Technical Articles and Product Descriptions

Reference Materials

Additional Products

Race Engine Technology Magazine

Perhaps the most important point to be taken from this presentation is that PSRU’s, propellers, aircraft engines and aircraft engine mounts are very complex devices which, although they look simple, have mostly evolved to current reliability on the back of many disasters. Uncertified PSRU’s are the incarnation of EXPERIMENTAL. You should keep in mind that there is no significant body of accumulated experience behind most of them.

It should be clear to anyone contemplating the use of an engine conversion which uses a PSRU that the PSRU is one of THE most critical components, upon which their ultimate safety depends. PSRU’s appear to be quite simple, but they actually involve several complex subjects, some of which seem to have been overlooked in certain PSRU products.

This presentation will describe several of the critical concepts involved in PSRU design, with special emphasis placed on the methodology which EPI used to solve those problems. The EPI implementations are not necessarily the only solutions, but the problems are common to all. The intent is to provide the potential buyer of a PSRU with enough understanding to ask the right questions, detect the wrong answers, and to make a wise (SAFE) selection.

Before beginning on this presentation, it will be helpful if you fully understand the relationship between TORQUE and HORSEPOWER. We suggest that you visit the EXPLANATION of POWER and TORQUE page which gives a brief clarification of those often misunderstood terms.

WHY IS A PSRU NEEDED??

The simple answer is: to mate an engine whose crankshaft runs at a high speed to the slower speed required by most propellers.

But why must the propeller turn slower than the crankshaft? There are really two questions there: (1) Why run the crankshaft faster? And (2) Why limit the propeller speed?

Let’s examine question 1 first. The second-most important property of an aircraft engine is specific power, the number of horsepower per pound of engine weight. (The most important property is clearly reliability.)

Generally speaking, it takes lots of cubic inches of engine displacement to produce the torque necessary for high power at low crankshaft speed. (Remember that POWER is "work done per unit time", and HORSEPOWER is TORQUE times RPM divided by 5252, as shown in EXPLANATION OF POWER and TORQUE).

And (generally speaking) adding displacement adds weight.

So the obvious answer to the problem of extracting more power from each cubic inch of engine displacement (assuming roughly equivalent limits to BMEP) is to change the design of the engine so it produces its peak torque at a higher RPM, thus moving the maximum output range to a higher RPM and increasing the power output for the same torque.

For example, a normally-aspirated Continental IO-520 (520 cubic inches) produces a max of about 285 HP at 2700 RPM. Turbocharged, the same engine reaches it’s heat-rejection and internal stress limits at about 325 HP at 2700 RPM (although certain versions are rated at 350HP). But the specific power (HP per pound) remains about the same.

However, by increasing the crankshaft speed to 3300 RPM with roughly the same torque output, Continental increased the power output of the GTSIO-520 to a max of 435 HP. That required the addition of a gearbox to reduce the prop RPM to 2200, and the added weight actually decreased the specific power (and increased the heat rejection load on an already strained design).

Taking it one step further, the EPI Gen-1 liquid-cooled V8 aircraft engine produces 500 HP at 4600 RPM (both turbocharged and normally aspirated). That powerplant includes a PSRU, and weighs considerably less than a GTSIO-520. (see AIRCRAFT ENGINE DEVELOPMENT).

All turboprop engines have PSRU’s, because the turbine sections turn at very high RPM. For example, the Garrett TPE-331 engines produce from 500 to well over 750 HP from a turbine section which turns at 41,000 RPM while the prop turns at 1591 RPM.

But why can’t the props just run faster? The propeller limitations arise from a variety of factors including blade and hub stress levels and blade efficiency (SEE PROPELLER TECHNOLOGY for more information). In fact, the 2700 RPM limit familiar to most general aviation pilots is really a compromise worked out to accommodate engines which have the propeller attached directly to the crankshaft. That compromise allows reasonably high engine speeds in order to achieve better specific power (HP per pound), while producing reasonable efficiency numbers at the relatively low airspeeds achieved by most aircraft using those engines.

The peak efficiency for contemporary subsonic blades usually occurs at a tip velocity of approximately mach 0.85 - 0.87. Note that tip velocity is not just the rotation speed. It is the helical velocity of the prop, which is the vector sum of rotational and translational velocities.

As the altitude of a propeller-driven aircraft increases, the propeller RPM for max efficiency decreases, and can become quite a bit lower than it was for sea-level takeoff.

There are two reasons for that. First, the speed of sound (Mach 1.0) decreases with temperature, and air temperature (usually) decreases as altitude increases. So for a constant tip velocity, a lower temperature produces a higher tip Mach number.

Second, as density-altitude increases, airframe drag decreases and true airspeed increases. If propeller RPM is held constant, the velocity of the prop tip increases as TAS increases. The effect is more noticeable with engines which can maintain a constant power output with increasing altitude, but is also a factor in normally-aspirated engines.

The problem of prop speed gets a bit worse as we begin to move in the direction of higher power engines. The prop diameter is one parameter which often increases with the power rating of the propeller. Larger diameter propellers must turn at relatively lower RPM to observe tip-speed limits.

Given that quick background, it is clear that if the engine power section is designed to run faster than the prop can efficiently turn, a reduction unit is necessary to turn the prop within it’s range of good efficiency.

For a more detailed explanation of propeller performance, see our Propeller Performance Factors page..