During the summe of 2010, my homebuilt, Melmoth 2, which had been flying for almost eight years, began to display a mystifying symptom. As I turned off the runway after landing, the main landing gear oleo on the outside of the turn would seem to collapse. The wing on that side would slump toward the ground, and because Melmoth’s wings have relatively little dihedral and are completely wet from root to tip, fuel would sometimes even dribble out of the tank vent at the wingtip. Once or twice the local controller at my home field told me that I had a flat tire, but after seeing me taxi off listing drunkenly to starboard a few times, he must have decided that nobody could have that many flat tires.
But here was the bizarre thing: As I continued to taxi, the strut would slowly return to its normal height.
When I returned to my hangar I would bounce the wings up and down. The struts felt and looked normal, and would settle at their proper extension. There was no sign of leakage, either of hydraulic fluid or of air.
Now, an oleo strut — at least the kind of strut likely to be found on a light airplane — is a very simple thing. It should not harbor mysteries. It consists, basically, of a cylindrical chamber — the visible external shell of the strut, inside which a piston (the chrome-plated shaft to which the fork and tire are attached) slides up and down — sealed by an O-ring.
Within the cylinder is a quantity of hydraulic oil and some empty space filled with air. As the piston moves up into the cylinder, it displaces the hydraulic oil and reduces the volume of the air. The oil is incompressible, but the air is not, and it acts as a spring. The air pressure in the cylinder at rest is whatever is required to support the airplane. In the case of my struts, the piston is 1.75 inches in diameter and each main gear leg supports a static load of about 600 pounds. The cross-sectional area of the piston being 2.4 square inches, the static pressure in the strut has to be 600/2.4, or 250 pounds per square inch.
If you make a very hard landing, the piston is driven far up into the cylinder and the air is squeezed to a much higher pressure. The piston then rebounds, and might toss the airplane back into the air were it not that the displaced hydraulic fluid has to pass, in both directions, through a small hole. This orifice, as it is called, produces frictional resistance, so that the piston’s movement is damped: It does not bounce back with as much energy as drove it into the cylinder in the first place.
Now, taxiing does not subject an oleo strut to large loads, and a gentle turn off a runway doesn’t either. So why was my strut first compressing and then gradually re-extending itself? In fact, how could it overcompress at all, since in order to compress it would be fighting against ever-increasing air pressure? It seemed physically impossible, and yet there it was — it happened again and again. And it happened only just after landing; it would never happen while taxiing before takeoff, and would never happen a second time once the strut had regained its proper height.
I discussed this conundrum with a number of people who had extensive aviation experience, including pilots, mechanics and engineers. One of them had even designed oleo struts professionally and had written quite a complex spreadsheet for oleo sizing. Nobody could explain what might be going on. Eventually, I took the right main strut out of the airplane and dismantled it, expecting that I would find something highly peculiar inside. Instead I found nothing unusual, but I did realize something about my struts that I had forgotten.
When I built this airplane, I used a pair of Cherokee Arrow main gear struts, which I modified considerably. One change I made was to move the filling port, originally a couple of inches below the top of the cylinder, up to the very top. The procedure for filling the struts with the proper amount of fluid depended, however, on the opening being where it originally was. Over the 20 years since I had assembled the struts I had forgotten this, and had at one point added too much fluid. Could this be the problem? It didn’t make much sense — I thought reduced air volume ought to make the gear stiffer, not softer — but I had high hopes that somehow or other this error might hold the key to the problem.
No such luck. The first time I flew after overhauling the strut and filling it correctly, it slumped the same as before.
In January of this year I e-mailed an aeronautical-engineer friend to ask whether he knew any oleo experts. He gave me a name, and a few minutes later sent me a link to a discussion on a Piper owners’ forum of precisely my problem. Why hadn’t I ever thought, as he immediately had, to Google “Piper Cherokee oleo strut sags”? But there it was. I wasn’t the only one; others had the same problem.
The discussion went on for a couple of pages. A lot of it was off the point, with irrelevant suggestions about replacing this or that and checking air pressure and fluid level and so on, but one person said, in so many words, “It’s not that your strut is sagging; it’s that the other one is sticking.”
Elementary, my dear Watson.
I wish I could say that the clouds parted in an instant and heavenly choirs gave voice, but the subjective impression of one strut going down — “sagging” — had been so powerful, and I had thought about the phenomenon in those terms for so long, that it took a few minutes for the fog to lift from my brain and for me to see that an opposite strut remaining extended, while the other settled to its proper static height, would produce the same sensations in the pilot, and would not violate any laws of physics in the process.
A week after this revelation had sunk in, I attached a small video camera to the retractable boarding step on the left side of the airplane and made a trip around the pattern.
Studying the video later, I could see what was happening. The full extension of the piston is about eight inches. On touching down, it slowly compressed to about four to five inches as weight settled on it, and stayed there as I turned off the runway. Only a minute later, as I made a right turn from the parallel taxiway, did the left strut finally settle all the way down to its static height.
So the mystery was elucidated — partially. The struts were not settling evenly. But why had this begun to happen only after Melmoth 2 had been flying for nearly eight years?
What had changed, I realized, was that in 2010 I had finally gotten around to actuating my big Fowler flaps and was using them on every landing. They reduced the touchdown speed from 68 knots to 53. But they also produced a great deal of residual lift during the landing rollout. That reduced braking effectiveness, and in order to enhance it I had gotten into the habit of pushing the stick forward to lessen the angle of attack; but lifting the tail may have canceled the benefit. At the same time, I was braking heavily in order to make the first turnoff; this was just a matter of pride, but it had the effect of increasing the friction between the pistons and the cylinders in the oleo struts.
It seems that what was happening was simply that the weight of the airplane was settling so gradually onto the landing gear that the struts remained excessively extended until I turned off the runway, at which point the one on the outside of the turn would compress to its static height. Because the gear track is relatively narrow — 82 inches — a difference of two or three inches in strut extension could make a large difference in the angle of the wings, and this could be exaggerated by fuel shifting in each bay of the 15-foot-long tanks.
There were some life lessons here. The way you describe a problem — in this case, my habitually saying “sag” or “collapse” rather than “unequal height” — conditions the answers that occur to you and that people you consult are likely to think of. Subjective impressions may have multiple explanations, but favoring one blinds you to the others. I’m not as smart as I thought. Google first, ask questions later.
The oleo expert to whom my engineer friend had referred me animadverted at length upon the Arrow gear. It is, he said, too heavily damped in compression, and so does not settle to its static height as readily as it should.
His solution? Land harder!
