Designing a Planetary Gearbox to increase speed?

[scuba1000]

I have been researching ways to design a planetary gearbox. It seems that the main configuration of the gears is to have the ring gear fixed, output the rotation of the planet gear carrier, and use the sun gear as the input. This would produce a reduction.

I have found little information about using a planetary gearbox the other way around to multiply speed. Is it commonly done or is there some other alternative to this? The design would be to have the ring fixed, the gear carrier as the input, and the sun as the output. Is the ratio just the reciprocal of the normal ratio.

I did see that the Wikipedia article on planetary gearboxes said that the ratio for what I'm talking about is 1+ Nring/Nsun. But it said a citation is needed, so I just wanted to double check.

[MechEng83]

The most common example is an old-school pencil sharpener (although maybe not as common anymore). The handle is attached to the carrier and the planets spin relatively fast compared to the crank handle, which causes several rotations of the sharpening cylinders around the pencil. There is no sun gear in this case, as the desired output is shearing of pencil pieces, rather than shaft output speed.

Yes, the ratio with a fixed ring gear with the sun as the output is 1 : (1+Nring/Nsun). You can do any of the 6 combinations of sun, carrier, or ring fixed with the other 2 being input and output. The math gets kinda fun, and I'm sure I could dig up an old textbook to cite. Certain configurations even produce backwards rotation.

[Tom Ore]

Ns * Ss + Nr * Sr = (Ns+Nr) * Sc

where N is number of teeth and S is speed, s, r, and c are sun, ring, and carrier.

If the speed of the Ring (Sr) is zero then:

Ns * Ss = (Ns + Nr) * Sc

or

Sc= Ns / (Ns + Nr) * Ss

(This is for a simple planetary)

[IKE]

Originally Posted by Tom Ore

Ns * Ss + Nr * Sr = (Ns+Nr) * Sc

where N is number of teeth and S is speed, s, r, and c are sun, ring, and carrier.

If the speed of the Ring (Sr) is zero then:

Ns * Ss = (Ns + Nr) * Sc

or

Sc= Ns / (Ns + Nr) * Ss

(This is for a simple planetary)

As Tom is showing, technically, all 3 can spin at a given rate. One of the more recent and famous versions of this is the drivetrain of a Prius:
http://en.wikipedia.org/wiki/Hybrid_Synergy_Drive

In your scenario, you are used to seeing a PSR or Planetary speed reducer. This is usually because you have a high speed source (like an electric motor) and want high torque to do some useful work.

Windmills are the opposite and have PSI or planetary speed increasers as they blades produce a lot of torque, but the generator wants higher speeds.

[kevin.li.rit]

Originally Posted by IKE

As Tom is showing, technically, all 3 can spin at a given rate. One of the more recent and famous versions of this is the drivetrain of a Prius:
http://en.wikipedia.org/wiki/Hybrid_Synergy_Drive

From Wikipedia

Originally Posted by Wikipedia

This limited gear-ratio set, forces the engine crankshaft to rotate at speeds where the ICE is less efficient, i.e. where a liter of fuel produces less joules.

Does this make sense? Does a limited gear-ratio set on a ICE produce incomplete combustion?

[DonRotolo]

Originally Posted by kevin.li.rit

Does this make sense? Does a limited gear-ratio set on a ICE produce incomplete combustion?

I did not check the article, but it is a fact that certain IC engine speeds are more efficient than others. Depends on the engine of course but efficiency generally peaks around 2500-3000 RPM. If you put a stupid gear ratio in you can force efficiency down.

[kevin.li.rit]

Originally Posted by DonRotolo

I did not check the article, but it is a fact that certain IC engine speeds are more efficient than others. Depends on the engine of course but efficiency generally peaks around 2500-3000 RPM. If you put a stupid gear ratio in you can force efficiency down.

You can force efficiency down, but does a litre of fuel produce less joules as stated? I assume this is only possible due to incomplete comubstion?

[Ether]

Originally Posted by kevin.li.rit

You can force efficiency down, but does a litre of fuel produce less joules as stated?

No. It produces fewer joules.

I assume this is only possible due to incomplete comubstion?

The joules being referred to are mechanical output joules at the engine crankshaft, not joules of combustion heat.

[DampRobot]

Originally Posted by Ether

The joules being referred to are mechanical output joules at the engine crankshaft, not joules of combustion heat.

I'd presume that fewer joules of the chemical are released at all--IE some of those joules are turned into entropy.

[Ether]

For a given required mechanical power output at the engine crankshaft, there is a continuum of engine_speed + fuel_flow_rate combinations which will produce that power. Only one of those pairs minimizes the fuel_flow_rate.

[Clem1640]

Engine efficiency is not (necessarily/generally) a function of incomplete fuel combustion. Engines do not provide 100% of the energy produced by burning fuel as useful mechanical energy. The balance of the energy is released as heat.

A modern gasoline engine has an efficiency typically in the range of 25-30%.

Engines operate most efficiently within a certain rpm range. This provides the connection between gear ratios and engine efficiency.

[IKE]

Originally Posted by kevin.li.rit

From Wikipedia



Does this make sense? Does a limited gear-ratio set on a ICE produce incomplete combustion?

Here is a link to some BSFC maps:
http://ecomodder.com/wiki/index.php/...on_(BSFC)_Maps
BSFC or Brake Specific fuel consumption shows how much fuel is consumed for each useable work unit produced at a given operating point.
If you look at the:
Mercedes Benz 300SD OM617 5 cylinder turbo diesel map,

You will see behaviour that Don R is talking about. where there is a sweet spot to run the engine from a (fuel in)/(work out). For that engine, the sweet spot is between about 1900 and 2600 rpm and the engine is about 60-90% fully loaded.
For gas and diesel engines, the general rule of thumb for peak efficiency is around 50% max rpm and at and above 50% load. This is a very general statement.

If you had a 1 speed gearbox, you would figure out the most important attribute,a nd size your gearbox to match that (acceleration, top speed, fuel economy). For Supermileage, we had a single speed drivetrain.

If you want it all, then you would go for a CVT or continuously variable transmission that would (in theory), give you the optimal ratio for whatever scenario you are driving. These gearboxes tend to have other issues (efficiency, cost, longevity, weight....).

The Toyota Hybrid system does a really good job of acting like a CVT without some of the draw-backs, and with some other benefits. Its not "perfect", but it is a really good system for a car, and a pretty neat one to figure out.

BTW, you can replace the generator/electric motor with Hydrualic pump/Hydrualic motor and achieve a similar set up... Of pressurized air... Or... Check out this Dana HVT. It is a 3 range IVT (infinitely variable transmission) powersplit...
http://www.youtube.com/watch?v=jsgw7ozlfJ0

[Ether]

Originally Posted by IKE

If you want it all, then you would go for a CVT or continuously variable transmission that would (in theory), give you the optimal ratio for whatever scenario you are driving. These gearboxes tend to have other issues (efficiency, cost, longevity, weight....).

Does anyone know: What's the word on the street about the CVT in Nissan Altima concerning the issues mentioned above?

[IKE]

Originally Posted by Ether

Does anyone know: What's the word on the street about the CVT in Nissan Altima concerning the issues mentioned above?

Nissan and CVTs have a pretty long history, so a little searching should give you some background. I believe they use a "push style" belt which is pretty neat. Check out "Van Doorne" for some history on the push belt.

[scuba1000]

Excellent. Thank you both.