Quantcast

CVT belt for Reduction Drive

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,124
Location
Saline Michigan
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.
Well, if you wanted single speed out of the CVT, you still need the engineered surfaces on the sheaves, a method to maintain the grip forces, cases, seals, etc. With a fixed speed ratio, I suspect one sheave can be made solid, but the other still has to be loaded - factory pump was around 1000 psi, and the hydraulic diameter of the sheave was large, so the forces are big indeed. You may be able to come up with enough force with a BIG Bellville washer and a big nut to close the thing. If you keep the hydraulic actuation, you will also need to keep the 1000 psi oil pump. It can be made to work with enough effort and time and money. I doubt you will save any weight and it will be way more complicated than one of the existing PSRU with good service history.

Now start changing gear ratios in search of getting more power over a wider airplane speed range? Nope, prop mechanics are working against you. Torque you can actually apply through a prop goes with rpm squared, power with rpm cubed. You can not slow down a fixed pitch prop without greatly reducing the torque and power that gets expressed. Likewise, you can not speed up a fixed pitch prop without greatly increasing the torque and power applied to it. Change prop blade angles and sure you can shift to another place on the engine's speed vs torque map with out changing gear ratios. All this stuff was considered back in the early 1930's and again with turboprops (it is jet with a power turbine and a gearbox, so adding another gear set, a clutch and a one-way-clutch is easy), and they all went with constant speed props...

The advantage of CVT's in cars was they had more range of gear ratio than four speed automatics and let you run right on the fuel island, both improving fuel economy. Downside was the fuel used to maintain the 1000 psi to grip the belt/chain was significant, but still a net gain over four speeds. From the advent of six speed automatics, wider ratio range became available with lower pumping loss, and the fuel island has become wider with direct fuel injection too, the CVT's era passed. With the now current 8-9-10 speeds, even manual trans vehicles are not as good as automatics. Sigh. I like my own gear change.

Anyway, the big measure for which engine and PSRU to run in my mind is how big is the fleet and how well is the fleet doing for reliability? Pick from among the best, and cry over the cost only once.

Billski
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
5,469
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.
As I mentioned, it's been tried and found wanting. Aviation is 117 years old now, at least, and millions of ideas have been tried and what we have now is generally what works safely and economically. That's why we have it. I know, I know, guys will say that my sort of thinking actually holds things back, but they don't know that I've designed and invented and built machines in another industry before I became an aircraft mechanic. I looked for lots of ways to improve airplanes, but outside of small changes, they're difficult to find. The STC market indicates that, too, and I've yet to see an STC that reduces costs overall, except perhaps the Mogas STCs. One needs technologies and materials that aren't available yet for any real advances. The fixed-pitch propeller has a really narrow range of efficiency, all of it near its designed redline, and running it outside that RPM just costs efficiency, enough that you gain nothing but weight and cost and more failure points. The constant-speed or controllable prop circumvents so many of the CVT's drawbacks.
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
I'm not arguing about practicality, but about what you get out of the prop. Let's say you have an airplane with a cruise prop which works best at 8,000 feet, at about 75 percent of rated power because you don't have a turbo. You would like to climb at a significantly lower speed so you get a higher climb rate. When climbing at low altitude, the air is much denser and the airspeed is much lower, so you can't spin the prop as fast. This is made worse by ths engine turning slower and therefore putting out less power. You could climb faster by putting in a slightly larger engine, which will spin that prop a little faster. However, with the CVT, you can gear down and put the full rated power of the original engine into the prop. I don't understand why spinning the prop at a particular rpm is supposed to produce better results with a big engine that is bogged down as opposed to the same rpm from a slightly smaller engine that isn't bogged down. Or are you saying the bigger engine won't improve things either? I don't think props are smart enough to take exception to whatever is spinning them. Particularly fixed pitch props.

