrates the need for being prepared for any emergency while in flight.

Leave a Reply

Your email address will not be published. Required fields are marked *