Every RPM reduction gearbox I look at seems to have more parts in it than a Swiss watch. Great for keeping time but many parts to wear and breakdown.

What seems especially worrisome is the direct coupling of the engine drive shaft, through the gear, to the prop. This seems fine when the engine is turning the prop, but potentially damaging if the prop starts turning the engine.

The torque converter, with it hydraulic coupling, seems to offer a safer approach because the prop is not directly connected to the engine power output shaft.

Hydraulic torque converters are commonly used with automobile automatic transmissions. Are any aircraft using them?

As smaller, higher RPM engines become more common, is there a chance they will be a market for them in aviation?

  • 2
    $\begingroup$ That's basically a free-turbine turboshaft engine, where the coupling medium is gasses. $\endgroup$
    – user71659
    Apr 16 at 21:35
  • $\begingroup$ @user71659 So I have heard. Could a piston engine try it? $\endgroup$ Apr 16 at 21:39
  • $\begingroup$ You are looking at mechanical devices that convert the torque of an electric motor into the rotation of a propeller. Is it possible to use the current mechanical system and solve problems (or at least minimize their impact) by developing a special motor control system? $\endgroup$
    – dtn
    Apr 17 at 9:29
  • $\begingroup$ @dtn Flexidynes can be used on any driven system. Your question could be posted as a follow up. $\endgroup$ Apr 17 at 11:48
  • $\begingroup$ "What seems especially worrisome is the direct coupling of the engine drive shaft, through the gear, to the prop". Are you sure about that? At least in the helicopter world there's always a freewheel inbetween. $\endgroup$
    – sophit
    Apr 17 at 13:12

3 Answers 3


Fluid torque converters would fix the torsional resonance problems on gear reduction systems, but they are inefficient (significant power losses) and fairly heavy.

When Molt Taylor designed his flying car, the Taylor Aerocar, which the FAA (nee CAA) certified as an aircraft in the early 50s, it used a Lycoming O-320 for power, and to deal with the long drive shaft, which has similar torsional resonance problems as a gearbox, he went with the Flexidyne Dry Fluid Coupling. They are still widely used in industrial equipment and appliances like washing machines.

The flexidyne is a housing filled with small balls (steel shot) and the output shaft is driven through a wavy plate buried in the ball bearings. When it spins, the ball bearings become semi solid from centrifugal force, with just enough give to create a compliant energy absorbing drive transfer with much more efficiency than a fluid torque converter.

  • $\begingroup$ Yes, indeed! See the Lesher Nomad. $\endgroup$ Apr 17 at 7:52
  • 1
    $\begingroup$ They get hot in this application because the steel shot is constantly being displaced tiny amounts on every power pulse during a resonant phase. I think you still need to mitigate it as much as possible in design (drive shaft stiffness, gear lash tuning etc, so the Flexidyne only has to work through resonant RPM bands you avoid continuous operation in. $\endgroup$
    – John K
    Apr 17 at 12:59

In addition to John K's answer there is one more consideration: the heat the torque converter produces. Cars run at about 15% "power setting" when cruising, airplanes run at about 75%. TC will run hot at these conditions, especially since a TC lockup can not be used since this bypasses the TC.

Cooling would have to be arranged, adding complexity / points of failure, and weight to the system.

  • $\begingroup$ @Jpe61.You are right about the heating issues. Cars use "lockup" to improve fuel economy. But I am sold on looking into the Flexidyne, and I think the Lesher Teal could fly off vs the Rutan Vari-eze. $\endgroup$ Apr 17 at 12:04
  • $\begingroup$ Note that current Flexidyne couplings are not meant to run on constant slippage. They are soft start / overload protection. $\endgroup$
    – Jpe61
    Apr 17 at 12:29
  • 1
    $\begingroup$ @Jpe61 yes IIRC they got quite hot when Taylor used them in the Aerocar and Mini IMP if the resonance problem was severe, so you have to mitigate resonance problems as much as possible and use the Flexidyne to trim off the last bits. $\endgroup$
    – John K
    Apr 17 at 12:54
  • 1
    $\begingroup$ @RobertDiGiovanni On reflection, it seems that you are unaware that hydraulic torque converters, unlike simple fluid couplings, do perform torque amplification, with a corresponding reduction in RPM, like reduction gears, though less efficiently and only over a limited range of ratios. This is accomplished by having a third set of vanes, attached to the stator rather than either the input or output shafts. $\endgroup$
    – sdenham
    May 3 at 2:00
  • 1
    $\begingroup$ Just as @sdenham wrote torque converters by definition alter the torque. A fluid coupling is a different device. When deciding which damping application is best for a specific case, a lot of variables need to be taken into consideration. Transferrable peak torque and torque range, frequency and amplitude of vibration, time scale, acceptable losses, mtbf, serviceability, etc. etc... My more than less educated guess is, that if torque converters were suitable for aviation applications, they surely would be widely used by now. $\endgroup$
    – Jpe61
    May 3 at 12:27

