Doug Fir and I are working on this Continuously Variable
Transmission Turbo Compound system (CVT-TC) that will extract the
maximum energy from the exhaust of any engine. It will
first be tested in Doug's Sea Wind amphibian aircraft.
This is one step up from the widely used Curtiss Wright R3350 turbo
compound engine that had a fixed gear ratio between turbine
and crankshaft. The CW R3350TC achieved a .38 BSFC at the
design point however off the optimum design point the BSFC was
somewhat higher. Typical car and aircraft engine BSFC's are around
.45 or above. This design is an attempt to optimizes the
energy recovered from the exhaust at all RPMs and power
levels.
The first stage is a knock off of a planetary reduction
found in almost all large aircraft turbine powered APU's.
Silicon Nitride ball bearings commonly found in turbo chargers
and machine tools will be used in this stage because of the high
RPM's
involved. Reduction ratio is about 11 to 1 reducing the 100K RPM
from the turbo charger down to about 10K RPM or so on the input to
the
CVT.
The torque at 100K RPM and 30 HP out of the turbine is about:
HP = Torque X RPM/5252
Therefore HP X 5252 / RPM = Torque = 1.57 foot pounds.
This low torque value is the reason a tiny shaft can be used
to extract 30 HP from the turbo charger.
At the input to the CVT this torque is increase to about 17 foot
pounds
at about 10K RPM. At a two to one final belt drive to the e-shaft
torque is increase again to about 35 foot pounds. Since the max
torque
on a RX8 engine is 160 foot pounds this represents a considerable
free lunch in terms of torque. I am assuming here the turbine is
capable of 30 HP over and above the HP necessary to super
charge the engine. (300 HP would be normally available from a turbo
charged RX8 engine.) This is a reasonably assumption as roughly 50%
of
the
energy in a gallon of gas goes out the exhaust in the form of
waste heat in all internal combustion engines. Roughly 20% is lost
in cooling the engine leaving only 30% to come out of the shaft in
the form of useful mechanical energy. To further put this
into perspective on a 300 HP engine, 200 HP is lost to the cooling
system and 500 HP is lost to the exhaust. If we can recover
just 10% of that lost exhaust energy that ups the over all
efficiency of the engine from 30% to 35%. A car now achieving
25 MPG on the highway in theory could achieve 30 MPG.
All of this must be tested to determine actual numbers as these
are rough first iteration calculation based on what
Allison, Curtiss Wright and Pratt & Whitney aircraft piston engine
builders where able to achieve in WW II with turbo compound engines.
A Lockheed P2V twin engine Naval aircraft powered by R3350TC's flew
from Australia to Ohio unrefueled in the 1960's. A feat not even
a Boeing 747 can achieve to this day.
The toroidal type CVT itself is a miniaturized version of CVT's used
in some small cars. The final belt drive isolates the CVT from the
torque variations of the rotary engine. Clamping force is applied to
the
CVT toroids by engine oil pressure. This is achieved by a annular
hydraulic cylinder throw out type bearing/clutch cylinder
combination
now widely used on manual transmission cars. I invented this around
1971. I did not patent it but it was first sold by the new
owners of Lamar Engineering for race car use.
Controlling the oil pressure in the annular cylinder with a solenoid
spool valve can make the CVT do double duty as a clutch. Note the
higher
the engine RPM the higher the oil pressure and clamping force
on the toroids. The gear ratio change of the CVT is controlled by a
small stepper or DC electric motor through a non reversible worm
gear drive. For aircraft use a simple DPDT momentary contact switch
and DC motor can be manually adjusted to optimize the CVT gear
ratio.
This device should weigh less than 20 pounds.
The cost is far far less than the cost of hybrid systems
with high cost power semiconductors.
There is nothing about this design that is incompatible with
an electric assisted turbo charger. The electric motor driving
the turbo is shut off after about one second when the turbo
reaches steady state speed. At that point the exhaust pressure
is up and additional HP can be extracted from the turbine.
An overrunning clutch can be used on the turbine output
shaft to disconnect the rotating inertia of the CVT for rapid
turbo RPM acceleration and boost pressure purposes.
This results in about a 20% increase in car MPG if the same
size engine is used. If a smaller and lighter car engine is used
further improvements in car MPG can be achieved without
sacrificing acceleration. This along with a stop start
starter/alternator now scheduled for installation in GM cars
should come close to the MPG achieved in the current hybrids
without the 700 pound excess weight penalty and much higher
cost penalty.
I hope I covered all the features. If anything is not clear
please feel free to ask questions.
Paul Lamar
Opps sent out the wrong last 3D. Here it is.
Paul Lamar
The famous (or infamous) Nomad engine also used a CVT in the recovery
system, and also was not
what I would call simple. To get the 30 mpg this has to operate with
little
loss transfering
about 3 hp (10 percent of the 30 to drive the car) it looks to me like
the
friction loss in
all those gears and friction drive CVT would be over the 3 hp. The
control
system to keep
all this optimized makes the electric system look very simple by
comparison.
Actually an
electric power recovery turbine in a Hybrid mode would seem to have a
lot of
potential -
you can buy a lot of semiconductor for the cost of a 100,000 rpm gear
set.
Vance
Come on Vance it is not going to suck 3 HP when transferring only 3 HP.
I figure over all efficiency of the drive to be about 90%. That includes
transferring 3 HP. 3 HP in 2.7 HP out. The CVT in a car gives 20% more MPG
than a conventional auto transmission. They could not be all that
inefficient.
The CPU in most cars is now more powerful than the CPU in your desk top.
Lets see.... 30 HP electric motor in a Toyota Hybrid.
30 HP times 746 watts per HP or 22,380 watts! 80% efficient say 30KW.
I don't think you have priced 30KW power semicondutors :)
Paul Lamar
Mark wote,
I'd look at getting the biggest turbo you can possibly find, to try and keep
the rpm down. At 100K you will have a great deal of trouble with that long
extension shaft bending. The slower the turbo the more likely you will get
this design to work.
Mark Wrathall
We can shorten the 1/2 inch shaft Mark. Good point. Or make it from carbon fiber.
That lowers the mass and reduces the off center whirl effect.
We can also attach the gear box to the compressor for better alignment.
Paul Lamar
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://home.earthlink.net/~rotaryeng/
Copyright 1998-2004 All world wide rights reserved.