Reduction drives

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

alr

Active Member
Joined
Nov 24, 2004
Messages
35
Are you saying that for every 100 rpm the engine turns in cruise, the propeller will turn at 60 rpm because its more efficient for the torque converter?? :think:
That is close to what I am trying to say, but not quite. Relating my response to what you wrote, it is not that the propeller turns at 60% of the engine rpm BECAUSE it is the most efficient operating point, but rather it turns at 60% because the torque converter and the load are matched when the system is designed. In other words, operating at the most efficient condition is a consequence of an appropriate design, not a cause.

If the torque converter does not match the load then the speed ratio under operation will not be at the optimal efficiency point but at some other point, either a higher or lower ratio than optimal. This means that one must pick the appropriate torque converter so it matches the load so the rpm ratio is at the most efficient operating point of the torque converter. It may also mean messing around a bit with the diameter and pitch of the propeller to fine tune the load so everything is matched. In a way it is a lot like matching a prop to an aircraft but a little more complicated.

As far as matching is concerned, one reasonable strategy would be to set the design point so the torque converter operates at its optimal efficiency point under cruise conditions.

Another strategy would be to choose a slightly higher rpm ratio than optimal when the aircraft is cruising. I'm not talking about a much higher rpm ratio because that would get you too far from the most efficient operating point, but rather just enough so your efficiency is down by 1 or 2%. Then when the aircraft is in take off or climb mode the torque converter will be operating at a slightly higher efficiency than if one were to choose the design point mentioned in the previous paragraph.

The best choice would depend somewhat on whether you want to give more emphasis to cruise or take off/climb, kind of like when you choose a propeller.
 

alr

Active Member
Joined
Nov 24, 2004
Messages
35
The CVT is automatic. Flyweights in the drive pulley, plus loading on the belt, control the positions of the two pulleys. If you've ever run a snowmobile you'll know how awesome it works. The acceleration can be scary for the uninitiated.

Dan
That is interesting. I have only driven snowmobiles a few times in my life, and I had not learned about the transmission technology they used.
 

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
The planetary drive we were working on was in the end a success on the function end of things but we were still not happy with the life expectancy of the gears and bearings. We were using a planetary set from a C-6 transmission which at first looked to be good for about a 700 hour life, possibly as much as a 1,000. However, a closer look by a more experience drivetrain designer revealed that while the gears may possibly reach that life, the bearings would most likely have a much shorter useful existence. Given that conclusion I decided to change the design in order to gain a more robust and longer lived configuration. One of the changes was to go from the C-6 gears to a set from a commercial Allison unit (don't recall which one). The larger gear set was manufactured to a somewhat higher gear standard with substantially stronger materials. Also, the bearings did better with the high speeds of the rotary. But in the end the company that was behind the effort ended up running out of development funds and eventually went out of business. I don't know what the end result was of the prototype parts we did make.

Regarding your reduction, keep in mind that the design of the reduction must take into account not only the gear loads but also those of the prop. This includes not only the thrust and drag loads but also those of the gyroscopic moments. This is where I find most of the reductions on the market fall short simply because they seem to fail to take into consideration that the flight loads must be able to transfer into the engine block and into the airframe in a smooth and predictable manner. Mounting a structure that has those loads onto a simple flat plate is just poor design. As an example of good design, look at the housing and load transfer of the reduction drive of the Mistral rotary. This type of structural configuration should be the goal regardless of whether you're looking to design something for the rotary or for a V-8.
 

RJW

Well-Known Member
Joined
Feb 9, 2011
Messages
708
Location
Wisconsin and Kansas
Thanks Kristoffon,

I’ll check out the MitCalc program.

I’ve already studied all of Jack Kane’s stuff and read the RWS page. Both of these sources seem to have reliable information.

Orion,

Thanks for the reply. I’m glad to hear that in the end your design made the cut. It must have been a satisfying experience. Sorry to hear that the money ran out. But that seems like the rule in most of this stuff nowadays.

The Mistral box looks like a work of art. Thanks for the link.

Attached is a quick redraw of my box. Again this is just a sketch. Does this seem like an improvement?

Rob

PSRU R 4.jpg
 

PeterJC

Well-Known Member
Joined
Sep 12, 2010
Messages
48
Location
Helsinki, Finland
However, a closer look by a more experience drivetrain designer revealed that while the gears may possibly reach that life, the bearings would most likely have a much shorter useful existence.
This has been my concern too. I was going to order one 6-planet unit and disassemble it to get exact dimensions of bearings - to make some sense about this TBO issue.
There is presumably needle bearings inside, and from their dimensions can be made an educated guess of how long they last on a given torque/rpm mix.