Whenever you mention rpm, I wish you would say whether it is prop rpm or engine rpm. Also, we are NOT setting the prop rpm with the CVT, we are setting the CVT to allow the engine to get enough rpm's to put out maximum power at the current conditions.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,124
Location
Saline Michigan
However, with the CVT, you can gear down and put the full rated power of the original engine into the prop. I don't understand why spinning the prop at a particular rpm is supposed to produce better results with a big engine that is bogged down as opposed to the same rpm from a slightly smaller engine that isn't bogged down. Or are you saying the bigger engine won't improve things either? I don't think props are smart enough to take exception to whatever is spinning them. Particularly fixed pitch props.
I went through this, but I will try it a different way - Any fixed pitch prop has a torque-speed relationship. At any given fluid density and input velocity, for the prop to run at a given speed, a certain torque MUST be applied. Less torque applied and it slows down, more torque applied and it speeds up. The relationship is torque and speed squared go together. And torque times speed is power, so they all end up linked.

I like to set up prop, gearing, and engine conditions, and see what it takes to do something. Let's make a couple assumptions:
  • We are climbing at constant airspeed, so changes in power at the prop show up in climb rate;
  • We can get 100% torque at any rpm selected - this is a modest error, as most modern automotive engines are close to this in the range we would use them in airplanes;
  • The gearing has the same efficiency regardless of ratio selected;
  • Engine power can not go above 100% power ever.
You can fix either engine speed or prop speed, and calculate through the gearing what the other must deliver. I like to use % as it is easier to keep everything in shape. Remember that the engine can deliver up to 100% torque, but no more than that, while the engine exceed 100% rpm only when power is below 100%.

I run a base condition, say 100% power, rpm, and torque for the engine, which gives 100% for the prop too.

Then I change the gearing to 1.1:1, and the prop slows down to 91%, the prop can only transmit 83% of its original torque, and 75% of its original power. Power at the engine is thus 75%, the engine is still turning 100% rpm, but only needs 75% of max torque to do it. If spinning the engine at redline speed and 75% torque is OK, go for it. Notice that if you leave the gearing alone, you can go to 75% power simply by reducing to 91% rpm and 83% torque.

Let's say you want to maintain 100% power, and we go for 1.1:1 gearing. We need to hold 100% prop torque and prop speed to hold 100% power, but at the engine, we are going to 111% rpm and 91% torque to get it. If spinning at 111% of redline to get it is acceptable, go for it. With a real engine, holding 91% torque at 111% rpm may not be available...

Use the suggested 1.25:1 change in ratio, and now you have to rev the engine to 125% at 80% torque to hold 100% power, and 114% and 66% to hold 75% power. I suggest that these sorts of over-rev are tough to accept.

You can scale all this to 75% at the beginning, and get a new set of numbers, but you always get the same result with a ratio change - To maintain power, the prop MUST spin as fast as before, but now the engine has to be spun faster and engine torque can be a little lower. In a real engine, that much torque at the high rpm may be hard to come by or may dangerously shorten engine life.

Most of us say a conversion should stick to original redline rpm and power, and for good reason. I can get on my soapbox about crank resonance, but let's just assume that there really is not a lot of margin for extending car engine rpm ranges while maintaining engine life.

Whenever you mention rpm, I wish you would say whether it is prop rpm or engine rpm. Also, we are NOT setting the prop rpm with the CVT, we are setting the CVT to allow the engine to get enough rpm's to put out maximum power at the current conditions.
Re-read my prior posts, I was specific about which gadget we are addressing... As to whether you set engine rpm or prop rpm, the problem still works out the same way... And I ran those numbers above by setting prop and running the engine to match. It does not change the prop mechanics...

Buy a constant speed prop and be happy or develop a dual ratio gearbox and spend an order of magnitude more money.

Billski
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
Where in your analysis do you account for changes in airspeed and air density? At lower airspeed and higher density, the prop will absorb a lot more power for a given rpm than it will in cruise.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,124
Location
Saline Michigan
Where in your analysis do you account for changes in airspeed and air density? At lower airspeed and higher density, the prop will absorb a lot more power for a given rpm than it will in cruise.
Changing the gear ratio does not make more thrust or power available at a the prop. The prop has a force vs rpm relationship that dominates the system behaviour. Other variables are present and do have effect.

