Subject: Radial inflow turbine HP & turbo compound in cars.
From: Rotary Engine
Date: 4/3/2007, 1:15 PM
To: AARotary Engine



    snips...

    Doc and Paul,


    I'm not so sure turbo compounding has much to offer for cars. The
    main reason is that cars, unlike airplanes, cruise at a very low
    power levels, probably about 25 percent. There is little energy in
    the exhaust at low power levels, so not much power to be recovered.

    Another factor is ambient pressure. The NACA paper -- 786 I believe
    -- proved that the power recovered by the turbine is a direct
    function of the pressure gradient between engine intake manifold and
    turbine exhaust, which is at ambient pressure. The steeper the
    gradient, or delta P, the more power the turbine wheel will see.
    This should be intuitive -- the bigger the difference in pressure,
    the faster the gas wil accelerate from the high pressure side to the
    low pressure side, and the more momentum, ergo energy, it will bring
    to bear on the wheel.

    At sea level and without boost, there is no pressure gradient. In
    fact, intake manifold pressure will be lower, even at full throttle,
    than ambient pressure, due to pressure losses in the intake tract.
    At part throttle, such as cruise, the intake pressure will be quite
    a bit lower than ambient. There is nothing to be gained here.

    If you boost the engine significantly, you will get a pressure
    gradient. For example if you boost the engine 15 psi, you will have
    2 : 1 pressure gradient. In an aircraft flying at 20,000 feet, the
    ambient baro pressure is less than 7 psi. Even if you only have a
    turbo normalized setup that brings your manifold pressure to sea
    level pressure of 14.7 psi, you have a pressure gradient of more
    than 2 : 1.

    So I would say that turbo compounding would only make sense for
    high-performance vehicles that are highly boosted. In this situation
    you will get extra power from the recovery turbine, above and beyond
    what you get from the boost. So I think it is a niche technology at
    best for the automotive world.

    Also, I absolutely agree with Paul that turbo compounding is only for
    the Wankel engine. Forget about it for piston engines, for all the
    reasons stated above.

    But for an aircraft engine, the turbo compound rotary could be
    revolutionary, no doubt about it. Think of an 800 hp 4-rotor, that
    weighs less than a Pratt PT6, and sips half the fuel. Think what
    that engine would do in a plane like the TBM 850, or better yet, a
    plane specifically designed to take advantage of the lighter fuel
    load and engine.



    Regards,

    Gordon.


    Billions have been spent over the years getting one percent more
    efficiency out of car engines. Curtis Wright (no less), John Deere,
    Rotary Power International, Mazda and lately Mistral have all spent
    as much as a 100 million dollars or more (in total) on the rotary
    engine attempting to get the BSFC down from .47 BSFC to .44 BSFC.

    IMHO by downsizing the engine and the turbine during cruise
    the rotary powered car fuel consumption can be improved by as much
    as 40%. From the RX8's 25 MPG to 35 MPG using the turbo compound
    principle. Here in LA we spend as much as 80% of our travel at a
    constant80 MPH on the freeway. You don't need a 250 HP engine to go
    80 MPH. Here is an How-to in pdf format and html.

    http://www.rotaryeng.net/RX8-TC-eng.html


    Paul Lamar ...No rotor no motor.

    Paul,

    Thanks for that article. That's a neat idea, with the small rotor.

    Another approach might be to use a very small, highly boosted
    engine. Here's a paper about an engine where they ran boost of up to
    3 bar. They used ethanol direct injection for detonation control,
    thanks to the evaporative cooling effect.

    The advantage of the boost is that it gives a steep pressure
    gradient across the turbine that really kicks up the recovery power.

    I've included a copy of that paper here.


    Regards,

    Gordon.


    Gordon,

    Approaching this from another direction.
    My guess on the frontal area of the RX8 is 25 square feet.
    Five feet high and six feet wide by .8 (old rule of thumb).
    The C sub d is around .3 so flat plate drag is 7.5 square
    feet. Aero drag will be 7.5 X .0026 X 80 MPH ^2
    or 125 pounds. Drag HP will then be:
    (117 FPS X 125 ) / 550 ft/lb sec. or 25.5 HP.
    Give it another 11 HP for rolling drag, gear eff. and
    accessories. The attached charts are some measurements
    I made of total RX8 drag and HP required using the coast down
    technique so this number correlates rather well.

    Since we now know the optimized BSFC of the leaned rotary engine
    is .47 and a car requires 36 HP at a steady 80 MPH and if we
    optimized the SIZE of the rotary engine for cruise at 80 MPH and
    36 HP we should be burning 17 pounds per hour or 2.8 GPH. That
    equates to 31 MPG. Right now the RX8 is getting about 20 MPG at
    a steady 80 MPH. I have a Car Chip to measure it at 80 MPH but
    I haven't got around to measure that accurately yet.

    http://www.davisnet.com/drive/products/carchip_products.asp

    I have measured it at 55 MPH, 65 MPH and 70 MPH. It appears to
    be about 25 MPG in this range. The MPG numbers assumed the mixture
    was set at 14.7:1 which I think the RX8 computer maintains.
    To calculate fuel burn you take the pounds of air burned per minute
    from these charts and calculate the fuel required.