Bill, regarding that short life expectancy, do you recall what was the power level and engine? 200 hp 13B-MSP renesis? 250 hp Turbo 13B-REW? 20B?
 

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
Yes, we were designing for a normally aspirated 13B so the minimum power level was set at 200 hp. Variants also considered ranged up to 250 hp.
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
5,278
That is close to what I am trying to say, but not quite. Relating my response to what you wrote, it is not that the propeller turns at 60% of the engine rpm BECAUSE it is the most efficient operating point, but rather it turns at 60% because the torque converter and the load are matched when the system is designed. In other words, operating at the most efficient condition is a consequence of an appropriate design, not a cause.

If the torque converter does not match the load then the speed ratio under operation will not be at the optimal efficiency point but at some other point, either a higher or lower ratio than optimal. This means that one must pick the appropriate torque converter so it matches the load so the rpm ratio is at the most efficient operating point of the torque converter. It may also mean messing around a bit with the diameter and pitch of the propeller to fine tune the load so everything is matched. In a way it is a lot like matching a prop to an aircraft but a little more complicated.

As far as matching is concerned, one reasonable strategy would be to set the design point so the torque converter operates at its optimal efficiency point under cruise conditions.

Another strategy would be to choose a slightly higher rpm ratio than optimal when the aircraft is cruising. I'm not talking about a much higher rpm ratio because that would get you too far from the most efficient operating point, but rather just enough so your efficiency is down by 1 or 2%. Then when the aircraft is in take off or climb mode the torque converter will be operating at a slightly higher efficiency than if one were to choose the design point mentioned in the previous paragraph.

The best choice would depend somewhat on whether you want to give more emphasis to cruise or take off/climb, kind of like when you choose a propeller.

I still fail to see how a fluid-type torque converter can provide an RPM reduction without losing a lot of power in the process. Oil shearing creates a lot of heat. A positive-displacement hydraulic pump driving a hydraulic motor can be made to give any handy reduction ratio just through displacement-per-rev differences, but I've never heard of a torque converter doing such a thing. When it's slipping it's losing energy somewhere, unless there's some magic inside that I've never seen.

Lots of newbies have proposed the hydraulic-pump-and-motor affair for redrives, too, BTW. But go to the industrial suppliers and see how much that stuff weighs, and realize that it, too, gets hot enough to require oil coolers. Works great for Bobcats and excavators, machinery that need the weight anyway for traction and stability and require the control finesse it offers, but those machines don't have to fly on affordable power. And by the time one bought all that stuff he'd have spent a big chunk of a used Lycoming's price.

Dan
 

alr

Active Member
Joined
Nov 24, 2004
Messages
35
I still fail to see how a fluid-type torque converter can provide an RPM reduction without losing a lot of power in the process. Oil shearing creates a lot of heat. A positive-displacement hydraulic pump driving a hydraulic motor can be made to give any handy reduction ratio just through displacement-per-rev differences, but I've never heard of a torque converter doing such a thing. When it's slipping it's losing energy somewhere, unless there's some magic inside that I've never seen. ...

Dan
The magic inside a torque converter that gives RPM reduction without loosing a lot of power is called the stator.

Basically, a simple torque converter is composed of three solid parts, a driving member (impeller), a driven element (turbine), and a stationary element (stator). The engine is connected to the impeller. The stator stands more or less between the impeller and the turbine. The load is connected to the turbine. All three of these elements have vanes to interact with fluid contained in the device.

The impeller sets the fluid into motion. The stator redirects the fluid motion. The flowing fluid which has been redirected by the stator then interacts with the turbine, transferring power to it.

The magic in all this is that because of the way the stator redirects the fluid flow there is more torque applied to the driven member than was supplied by the driving member. You can't get something for nothing, and the trade off you make for more torque is lower rpm. There is a kind of mechanical advantage in operation here, but it is not a hard-coupled mechanical advantage as you have with a lever or gear, but a soft-coupled mechanical advantage resulting from the fact that the power is transferred by vanes and a liquid.

There is power loss with a torque converter. However, when it is operating at its optimal operating point (i.e. the optimum rpm ratio) the power loss is surprisingly low, somewhere in the range of 10% power loss in a reasonably efficient unit. Some manufacturers claim even lower losses. When running under non-optimal conditions the power loss is higher, reaching 100% loss when the speed ratio is zero (the unit is stalled) and also reaching 100% loss when the input and output shafts are rotating at the same rpm.

