- Propeller Vibration -
Some Critical Information About Propeller Safety
There are two primary reasons why it is important to suppress engine torque pulses. The first (discussed in PSRU Technology) is to reduce the fatigue loadings on the gearbox components.
However, metal propeller blades are especially unforgiving of being excited near a resonant frequency. So the more important reason to suppress engine torsional excitation is to eliminate the pulse excitations fed through the gearbox, multiplied by the gear ratio, and applied to the propeller blades.
Each prop blade has more than one resonant frequency. The frequency excited by thrust vibrations is different (typically) from the one excited by torsional vibrations.
Metal prop blades are especially susceptible to destructive vibration if they are excited near a resonant frequency. One reason is that a metal blade contains very little internal damping, so resonant vibrations can build in amplitude very quickly. Another reason is that the blade material (typically aluminum) HAS NO ENDURANCE LIMIT in a fatigue environment. Therefore, the fact that an aluminum part has run for x-number of fatigue cycles is NO GUARANTEE that it will not fail during the next hundred similar cycles.
Manufacturers of certified props go through extensive analysis and testing to be sure that a particular prop will survive the fatigue environment produced by a particular engine on a particular airframe. Get it wrong, and blade pieces will be departing the aircraft. It’s just a matter of time.
In addition to being susceptible to engine vibration, a propeller produces torsional excitation which varies with rotational and translational speed, flight attitude, airframe characteristics, and the specific properties of the engine torsional excitation which are applied to the propeller.
For purposes of illustration, consider just one mode of propeller-blade vibration. Picture yourself in the pilot seat, looking forward over the cowl watching the rotating propeller, being driven by your 4-cylinder Lycoming. (For the purposes of this experiment, you will have acquired very-high-speed vision, which enables you to see fast-moving things in slow motion.)
In normal vision mode, you see the prop blades rotating clockwise across your field of vision. Now, switch into high-speed-vision mode. Here is the slow-motion picture you see: A prop blade rises into view just as one cylinder of the engine begins it’s compression stroke (remember the instantaneous torque curve for a 4-cylinder engine in the TORSIONAL EXCITATION BY PISTON ENGINES section). As that progresses, the instantaneous torque is at its minimum (actually negative in a 4-cylinder engine). Since the torque is minimized, the engine decelerates slightly and is not keeping up with the rotation of the prop. The blade, being a heavy, fast-moving thing, tries to maintain its speed, but the prop hub is slowing down. The blade, also being a springy thing, deflects clockwise in a nice curve as the result of the blade momentum being opposed by the decelerating hub.
Now, just as the blade reaches it’s maximum clockwise deflection, the cylinder fires and the crankshaft torque quickly reaches it’s maximum positive value, and the crankshaft now tries to accelerate the prop hub, which in turn, tries to accelerate the blade we are watching. The blade responds by starting to bend in the opposite direction (counterclockwise), but it was already in motion, springing back from the clockwise deflection we just saw. That motion, added to the motion induced by the new pulse, causes this blade deflection to be greater that the one which occurred last rotation.
This is an example of blade excitation at a frequency too close to resonance. Each cycle imparts energy into the system, and the spring-mass properties of the blades stores some of the energy and adds it back into the system in phase with the engine. The forces increase with each cycle. The deflections continue to increase as a result of the energy added during each cycle, until something breaks.
There are more complex modes of propeller blade vibration, but this simple one serves to illustrate the point.
The more likely scenario is that high blade stresses induced by vibratory bending near resonance exceed the endurance limit of the blade material, and the fatigue cycles pile up, and at some point, a piece of blade departs. That’s why certain direct-drive aircraft engines have a band (2200-2400 RPM, for example) where continuous operation is not allowed.
If a large enough section of blade breaks off, the resulting vibration could easily tear the engine off its mounts. Consider how the resulting CG might alter the rest of your day.
The propulsion systems of certificated aircraft are approved as a powerplant/propeller/airframe combination. Torsional interactions between engines and propellers are subject to close engineering scrutiny. In order for a new propeller to be approved for a certified aircraft, it must be shown that the engine-propeller-airframe combination does not experience dangerous levels of stress from vibration. That is accomplished by design and development engineering, and verified by an extensive vibration survey, performed in flight, using real-time data acquisition equipment.
Metal-blade propellers are especially critical because they have very little internal damping and closely resemble perfect springs. Wood and composite propellers have varying degrees of internal damping, so they tend to be more tolerant of torsional excitation. The point is this: each system is different, and needs to be investigated.
The need for such scrutiny is illustrated by a few examples.
- A certain propeller which survives quite well on a Continental IO-520 (6-cylinder horizontally opposed, 520 cubic inches, 554 lb.-ft. mean torque, torsionally counterweighted) has unacceptable vibration problems on a Lycoming IO-540 (6-cylinder horizontally opposed, 540 cubic inches, 585 lb.-ft. mean torque, torsionally counterweighted).
- A recent Hartzell vibration survey revealed an interesting situation. There is a certain Hartzell propeller which is certified on a Lycoming IO-360-A3B6D for a specific aircraft. This engine is rated at 200HP, and is equipped with 8.7:1 compression ratio pistons (LW-10207-S) and 6th and 8th order torsional counterweights. On that engine, the propeller blade stresses were acceptable.
However, there is an STC for that aircraft which allows the installation of the Lycoming 10:1 pistons (LW-11487-S from the HIO-360-D1A) to increase the engine power output. Hartzell discovered that the installation of those STC’d pistons significantly altered the torsional characteristics of the IO-360-A engine, to the point that, on that same propeller, it drove the blade root stresses beyond the endurance limit of the material. That engine, with its pendulous torsional absorber counterweights, might be expected to be "torsionally tolerant" (for a 4-cylinder engine). The fact that a relatively small change in engine configuration caused such a significant change in vibratory stress gives some insight into just how critical this engine/propeller interaction can be.
While we are on the subject of propellers, an extremely dangerous practice in the experimental community is that of shortening the blades of a given propeller to fit a new situation. For example, a certain popular certificated Hartzell propeller has a minimum diameter of 72". If the blades are shortened below that limit, the resonant frequency of the blades will be increased to the point that engine torsionals can excite vibratory stresses which exceed the endurance limit of the blade alloy. What that means is that sooner or later, one of those blades is going to shed a piece, which could easily tear the entire engine off the airframe. Again, consider how the resulting CG might alter the rest of your day.
We have read reports on conversion projects in which propellers from turboprop applications have been shortened and attached to a V8-with-PSRU. That should give someone nightmares, since the vibratory modes which a turboprop propeller is designed to handle are significantly different from those generated by a high-output piston engine, and the PSRU's used in these projects had no torsional pulse suppression characteristics whatsoever.
There are many complex modes of propeller blade vibration. These examples are given to emphasize the fact that making a propeller survive on a piston engine is a challenging task, demanding very specialized skills and equipment. The intent here is to discourage the practice of willy-nilly mix-&-match attachment of propellers to engines, with the compatibility of the bolt pattern being the only criterion.
A deeper insight into propeller considerations is presented in Sport Aviation Magazine (July, 1994) by Brian Meyer, a propeller engineer. (Access to the Sport Aviation online archive is restricted to EAA members.)
For more information, visit the Hartzell website dedicated to technical issues relating to propeller selection for experimental airplanes.