- PSRU's and Torsional Excitation -
A summary of the disasterous problems it causes
One of the most poorly understood requirements of a propeller reduction gearbox is the damaging effect that the torsional excitation produced by piston engines can have on gears, shafts and propellers. If you are unfamiliar with piston engine excitation, it would be extremely helpful to read about it before continuing with this page.
The engineering literature is rich with information on the subject of propeller reduction gearboxes. In Volume 2 of his renowned reference work The Internal Combustion Engine in Theory and Practice, (ref-5:3:280) C. F. Taylor says this about PSRU’s:
"Particularly complex is a vibration system consisting of an engine with an air propeller mounted on its output shaft, since the propeller blades usually have important modes of vibration which tie in with crankshaft torsional vibration."
An American Gear Manufacturer's Association (AGMA) publication summarizes the problem as follows:
"The gearbox is one component of a system comprised of a power source, gearbox, driven equipment, and interconnecting shafts and couplings. The dynamic response of this system depends on the distribution of the masses, stiffnesses, and damping. In certain cases, a system may contain a torsional natural frequency close to an excitation frequency associated with an operating speed. Under these resonant conditions, the dynamic gear tooth loads may be very high, and operation near a system resonance is to be avoided.
Clearly, a PSRU cannot be treated as an isolated entity. It is one critical component in a vibrating system having at least two natural frequencies (in addition to the internal natural frequencies of major engine components such as the crankshaft system) and two major sources of excitation (engine and propeller). Any change to any part of the system can significantly alter the loadings imposed on the other parts of the system.
Here is a quick review of the torsional excitation a piston engine produces. The instantaneous torque output of an even-fire 8-cylinder engine looks like this at the crankshaft flange:
That graph shows the torque "peaks" to be nearly double the mean torque of the engine and the torque "valleys" to be roughly 10% of the mean torque of the engine. Those peaks and valleys occur four times per crankshaft revolution. As an example, one of our aircraft V8 engines produces a mean torque (measured on a dyno) of 625 lb.-ft. However, at that mean torque value, the instantaneous torque peaks at the crankshaft flange are about 1235 lb.-ft. and the instantaneous torque valleys are about 70 lb.-ft. That is a torque fluctuation of about 1165 lb.-ft. occurring four times per crankshaft revolution.
Engines having fewer than 8 cylinders and any engine with an uneven firing interval will produce excitations significantly worse than those shown for the 8-cylinder engine (CLICK HERE for a detailed presentation on that subject). On the other hand, the torsional excitation amplitude of an even-fire 10 or 12-cylinder engine is lower than that of an 8 cylinder engine, but occurs more frequently (5 or 6 times per revolution respectively).
For example, an even-fire V12 engine which produces 625 lb.-ft. of mean torque will generate torque peaks of 875 lb.-ft. and torque valleys of 375 lb.-ft., which is a torque fluctuation of 500 lb.-ft. occurring six times per revolution. However, the lower amplitude does not diminish the severity of the problem, and in fact, the higher frequency can exacerbate it.
Here are the four major problems caused by engine torsional pulsing.
- The torque peaks which the engine produces are applied to the gear teeth (or chain, or whatever), causing a substantial increase in fatigue loads (both the cyclic bending stress and the cyclic hertzian stress) the gears (or chain or whatever) experience. That can cause gear teeth (or chains or whatever) to break prematurely or to pit and shed metal into the oil system prematurely.
- In addition to the spikes mentioned above, a coupling system which does not eliminate the torsional excitation has properties which increase the dynamic load on the gears dramatically. If you are at all interested in propeller gearboxes, you should be familiar with the dynamic loads on gears. A gearbox which theoretically should carry a given torque but which has a dynamic gear load problem in the system can break apparently adequate gear teeth in a very short time.
- If the torsional excitation from the engine is anywhere near one of the engine-gearbox-propeller resonant frequencies, the forces applied to the gears, shafts and propeller will be amplified, not simply by the value of the torque excursion, but potentially by many times the peak values. That is not in intuitive fact, so to comprehend the gravity of that situation, you must understand the concept of transmissibility.