One way to think about the power loss is that it arises from a kind of slippage. This comes from the fact that it does not operate on a positive displacement principle (e.g. a hydraulic cylinder or a piston pump) but on a principal of vanes interacting with liquid (e.g. a centrifugal pump). In this respect it is also much like the propeller of an aircraft. A propeller is never 100% efficient because of its inherent design (slippage). However, when operating near its optimal operating point the power loss is a reasonably small fraction of the power delivered.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
12,823
Location
Memphis, TN
So what you want to do is run the thing in the stalled or below range, so you are using the stalled speed to govern the max speed driven by the prop? On another note how do you keep the prop from windmilling with low power but high speed like in a dive?
 

Bigorneau

Well-Known Member
Joined
Aug 24, 2009
Messages
47
Location
near LFOQ, Center, France
Hi every body,
you also may use this reduction drive, gear ratio 1.56/1, input from 200 hp to 300 hp output (propeller) of 1.70 meter diameter, pitch 2.30 meter, input V8 4.6 liters injection with mapping injection to have the best torque between 3000 to 3500 rpm (hight cruise speed (max rpm 4700)

Rgds

joe:tired:
 

Attachments

RJW

Well-Known Member
Joined
Feb 9, 2011
Messages
708
Location
Wisconsin and Kansas
Joe,

Is that a marine Casale gearbox? If so are there any features about the box itself that you don’t like? Casale boxes have been used on a few aircraft but I don’t know much about them.

Thanks,
Rob
 

Bigorneau

Well-Known Member
Joined
Aug 24, 2009
Messages
47
Location
near LFOQ, Center, France
Joe,

Is that a marine Casale gearbox? If so are there any features about the box itself that you don’t like? Casale boxes have been used on a few aircraft but I don’t know much about them.

Thanks,
Rob
yes it is a derivative of Casale gearbox i have on my e-racer.
All work correctly, may be add a better system after ball bearing to avoid exhaust of very small oil quantity but that the alone thing i see.
You can change reduction ratio if you need (the two helicoidal gear).
Rgds
Joe
 

pie_row

Well-Known Member
Joined
Nov 9, 2009
Messages
710
Location
salt lake city Ut
The planetary gears are made from powdered metal, then sintered into shape. This is a process of taking the powder and subjecting it to high pressure and temperature, thus causing the powder to turn into a semi-liquid state, forming the part. The process is well known and although the quality controls are good, I wonder if they are dependable enough for our application.
One thing that I have wondered about is using a brazing operation to fill the remaining micro holes in the metal structure as those are the starting points for cracks. I was thinking that the high temp nickel brazing stuff would be the stuff to use as it is 80,000 PSI. What do you think?
 

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
The boundary between the dissimilar metals (the major problem here is "E") is still a stress concentration so as far as longevity is concerned, spend the extra bucks and get a proper cut/honed/polished gear made, preferably to AGMA 12 or better. AGMA 16 would be ideal.
 

GaryBuster

Member
Joined
Jun 3, 2011
Messages
10
Location
Frankston, Texas
I've seen many a belt get thrown, and several "boxes" with spur gears get eaten up... I'll stick with the C6 planetary... It worked without a hitch on my mazda 12a, the subaru 2.2, and currently on the subaru 2.5.
 

pie_row

Well-Known Member
Joined
Nov 9, 2009
Messages
710
Location
salt lake city Ut
Actually, what I have in mind it a classical, i.e. non-locking torque converter.

(…)

One more thing. If we were to compare a torque converter (90% efficient if run under optimal conditions) to a v-belt drive (85% efficient according to some posts in this thread) the torque converter wins in efficiency. It loses compared to a geared drive (what is it, 97% efficient), but the torque converter has the advantage of torque multiplication at take off or other highly loaded operating conditions.
You have put together a well thought out argument for using a torque converter. One of those operates as a CVT. It also has a very effective torsional damper built in. I like it. Some Auto trance missions use the converter between one of the elements in a planetary gear set and the stationary housing. That will tend to cut your losses as the bulk of the power will go through the gear set so you will loose 10% of 20% of the power not of the total.
 

jhausch

Well-Known Member
Joined
Dec 10, 2008
Messages
135
Location
WI
I've been wondering what happened to this thread so I went searching.

Are you saying the best solution is a TC designed to act as a gear reducer, or to design one that is "locked" at design rpm?

