Had probably my strangest crash on Sunday - was flying my speed 400 sport plane on a low pass when there was a fairly loud bang, followed immediately by the motor ejecting itself from the nose!!!

...Probably the most telling factor was that the Graupner 6x3 folder had shed a blade.

Although it's possible the prop simply broke, that doesn't really match up with the events. It was also only the second flight since I'd last inspected the prop. The best guess so far is that the prop hit something, maybe a bee or other insect...

From : Don Stackhouse

Chris, I doubt it was a bug strike. In my "previous lifetime" as an engineer for full scale props, I got to see quite a bit of the results of birdstrike tests. In general, the loads at the blade roots are pretty low because of the "centrifugal stiffening" effect. Most of the loads and deflections show up in the mid-blade area. While the mass distribution of a model propeller would probably allow more load to reach the blade root, the relatively flexible retention system would probably take that out before it caused a blade failure. In addition, the probability of a bug actually hitting a blade while passing through the prop disc is VERY low. If you look at the ratio of the frontal area of the blade (i.e.: the blade THICKNESS, NOT chord, times the blade span) to the swept area of the disc, that's about the probability involved, and that assumes the bug goes through the disc to begin with, a very low probability in itself!

When we were originally developing certification test methods for composite propeller blades, we went through all these numbers, plus consulting with the engine companies regarding their experience with birdstrike tests. It was pretty obvious that we could fire a hundred or so chickens through a spinning prop (at a cost of several thousand dollars per shot) and not have a single blade strike to show for our efforts. We opted to go with a much more severe (because of the lack of centrifugal stiffening to help limit deflections) lab test of a stationary blade. Even so, the resulting loads on the blade roots were minimal.

The dive you put the model in before the failure would get the airspeed fairly high, which would drive the prop RPM up quite a bit. Centrifugal force is proportional to the SQUARE of the RPM, so the CF on the blades would be much higher than normal. Thrust and torque loads form the high power setting would tend to increase this static load, plus concentrate it on the face side of the blade root. In addition, the pull-out from your shallow dive at the entry into the low pass would create an angle of attack on the prop disc, which generates once-per-rev bending stresses. Your blade roots would now be subjected to a static overload (due to the high CF plus the high power setting) plus a high vibratory stress (due to the 1 per-rev), a perfect scenario for rapid growth of a fatigue crack. It's also possible (though doubtful in this case), if your rpm was high enough, that the crack in question could have initiated and grown to a fatal size all in that one high speed pass. Besides flight loads, such a crack could be intiated by snagging the blade on something and bending it forward during handling. Did you perhaps catch that blade on the carpet while pulling the fuselage out of the car?

Those folding blades have a major stress concentration on the face side of the blade in the crook of the elbow at their roots (caused by bending stresses from CF), and another nasty one around the bolt hole. Either spot would be a prime location for a failure. The inspection you described in your subsequent post would probably not be likely to find a hairline fatigue crack in a blade root. The beginnings of such a failure could very well be invisible to the naked eye, and might not even be visible under a 10X magnifying glass. This is especially true of a non-metal blade on an electric model, since there would not be any of the oxidized metal particles and oil residue that sometimes highlights such cracks on full-scale props.

The CF on a propeller blade is incredibly high. The centrifugal force on a full scale metal propeller blade for something like a Beech King Air or a Pitts Special (because the rpms are higher for the smaller props, the CF seems to be about the same regardless of size) is typically around 40,000 to 50,000 pounds! Composite full scale props are somewhat lower because of the lighter blades, typically about 25,000 pounds. In any case, the loss of a blade almost always immediately results in either folding the engine mount about 90 degrees or the total departure of the engine from the airframe.

I don't have the numbers handy for a model propeller, but as soon as I can work it in I'll try measuring a 6-3 folder blade and calculate some loads and stresses. If I can find the time (and that might be quite a while), I can try running some finite element models of the blades and calculate the stresses around those critical areas in the blade roots. I've thought about doing this before, in fact it would make a nice magazine article, with color plots of the stress patterns, etc. I think all of you would find the results quite amazing, perhaps even alarming. Maybe after I finish getting the Speed 600 electric 2-meter Chrysalis into production...

Don Stackhouse
DJ Aerotech