CVT belt for Reduction Drive

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
It can't be too bad, if done right, since they use it in some cars that get very good mileage. If you can keep the CVT feature, it may be like having variable pitch. If it was my project, I think I'd be looking into how heavy and/or big a flywheel would be required.
 

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
6,991
Location
Rocky Mountains
We have a whole thread somewhere here on this very subject.

The basic idea is sound. The implementation isn't. Long story short - doesn't work in aircraft. That is why we use variable pitch props.
Tires and tracks have near 100% linkage with the ground. Propellers, maybe 85% with the air.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,694
Location
Saline Michigan
Van Doorne belt and LuK chains require significant pressure in the piston chambers of both sheaves. Pump,seals, etc. Also the running surfaces on the sheaves are specifically engineered to each type. so, using a bunch of the tranny pieces would be needed, including ratio change software and controls, but with your own case, Ugh! Heavy and complicated...

Now if you selected a specific ratio, you might get by with spring loaded sheaves of specific sizes and much more compact set of cases. Still heavy and complicated.

The need for rpm change is handled way better with controllable props.
 
Last edited:

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
12,803
Location
Memphis, TN
It works backwards. The prop can’t be effective at multiple rpms. CVT transmission keeps engine at optimum varies wheel speed. It needs to be flipped. You want one prop RPM and vary engine to maintain that load. That means you need a double horsepower engine to have the horsepower when struggling and then you have way too much of everything you have to carry, once going. It’s pretty much how a helicopter operates power wise. Blades don’t change RPM. You load up or unload engine with need. Hover is max load. Cruise is less load.
 

Aviator168

Well-Known Member
Joined
Sep 21, 2012
Messages
2,214
Location
Brookville, NY. USA
The way I understand helicopters rotor is. If you change the collective, engine power is automatically changed so the same rotor rpm is maintained. Even with a true CVT, how are you going to change the load if the pitch of propeller remains the same?
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
5,275
A propeller has a pretty narrow useful RPM range. See it by running the engine at idle, 650 RPM, with the brakes off; the airplane probably won't move, or not much at all. Double the RPM to 1300 and you get a fast taxi. Double it again and the airplane takes off and climbs pretty handily. So at one-quarter of the takeoff RPM we get nothing, no speed, not one-quarter of takeoff speed. At half takeoff RPM we can't take off, much less get half the climb rate. That prop is only producing well when it's spinning fast.

But then there's a limit to how fast it can spin. Not much faster than that 2600 RPM and the tips start going supersonic, where drag goes way up and thrust starts to fall off. Useless, that extra RPM.

And that's why we have constant-speed props. That prop stays in the ideal RPM range and its pitch changes to keep the angle of attack of its blades in ideal places for whatever speed the airplane is moving. The engine is able to reach redline, where its max power output is.

The airplane's drivetrain requirements are entirely different from that of a car. Propellers aren't wheels. The CVT has been tried in ariplanes, anyway. The first iterations of the Bede BD-5 used it, and it was abandoned.
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
Dan:
What you're saying is equivalent to saying that somewhat lower rpm engines won't work well in aircraft. I don't believe it. Lower rpm engines have lower rpm's but more torque, just like an engine with a CVT geared down for climbing.

I can believe that getting a CVT to work with a propeller might be very difficult due to issues like torsional vibration, but that's another matter.
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
P.S. I'm only talking about relatively small differences in gearing. Like maybe 2:1 for cruise and 2.5:1 for climb. Assuming an automotive engine with the CVT being used as a redrive.
 

cheapracer

Well-Known Member
Log Member
Joined
Sep 8, 2013
Messages
6,050
Location
Australian
Dan:
What you're saying is equivalent to saying that somewhat lower rpm engines won't work well in aircraft. I don't believe it.
I believe he is correlating the engine rpms to the actual prop rpms, as in a direct drive. Relatively speaking, the engine rpms are irrelevant, he is comparing the prop efficiency at various rpms.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,694
Location
Saline Michigan
P.S. I'm only talking about relatively small differences in gearing. Like maybe 2:1 for cruise and 2.5:1 for climb. Assuming an automotive engine with the CVT being used as a redrive.
You have propeller mechanics working against you. A fixed pitch props, fans, even centrifugal pumps all behave in the following way:

The torque a propeller can react goes with the square of its rotation speed. Since power is torque times rotation speed, power then is with the cube of rotation speed. So let's doa quick thought experiment:

Size a prop to convert 100% power while running at 100% rpm , and it is running at 100% torque. Direct drive airplane engines are close to this, so the approximation is a decent example. Pull the throttle back to 95% rpm, and you are at 86% power. Come back to 90% rpm, and you are down to 73%, which is usually about where most of us will cruise. Get to 80% rpm, and you are at 50% power - this is approach power or the power setting while you are parked in a hold.

So, let's put your prop on the front, and run it at 100% rpm and torque for 100% power for takeoff and climb.

Then you get to altitude and want cruise power, so let's just say that we use a 10% higher gear ratio for cruise. If the engine is still making 100% rpm, the prop has to turn 110% rpm, consuming 133% power, but we only have 100%.

So let's let it slow down to 100% power, 100% prop rpm, but now the engine is making 91% rpm, so we only have 91% power to the system, so we have to slow the prop some. Let's say 97% prop rpm, which means the prop is using 91% power, but now the engine is now at 88% rpm. Even if it had a flat torque curve with 100% torque available there, that means only 88% power is available.

Well, pull the rpm back some more, say 95% prop rpm, which is 86% power, but now the engine is at 86% rpm. If the engine can make 100% torque here, you are just about there. But to do this, the engine throttle is still wide open, but you are still in mixture enrichment to keep the exhaust valves in the heads.

You can repeat this exercise at other gear ratios, but the same problem occurs. Then you can do the rest of the real world issues - as you climb the air becomes thinner and the engine can not make as much torque as it made at sea level, so you have to back off even more.

And to do all this, you need a gearbox of some sort. A straight gearbox is simple, but weighs something, has to have a shift mechanism, lube, cases, seals, etc. If instead the gearbox becomes a salvage job, the sheaves, pump, oil pressure control, speed sensing, a method to adjust the speed ratio, cooling... UGH!

A controllable prop is WAY lighter and simple and works great and you do not have to figure out everything in that CVT that Ford or ZF or VW or Porsche has inside that boneyard gearbox and ECU...

Billski
 

Heliano

Well-Known Member
Joined
Dec 24, 2015
Messages
83
Location
Campinas, SP, Brazil
Good point, wsimpso1, when you say "So let's slow down (....) 88% power is available". However I would make a slight correction: when RPM is reduced, your calculation is absolutely correct, since power= CP x ro x N^3 x D^5, therefore power is in fact directly proportional to cube of rotation speed. However when you reduce RPM, speed will drop a little, and so does the advance ratio V/ND, which causes a slight increase in CP, so a little more power is required to maintain the new RPM. I say "a little" because it is in fact is a very small change. Ref: NACA Report 640
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,694
Location
Saline Michigan
Good point, wsimpso1, when you say "So let's slow down (....) 88% power is available". However I would make a slight correction: when RPM is reduced, your calculation is absolutely correct, since power= CP x ro x N^3 x D^5, therefore power is in fact directly proportional to cube of rotation speed. However when you reduce RPM, speed will drop a little, and so does the advance ratio V/ND, which causes a slight increase in CP, so a little more power is required to maintain the new RPM. I say "a little" because it is in fact is a very small change. Ref: NACA Report 640
I was simply trying to get through the basics of why even modest gear ratio changes do not work well, and why it is just not done in airplanes nor in boats. Get into the reductions to the inflow velocity with decreased speed when the expressed power is reduced and yeah, things get even worse.