To do a thorough job on this, we would have to define the engine, prop, test cases representative of whole mission, then do a downtown job on the analysis, share data and analysis, and prove the points. In general, when someone presents a theory, that theorist is the one with the most interest in demonstrating to other critical reviewers that the theory is valid.

I am not interested in performing the work of this "analytical experiment". I will suggest that the theorist who puts forward the theory has the most interest in proving it. Until then we have an unsubstantiated theory.

I look forward to checking the assumptions, math, characteristics, and the results obtained.

Billski
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
13,611
Location
Memphis, TN
You are still not accepting that driving a prop will flip the gearbox backwards of what you try to describe.

75% of engine power at 8000 ft is 100% available horsepower at that altitude.

With maintaining engine RPM and trying to climb, a CVT set up like a car will slow the prop as it climbs. It sees a load and is downshifting. Not what you want. Flip it around so you maintain 2700 prop RPM and you pull the nose up. The box is going to have to maintain 2700. The only way to do that and not bog is get more RPM from the engine. Does it have any reserve power to give? If you have to have a 150 hp engine at altitude to be able to have 100 hp on tap any time, wouldn’t you rather have the extra 50 hp driving all the time?

The company turbine helicopter is rated at 308hp to service ceiling. It’s a 405 hp engine. It will make 405 on the ground but it would blow the drive apart. 308 hp equates to 72 percent power at sea level. You give up almost 100 hp to have power at altitude. Hover is around 65 percent power and cruise flight is around 55 percent with reasonable densities. That’s a fixed gearbox with a governor running the engine management. That’s with an engine that will work it’s little heart out to give you the power you want. A piston engine will probably hit a wall pretty quick. A motorcycle engine might have the range to keep up with the prop demand. It’s going to be wild swings in needed power.

Changes in aerodynamic load, prop or helicopter rotor, changes power demand. Fixed pitch will slow the engine in a climb. Constant speed maintains rpm only because it flattens the pitch during the climb. Both times the engine is being worked hard. There is no excess, just different ways to deal with the demand. CVT can’t create energy so how are you going to feed the demands?
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
5,469
Where in your analysis do you account for changes in airspeed and air density? At lower airspeed and higher density, the prop will absorb a lot more power for a given rpm than it will in cruise.
I used to do the engine break-ins myself after installing an overhauled or new Lycoming in one of our six fleet airplanes. The TCDS gives the static RPM range that the engines (with fixed-pitch props) should be able to achieve on the ground. Brakes locked, full throttle, the RPM should fall with that range. If it's low, something's wrong with the engine. High, the prop is likely toast.

The airframe manufacturer specifies propellers acceptable on the engine in that particular airframe model, and with fixed-pitch, they normally want the engine to just reach redline in level flight at sea level at full throttle. I found that the relationship easily extended up to eight or ten thousand feet, since the air is thinner, reducing the load on the prop, at the same rate that the engine's horsepower falls off due to the same thing. The last half-hour of the 2.5-hour break-in was at full throttle, level flight, and it always coincided with redline.

History teaches us much. (It's really too bad that schools don't teach it much anymore, or they have the gall to revise it to suit some agenda. So society goes on to make all the same horrendous mistakes.) Studying aviation history reveals the many ideas that have been tried and abandoned and one starts to understand why we see what we see. We do not see, for instance, CVTs in airplanes. We see fixed gear ratios sometimes, but that allows an engine of a given displacement to generate more horses, since RPM times torque equals HP. The fixed-pitch prop on a geared engine is designed the same way as on the direct-drive engine, to reach engine redline at full throttle. Varying its ratio would just mess things up. To get more efficiency, we see constant-speed props because that's what works. The fixed-pitch propeller simply does not work well when you reduce its RPM. Even the constant-speed prop has to stay near redline to be efficient. In high-speed airplanes, the RPM will have to fall off some, with vastly increasing prop pitch, to keep the tip speeds within range, since the forward speed figures into the tip speed. You'll hear it in a turboprop commuter airplane.
 