Assuming the former, what sort of heat would need to be dissipated and would a radiator with built in tranny cooler be sufficient (since we are talking about this for auto conversions anyhow)

Would this have to be a constant speed prop? What happens with a dead or seized motor? The prop can't be allowed to windmill uncontrolled or it would be a huge drag.


I suppose if the overall weight penalty was "out-weighed" ;) by the benefits of a reduction unit, reliability, and no issues with violent torque impulses at start up and low RPM it might make sense.

The magic inside a torque converter that gives RPM reduction without loosing a lot of power is called the stator.

Basically, a simple torque converter is composed of three solid parts, a driving member (impeller), a driven element (turbine), and a stationary element (stator). The engine is connected to the impeller. The stator stands more or less between the impeller and the turbine. The load is connected to the turbine. All three of these elements have vanes to interact with fluid contained in the device.

The impeller sets the fluid into motion. The stator redirects the fluid motion. The flowing fluid which has been redirected by the stator then interacts with the turbine, transferring power to it.

The magic in all this is that because of the way the stator redirects the fluid flow there is more torque applied to the driven member than was supplied by the driving member. You can't get something for nothing, and the trade off you make for more torque is lower rpm. There is a kind of mechanical advantage in operation here, but it is not a hard-coupled mechanical advantage as you have with a lever or gear, but a soft-coupled mechanical advantage resulting from the fact that the power is transferred by vanes and a liquid.

There is power loss with a torque converter. However, when it is operating at its optimal operating point (i.e. the optimum rpm ratio) the power loss is surprisingly low, somewhere in the range of 10% power loss in a reasonably efficient unit. Some manufacturers claim even lower losses. When running under non-optimal conditions the power loss is higher, reaching 100% loss when the speed ratio is zero (the unit is stalled) and also reaching 100% loss when the input and output shafts are rotating at the same rpm.

One way to think about the power loss is that it arises from a kind of slippage. This comes from the fact that it does not operate on a positive displacement principle (e.g. a hydraulic cylinder or a piston pump) but on a principal of vanes interacting with liquid (e.g. a centrifugal pump). In this respect it is also much like the propeller of an aircraft. A propeller is never 100% efficient because of its inherent design (slippage). However, when operating near its optimal operating point the power loss is a reasonably small fraction of the power delivered.
 

blainepga

Well-Known Member
Joined
May 20, 2011
Messages
69
Location
Queen Creek, AZ 85142
Re: Heard about Honda and Lycoming ?

I think the best thing that could happen for us lighter homebuilt builders would be a merging of engineering talent from the auto engine industy with the aircraft engine industry. I read an article about Honda working with Lycoming possibly on a new series of aircraft engines. It is hard to believe Honda is interested in spending their time on something like this, maybe they think the small general aviation market will really take off when the Sport Pilot thing finally happens.

And its also hard to believe that Lycoming would want to have new competiion for its own "stone knife and bear skin" technology line of engines. Of course they may be seeing how many engines Rotax is selling and have decided that they want a piece of the under 110 HP uncertified engine action. Actually I am surprised they haven't started an uncertified engine subsidiary disting off the old blueprints and making their old 65 horse engines again.

I have owned several Hondas and would love to see Honda create a line of small aircraft engines. Especially knowing that they have partnered with a knowledgable (although hopelessly 20th century) aircraft engine company on the project. I am currently looking at the Geo/Suzuki engine and the Raven Reduction drive. I know the Lycoming/Continentals are reliable but I can't justify spending year 2020 cash for year 1930 technology even though it would be the easiest path to take.
If you want a Honda go to Firewall Forward Aero Engines. They have taken over the company that produced the CAM100,125,500 PSRU's developed for the honda engine.
 

delta

Well-Known Member
Joined
May 26, 2011
Messages
2,134
Location
Brookside Utah
I would love to use the existing cog belts and pulleys off a Toyota or Lexus for prop reduction, but I've heard that a 2 to 1 reduction is a bad deal due to some harmonic thing. That's always puzzled me because it seems the belt would absorb the harmonic resonances.
My ideal box would have counter rotating capability too.

Rick
 
Last edited:

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
Although belts can absorb a small amount of vibratory resonances, keep in mind that any absorption of type will manifest itself as heat in the belt. Too much and the belt will wear our very quickly and in the extreme cases, even come apart. Before resonance issues were serious considered, belt life in early reductions was as low as five hours. And yes, even numbers in reduction rations can be inoptimal due to the interaction of the engine and prop, especially if they have some form of 2:1 relationship also (4 cyls, two blades, etc.)
 
2
Top