- If engine torsional excitation is not eliminated, the torque pulses will be amplified by the gear ratio and applied to the prop hub and blades. Propellers, especially metal propellers, are quite unforgiving of torsional excitations which are near any of the blade resonant frequencies. A prop subjected to torsional excitation can easily shed a blade or a portion of a blade, followed almost immediately by the whole powerplant. This subject is covered in detail in the propeller vibration section.
Please do not be deluded into thinking that just because you can't FEEL any vibration, then none exists. Nothing could be further from the truth. In fact, many destructive vibrational forces occur at frequencies far beyond what the human body can detect.
It is also essential to avoid the delusion that the torsional absorber (aka "Harmonic Balancer") which is commonly found on the nose of long crankshafts will solve these problems. The Harmonic Balancer is there solely to subdue the amplitude of the destructive torsional deflections which occur at the nose (free end) of the crank when operating anywhere near the crankshaft resonant frequency or a small integer-multiple of it. Please understand that the Harmonic Balancer on the nose of the crankshaft MUST be there to keep the crankshaft in one piece, and its function HAS NOTHING WHATEVER to do with the torsional signature of the engine output. The effect of the Harmonic Balancer on the output end of the crankshaft is essentially insignificant. This subject is covered in detail in the crankshaft torsional absorber section of this site.
What can be done about these seemingly devastating problems? EPI has developed a very effective solution. EPI propeller gearboxes include, as part of the system, a proprietary engine coupling system, which typically consists of an engine flywheel (including the starter gear), a splined hub, and splined shafts connecting to the gearbox. We will not sell a gearbox without that system. The EPI coupling system removes over 99% of all engine torsional pulses which the engine produces at takeoff and cruise power. That makes the input to the gearbox nearly as smooth as that of a turbine, and provides the propeller with an operating environment which is essentially devoid of engine torsional excitation.
How? The mathematics which define the system make it abundantly clear that, if your goal is to minimize the engine torsional vibration transmitted to the gearbox (and thus to the propeller), then you must minimize the system transmissibility. We design the components, taking into account the properties of the engine, coupling, gears, shafts and propeller, so that the transmissibility between the engine and gearbox is less than 1%. We do that (a) by making the first mode torsional resonant frequency of the system at least four times less than the engine excitation in the operating range, and (b) by minimizing the damping. The details of this technology are presented in the transmissibility section of this site.
The effectiveness of our coupling system system is demonstrated by the fact that our Mark-9 gearbox can carry a continuous input from a V8 engine of 600 lb.-ft. of torque with a safety factor against gear tooth breakage over 2.0 and an expected wear life in excess of 10,000 hours, yet the driving gear in that gearbox has a pitch diameter of only 3.5 inches and a face width of only 1.5 inches.
Contrast that with the fact that other gearboxes which CLAIM a similar capacity, are known to break gears having pitch diameters of 4.4 inches and more, and face widths of 2.25 inches. (Of course, they don't have an effective torsional absorber between the engine and the PSRU.)
The simple bottom line is:
- Don't even think about using a flywheel with a low mass moment of inertia, and
- Don't even think about using a torsionally-rigid connection between the engine and gearbox.
There is a recent PSRU vendor who has cobbled together some off-the-shelf automotive parts and added a CrowerGlide ™ centrifugal clutch to the mix.
This vendor claims that, by some unexplained magic, the centrifugal clutch eliminates torsional vibration problems. What it actually does is add weight and complexity, and deludes the user into a believing he has actually solved the problem. Tragically, this same vendor recently suffered a failure of his product that was fatal.
Based on pictures of this drive, we would really like to see the calculations which show how the centrifugal clutch "...Resolves Torsional Harmonics..." (their words). Taking that one step further, it would also be revealing to see the calculations showing just how much combined torque, thrust, g-load and gyroscopic moment loading the propshaft and housing can actually withstand.
We sincerely hope these PSRU claims are based on more than intuition and the ownership of a CNC mill.