  • Like
Reactions: BJC

Aviator168

Well-Known Member
Joined
Sep 21, 2012
Messages
2,214
Location
Brookville, NY. USA
If you do not change engine power setting, transitioning to cruise from climb will unload the engine as the plane moves faster and faster. The way to keep the engine loaded is to 1) increase torque the prop can obsorb by increase prop pitch, 2) by increase prop rpm or 3) both. #1 is out of question since it is a fixed pitch prop. #2 is out of question since the engine was already in maximum rpm during climb. I would be very worry if I am running 2 stroke at this point. With a CVT, one can overdrive the engine to increase rpm to keep the power obsorbed by the prop equal to the engine's output. Of cause, the best is to use a variable pitch prop, but that can't be used on a LSA.
 

Niels

Well-Known Member
Joined
Aug 15, 2019
Messages
58
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.
Can a CVT at a fixed ratio do a better job than two gearwheels concerning torsional vibrations?
There is quite some creep,inefficiency and damping involved.
 

Aviator168

Well-Known Member
Joined
Sep 21, 2012
Messages
2,214
Location
Brookville, NY. USA
Can a CVT at a fixed ratio do a better job than two gearwheels concerning torsional vibrations?
There is quite some creep,inefficiency and damping involved.
Probably.
Torsional vibration on gearwheels is not a problem as long as the ratio is in such a way that every pulse hits a different tooth and on average all teeth are evenly used.
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
5,469
Probably.
Torsional vibration on gearwheels is not a problem as long as the ratio is in such a way that every pulse hits a different tooth and on average all teeth are evenly used.
There's a lot more to it than that. Billski made his living designing stuff that had to deal with it.
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
5,469
The only CVT in an airplane I'm aware of was in the initial BD-5. And that, I believe, was so it could turn a tiny propeller with a lot of pitch. High torque at low forward speeds, transitioning to high RPM at high forward speeds. The problem, as usual, is that the small prop is so inefficient, and such high pitch at low forward speed makes it half-stalled until the speed comes up. Takeoff runs get very long. The losses of the CVT don't help.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,124
Location
Saline Michigan
Can a CVT at a fixed ratio do a better job than two gearwheels concerning torsional vibrations?
There is quite some creep,inefficiency and damping involved.
Hmm, at Ford, we did a CVT with the variator, we were in a joint venture with ZF and even brought in their arch rival LuK for part of the design. Built a bunch of them at Batavia Ohio. I do not remember doing anything special to isolate the engine from the tranny and driveline. The torque converter had a pretty standard torque converter, clutch, and spring damper. I suspect that the variator has a little more inertia than the compound planetary and various clutch and brake plate sets, but I do not have any hard numbers.

In most PSRU, the gearset (or variator) and prop are isolated from the engine firing frequency and others by a "soft" element between the engine/flywheel and the gearsets or variator. The variator may be a bit stiffer than a gearset, but I suspect it is higher inertia too. Just to put in scale, the prop is a really big inertia in the system, then the engine/flywheel, then the gear set or variator. I suspect that the standard approach will work fine with it. No way would I carry any hope of running it without a "soft" element between engine and variator or of attempting to make it live as a "stiff" system.

As to the creep or inefficiency, the sheaves and belt or chain can NEVER be allowed to slip. These systems work fine, run at low losses and with decent durability unless it ever slips the belt or chain. That scratches up the traction surfaces and does inevitably result in a contamination of the pump and controls with hard steel solids and ends in a gearbox failure. You better believe that slip is precluded in the programming. LuK even has a patented device to help prevent slip with somewhat lower pump work to prevent slip. While these things do not slip, their efficiency is not as high as we would hope - the pump that maintains a safe margin of clamp load to prevent the sheaves and belt/chain from slipping on each other does take more power from the engine than in a conventional planetary automatic. The pump needs to make as much as 1000 psi to hold the variator, while conventional automatics were up to 300 psi, and more recently are running around half of that, all in an effort to keep the pump work down.