Simply put, it is a heavy, expensive, complicated way to get to a poor way to drive a prop. When you have a controllable propeller hung, you have a light, relatively simple CVT. Find a suitable hydraulic or electrically adjusted prop and be happy.

Billski
 

dino

Well-Known Member
HBA Supporter
Joined
Sep 18, 2007
Messages
661
Location
florida
It's sometimes done in boats but for a different reason. Boats with surface drive propellors benefit from a change in reduction ratio when ventilated than when fully submerged.
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
I don't think it's advantageous to select a climb prop for use with a CVT unless the rpm anticipated is low enough to keep the tip speed down. Better to use a cruise prop, and then use the CVT to low the full power of the,engine to be used during the climb. I imagine that it doesn't make sense to use a CVT unless you are using an automotive conversion or other high rpm engine. Then the question is how much more the CVT weighs than an alternative PSRU. It's probably a bad idea to take out the control electronics from the car instead of controlling directly.
Anyway, in climb the advantage of using a CVT is the same as having a somewhat more powerful engine spinning the same prop.
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
Oops. I meant to say leave in, not take out.

Here's another way to say the above. Let's optimize the prop for cruise at 100 knots at 8,000 feet. Now we want to climb at sea level at 75 knots. The engine can't reach full power rpm. Now we gear things down a bit. Now the engine CAN reach full rpm and spins the prop faster than it would have. My experience is that if you speed up the prop, you get more thrust.

BTW, I tried to look at this with Javaprop, but so far I haven't figured out how to get the performance at slower speed and greater air density using the prop optimised for 8,000 feet. And now my PC is ill.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,694
Location
Saline Michigan
It's sometimes done in boats but for a different reason. Boats with surface drive propellors benefit from a change in reduction ratio when ventilated than when fully submerged.
True and greatly different from our case...
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,694
Location
Saline Michigan
I don't think it's advantageous to select a climb prop for use with a CVT unless the rpm anticipated is low enough to keep the tip speed down. Better to use a cruise prop, and then use the CVT to low the full power of the,engine to be used during the climb. I imagine that it doesn't make sense to use a CVT unless you are using an automotive conversion or other high rpm engine. Then the question is how much more the CVT weighs than an alternative PSRU. It's probably a bad idea to take out the control electronics from the car instead of controlling directly.
Anyway, in climb the advantage of using a CVT is the same as having a somewhat more powerful engine spinning the same prop.
lr27 is proposing that if we progress through the same cases starting from the opposite direction, the results will be better. Think about that... Can that work?

Ran through some simple numbers anyway. Set up 75% power at 91% rpm (power goes with n^3, torque with n^2) and the resulting 83% torque at a base gearing. Now I drop the gearing to 1.1:1, let the engine run to 100% rpm, prop rpm goes to 91%, torque at the prop can be no more than 83% (rpm^2) and we are at 75% power... You can get to 100% power by letting the engine over-rev. With the 1.10 additional reduction, a 10% over rev and if the engine makes 91% torque at that speed, you have 100% power.

Cruise gearing only works in cars and trucks because we routinely run down the highway (level road) at modest power settings. We do not do that in airplanes, where we cruise level at 75% power and over 90% of WOT speed.

You can go get into all this using the more sophisticated equations, but they end up further reinforcing the same basics. If you really want to set rpm and torque independantly, go constant speed prop and be happy.

Billski
 

Aviator168

Well-Known Member
Joined
Sep 21, 2012
Messages
2,214
Location
Brookville, NY. USA
My original motivation for this topic is to explore the idea if CVT can replace a geared PSRU, simple setup and one speed only. OK, getting back to what you guys discussing. IMO, a CVT can keep the power to a fixed pitch propeller more constant. If you selected certain engine output, you can use engine power more effectively during different phases of a flight. Higher prop rpm in cruise, and lower rpm in climb. This is for a fixed pitch prop only of cause.
 
2
Top