    So I am guessing that down sizing the rotary engine is worth 11 MPG
    at 80 MPH. The turbo compound feature will only have to provide an
    additional 4 MPG to achieve our goal of a 35 MPG RX8 at a steady 80
    MPH. Four MPG is worth a lot in these days of $3 US a gallon gas.
    In the rest of the world where gas is $5 or $6 a gallon it
    should be worth a lot more.

    BTW this approach results in no loss of RX8 acceleration or top
    speed. In fact both may be increased. It is not a free lunch
    however as the cost of the engine will be increased.

    Paul Lamar


    The Rotary Engine NewsLetter. Powered by Linux.
    ACRE NL web site. http://www.rotaryeng.net
    Copyright 1998-2006 All world wide rights reserved.

    Paul,

    This is a fine idea, no question about it. And I have no reason to
    quibble with your numbers.

    I'm actually a little surprised at the equivalent flat plate area of
    7.5 square feet for the RX8. Just goes to show you how impressive
    airplanes are in terms of drag. The Piper Malibu / Meridian is a
    six-seater with a flat plate drag of just over 5 ft^2. Peter
    Garrison's Melmoth 2 is a 4-seater with flat plate of just 2.5 ft^2.

    I think a 747 is something like 125 ft^2, which is just 16 RX8s.
    Pretty amazing.

    Anyway, back to your idea. That small rotor working at 80 percent
    power will indeed make enough exhaust energy to drive a very small
    recovery turbine, as you figured. However, you can recover much more
    energy from that turbine if you boost the intake pressure of that
    small rotor.

    The reason is that turbines like a steep pressure ratio. Whether in
    a jet or a turboprop or turboshaft or turbocharger, the work done by
    the turbine is a function of the turbine pressure ratio, or TPR,
    which is the ratio of the pressure at the exit of turbine to the
    pressure at the entry of the turbine. Obviously this ratio must be
    less than 1 if the turbine is to do any work.

    We can also turn this ratio around and look at it as a ratio of the
    turbine entry to the turbine exit, in which case it has to be bigger
    than 1 in order for the turbine to do any work.

    Here is an equation for turbine work:

    TW = (nt * cp * *Tt)at turbine exit * (1 - TPR^((gam -1) / gam)

    Where TW is turbine work per mass of airflow; nt is efficiency
    factor, cp is specific heat at constant pressure, Tt is turbine
    temperature, TPR is turbine pressure ratio, and gamma is the ratio
    of specific heat at constant pressure to specific heat at constant
    volume. For air, gamma is 1.4.

     From this relation we see that the amount of work the turbine can
    do is directly proportional to the pressure ratio between the
    entrance of the turbine and the exit.

    Here is a link that explains it better:

    http://www.grc.nasa.gov/WWW/K-12/airplane/powtrbth.html


    Paul, you have read the NACA 786 paper on turbocompounding and saw
    the results, which illustrate the importance of TPR. The horsepower
    of the recovery turbine increased dramatically as the turbine outlet
    pressure decreased (like it would at altitude, because turbine
    outlet equals ambient.)

    Here is that graph and the text. Of course with a car we don't have
    the option of decreasing our ambient pressure by climbing, but we
    can accomplish the same thing -- the pressure ratio -- by boosting
    the pressure on the intake side.

    So if you can get several hp out of that little rotor hooked up to a
    recovery turbine naturally aspirated, you could get double that if
    you boosted that little rotor to 2 bar, or triple if you boosted to
    3 bar. Just look at the chart. Horspower goes from 40 to 80 to 120
    at pressure ratios of about 1, 2 and 3.


    Regards,

    Gordon.



Yes it would be a nifty small engine. No telling what the scale effects
would be as far as detonation is concerned. That would be quite
an R&D project on its own. I am counting on more modest gains
with out reinventing something. Probably downsizing alone
would be a major first step.

BTW here is a chart that Curtis Wright put out on the rate
of development on the turbo compound engine in 1955. Note the
curve is still going straight up.

Makes you wonder where we would be if Curtis Wright had applied the
turbo compound principle to the rotary back in the early 1960's
instead of wasting years trying to get it to run on diesel
fuel.

BTW I updated the website file.

file:///website/6-Advanced-Topics/35mpg-rx8/RX8-TC-eng3.html

Paul Lamar ...No rotor no motor.

The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://www.rotaryeng.net
Copyright 1998-2006 All world wide rights reserved.