Damping? Not so you would know it. Steel on steel in steel on steel. Pretty massive and springy. Then then soft element between flywheel and tranny tends to remove most of the input vibration.

Billski
 
Last edited:

vhhjr

Well-Known Member
HBA Supporter
Joined
Jun 29, 2003
Messages
92
A few years ago I built a rolling test stand that used a go kart CVT as a PSRU. I tested several different props at rolling speed up to 35 mph and wrote an article for Experimenter E-Magazine. I must say the article attracted quite a few naysayers. I did find the system worked in that setting the engine rpm as a constant the prop speed varied depending on the diameter and pitch as well as the forward speed of the test stand. Part of the reason for this excrsize was to investigate using a CVT as a poor man's constant speed prop. That and it might be a way to get the CSP effect without violating the Light Sport prohibition against them.

More recently I have been testing large ducted fans using the same rig. However, I had to replace the CVT with fixed pulleys because I swapped the 12 HP engine for a 35 HP 2si that over powered the CVT. I do have a snowmobile CVT that I would like to try if time permits.

I have attached some info from the earlier tests.

Vince Homer
 

Attachments

vhhjr

Well-Known Member
HBA Supporter
Joined
Jun 29, 2003
Messages
92
It appears that my post killed the thread. That, or everyone was tired of talking about the topic. I thought the Light Sport comment would light off something.

Vince Homer
 

trimtab

Well-Known Member
Joined
Apr 30, 2014
Messages
105
Location
rocky mountains, rocky, usa
Fixed pitch prop efficiency is highest in a limited range of airspeed and RPM combination where the L:D contributions of the blade are optimized. Controllable pitch props extend this range, but the twist of the blade along its length is still a.large compromise between static or low speed flight and cruise. The blades of the TU-95 in ight you can see on YouTube are a clear example of this. The twist is extreme to favor high cruise speeds, keeping more of the blade closer to a more efficient AOA at higher speeds than during takeoff (among other things...there is a lot going on with them as you'll see in the videos).

A CS prop from a production aircraft that cruises at 130 kts will not allow a white lightning to achieve interesting numbers at cruise, and a better prop for the white lightning would deteriorate the takeoff and climb performance of the pokey production plane even using the same engine (power, rpm's). The difference is in the inflow speed.

Nonetheless, there are efficiency gains of several percent to be had from gearing a fixed pitch prop as well:
 

Attachments

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,124
Location
Saline Michigan
If you do not change engine power setting, transitioning to cruise from climb will unload the engine as the plane moves faster and faster. The way to keep the engine loaded is to 1) increase torque the prop can obsorb by increase prop pitch, 2) by increase prop rpm or 3) both. #1 is out of question since it is a fixed pitch prop. #2 is out of question since the engine was already in maximum rpm during climb. I would be very worry if I am running 2 stroke at this point. With a CVT, one can overdrive the engine to increase rpm to keep the power obsorbed by the prop equal to the engine's output. Of cause, the best is to use a variable pitch prop, but that can't be used on a LSA.
I have not looked at this thread lately. Yeah, you might overdrive the prop and make it work. I get tired of doing other folks' work when they start "what-if"ing when they should actually run some example numbers and then make their case... same deal here. Folks are asserting this will work, well, postulate some adjustments, and figure out how much engine torque and rpm that would mean, and see if it makes sense, then present it to us.

Billski
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,124
Location
Saline Michigan
It appears that my post killed the thread. That, or everyone was tired of talking about the topic. I thought the Light Sport comment would light off something.

Vince Homer
It was kind of talked to death. And the folks proposing stuff were not doing their homework, so it stopped.

Your Rolling gadget with a CVT allows you to play with whatever prop you get, spin the prop at whatever speed the engine power will turn it, and then take data...cool.
 
Top