I recently explored each of the available methods used to couple high RPM engines to low RPM propellers. Takeoff HP, Cruise HP, Gross weight, Propeller mass and circumference, and aircraft maneuverability are critical issues. All these have an exponential relationship with forces, some squared, some cubed.

Now, back to your original question; it is a great question. If you try to reduce speed via a torque converter, you'll have to use the type that exchanges fluid with an external cooling radiator. I didn't calculate the size, fluid volume, exchange rate, etc. because I didn't intend to reduce RPM this way.

To reduce speed via the TC, there would be much engineering effort, although mostly dealing with the heat rejection system. My design is 300HP in RV10... I dismissed the TC idea early because the heat rejection system would be far too large, heavy, and risky. If you were working on something 100HP or less, and maybe half the weight, without aerobatics, it could be plausible enough to work on rough calculations.

A typical torque converter where RPM out = 65% of RPM in, would be 85% efficient, so the other 15% will be heat. So propeller horsepower from 100HP engine would be 85HP. And you have to get rid of about 12kW continuous heat. When you consider belt or geared drives will be 98-99% efficient, TC speed reduction is unpopular.

  • $\begingroup$ automotive torque converter designs solve efficiency issues with the ability to "lock" the converter in cruise. Some have pointed out that a turbojet turbine is essentially a torque converter using the exhaust as a fluid medium. Ah, yes, about the efficiency, pistons remain higher, and are heavier, jets maintain thrust at higher airspeeds. But true, automatic transmissions can get hot. But one might wonder if a fluid coupling/radiator is heavier than a gearbox. Which one would be more prone to breakdowns? $\endgroup$ May 1 at 6:33
  • $\begingroup$ In the locked position, there is very little heat, and very little loss, and RPM in = RPM out. I inferred from the original question the intention is to use only a TC to reduce RPM. If input <> output, the converter cannot be locked. Auto trans are planetary gear systems which are very efficient. If you use a TC to drive a planetary reduction, you still cannot lock the TC, because you loose the decoupling of engine pulses. I checked my calcs for 300hp, @18k ft. was min. 16" dia. at 107lbs for TC alone. Now add pump, radiator, liquid, lines. In my notes I noted 100lbs for a gear box +misc. $\endgroup$
    – Don Doyle
    May 2 at 2:36
  • $\begingroup$ A turbojet is not a TC. It is a jet engine combustion chamber which scavenges exhaust gas inertia with a turbine directly coupled to an intake compressor via a center shaft. Better described as a turbocharged jet engine. Torque is torsional force applied to a rotating body. A TC has a mechanical rotating input driver and mechanical output. The input impeller accelerates fluid from center to circumference, the turbine decelerates this fluid from circumference to center. The fluid is returned to the impeller through the stator, which changes its direction. Entirely different than turbojet. $\endgroup$
    – Don Doyle
    May 2 at 3:42
  • $\begingroup$ As for reliability, the question is far too general to answer well. If we assume the quality of engineering and development are equal, I can definitely answer. In my experiences, poor efficiency machines always suffer early failures. All the energy lost from in to out must be disposed of as heat, noise, etc... It's very handy the coupling fluid is also the cooling fluid. All the metal surfaces contacting fluid are coolest. Trouble is, the core of the same metal is warmer. This dynamic strains and releases each cycle causing fatigue failures within the material. Higher heat equal earlier fail. $\endgroup$
    – Don Doyle
    May 2 at 4:43
  • $\begingroup$ Are "planetary gears" needed for RPM reduction? Starting to wonder if two different size pullys and a slip clutch would work. Had on of those on a go-cart as a kid. I'm most interested in using theTC to save strain on the drive when changes are made to the prop, then re-locking it. $\endgroup$ May 2 at 10:37

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .