Emergency: Resolution

This is the final part of the Emergency series (part 6 of 5). After examination and testing of the propeller pitch control mechanism, I am confident that I know the cause if the propeller/engine failure.

GUFY's pieces were collected and removed to Red Lake. The salvage operator did some preliminary examination at my request. He confirmed:
The engine was not seized. It could be moved part of a revolution before being blocked.
The propeller pitch could be changed by hand.
The pitch control linkage was intact from cockpit through the bellcrank and to the forked rod that attaches to a bearing connected to the spinning propeller hub.

My working hypotheses was that something had failed in the pitch control mechanism that allowed both blades to feather together. If only one bladed feathered, that would set up an intense vibration. That never happened.

The forked rod goes through a hollow gear in the engine to the propeller hub. I had figured that the forked end had lost its attachment screw to the bellcrank. But it was still attached.

The forked rod attaches to the propeller hub through a ball bearing that has the prop hub shaft attached by a screw to the inner race of the bearing. I knew that screw could not fall out because the other end of the forked shaft allowed very little movement before being stopped.

So I asked the salvage operator to ship the propeller hub and linkage mechanism to me for further analysis. After sending him removal instructions he extracted the propeller hub easily.

But he forgot to perform the first step: remove the bolt attaching the forked shaft to the bellcrank. That step was necessary to allow the forked shaft to pass through the hollow gear. The shaft stayed behind and the hub easily came out since the screw attaching the hub to the bearing was completely unscrewed!

Aha. Although the head of the screw was prevented from unscrewing at its end, there was nothing preventing the propeller from unscrewing itself from the other end. Technically the bolt was specified to use blue Loctite to keep it in place. Blue Loctite was used but apparently failed.

After receiving the parts, I confirmed the inner race of the bearing still turned freely.

Ironically, to reconnect that screw to the prop hub requires removal of the forked rod first to gain access to the hex head of screw. The screw could be fully reattached to the prop hub.

To test if the prop hub could unscrew itself I separated the two parts and then turned the screw one turn before remounting the forked rod. After spinning the prop around the stationary bearing and forked rod, the attachment did loosen until the propeller hub finally separated and fell on the ground. I repeated the test with similar results.

Although it took some time for this separation to happen, at cruise speed there is a relative motion of approximately 40 revolutions per second which is considerably faster than my hand twirling of the prop.

When did the screw first become loose? Before the trip the propeller and hub had been removed for its 600 hour maintenance. The forked rod attachment to the bearing was verified to be tight at that time. Before the first flight of each day, the propeller blades are cycled by hand from fine pitch to feathered. As part of each preflight, with the engine operating, pitch is moved from fine to coarse and then back. When transitioning from climb to cruise flight pitch is again changed to coarse.

Engine load increases significantly for small pitch changes when the pitch is coarse (cruise mode). That was not observed on the final flight. So a gradual loosening of the screw during flight over even a few seconds did not occur.

Apparently a sudden removal of the screw enabled the propeller pitch to be changed. The only force available to change the pitch at that point was aerodynamic force on the propeller blades. It is equivalent to having the tail suddenly removed from a aircraft. This is an unstable situation where, depending on the angle of attack of the wing at that time, the aircraft will either immediately pitch up or pitch down. Similarly the propeller will either go to minimum pitch or fully feathered. Fine pitch would have been preferred: the throttle would have to be reduced and I would have to fly slower. Evidently my propeller slammed into full feather, the sudden load abruptly stopping the engine.

All aircraft using the Pipistrel Vario propeller could be susceptible to this unintended feathering failure. That includes the 80 hp Pipistrel Sinus and Virus series. The 100 hp versions use Woodcomp propellers that do not feather. Presumably they are not susceptible to this kind of failure.

The control linkage must be removed to change the blades, balance the prop or change the base pitch of the blades. I have changed blades five times and tweaked pitch more often than that.

So the screw needs to be removable when wanted. Looking at the design of the blade pitch control mechanism, space restrictions limit modifications to prevent unwanted screw loosening. Blue Loctite helps. One easy change would be to use left hand threads on the screw. Then propeller rotation would tend to retighten the screw rather than fling it away.

The pictures show the afflicted mechanism. The first is a shot of the screw in the bearing taken at Red Lake just after the propeller was (surprisingly) removed. The threads of the self-unscrewing machine screw are visible in the centre of the bearing. Note that the seal for the bearing is missing, along with a few balls. Apparently the suddenly unattached prop hub shaft rattled around a bit before exiting to its feathered location, banging into the press-on seal.

Curiously the seal was trapped in the engine prop mechanism. I received it with the rest of the prop hub assembly. The second picture shows the bearing after I popped the seal back in place. The third picture shows all the components: prop hub, hub extension, bearing case, forked rod, and the 'fork flange bolt'. The fourth picture shows the inside of the hub, including the mechanism that controls the pitch of each blade. All is normal here.

So what's next. I will forward my pics and thoughts to Transport Canada to add to their accident file as well as to Pipistrel and the Yahoo Pipistrel group.

As for me, I am recovering well from my broken leg. The impact broke my tibia near the ankle. They put a titanium tube down the centre of the bone and screwed each end into the bone. The fibular was more of a mess. It broke in three places, once near the top and 2 more places near the bottom. It was left to heal on its own.

Radiography (I do not like the term X ray when I am involved) were taken just before two months had elapsed since my operation. The tibia is mending well and I was surprised to see bone growth had bridged all the gaps in my fibular. I have been walking without a cane for some time, my limp gradually fading.

The bone doc said I will be able to do anything I wanted with my leg. I am happy to now being able to run marathons. Couldn't before.Read more