Subject: Kinsler Nozzles mechanical pump fuel injection.
From: Rotary Engine
Date: 12/8/2006, 2:23 PM
To: AARotary Engine


    Paul,

    I agree your slide throttle carb is an attractive way to go at very
    low cost. Just a little elbow grease to make one. (Okay, maybe more
    than a little...)

    But you still need a mechanical pump even at low pressure, unless
    you have high wing tanks. Maybe the stock diaphragm pump could work,
    but those diaphragm pumps are more prone to vapor lock. (I don't
    want to open that debate again; I know you can keep the fuel lines
    cool with a blast tube, routing away from hot spots etc...)

    If you decide to use a positive displacement mechanical pump like
    the Kinsler pump, are already half way to spending what it costs for
    the full mechanical injection. The pump is the biggest expense in
    the whole constant flow system.

    The only other things you would need are the barrel valve and two
    nozzles (for a P-port; more for more runners). Here is a Kinsler
    single-port barrel valve. You can run that to an AN Tee
    fitting that will split a fuel line to each runner.

    You would have to weld a nozzle boss on each runner, and make up
    some AN fuel lines and fittings. You would use the Parker valve and
    panel adjustable twist knob for leaning.

    Throttle would be with your slide throttle. The whole thing would
    probably cost less than an Ellison carb. Because of the higher
    pressure it would be much less susceptible to vapor lock than an
    Ellison. It also would not need carb heat.

    So compared to an Ellison carb it's a great deal. If you want to use
    a turbo or supercharger, it is a perfect way to go. Even for a
    normally aspirated engine it's a great way to go.

    Compared to EFI it's an even better deal. Even with stock EFI you
    need at least two high pressure electric pumps. They aren't cheap.
    And then what about an after market controller, like Tracy's unit or
    some others. These cost about $1,000 I believe.

    Then you've got your injectors, etc.

    And after all that money spent you're 100 percent dependent on
    electrical just to keep your engine running. If your juice stops
    flowing, you are coming down. This does not sound like a good deal
    to me.

    So while the constant flow injection is a little more expensive than
    the slide throttle carb, it is still a very good option when
    everything is considered.


    Regards,

    Gordon.

    I think you are going to run into trouble if you try to limit the output
    of a positive displacement mechanical pump with a needle valve. In
    car racing
    they could care less about leaning for cruise. In aircraft we need
    full power
    for take off and super lean for cruise. In car racing they need full
    power....
    PERIOD :) Race cars also don't idle. Do you have any 800 HP sprint
    car racing
    up there?

    Doug, next time you come to my house I'll take you to the sprint
    car races and into the pits after the race so you can see some of
    these systems. I can hear them from here every Saturday night :)

    Fuel delivery in these race car systems is only a function of RPM.
    You can not get more fuel out of a positive displacement pump if you
    lower the pressure on the output side. If you restrict the output
    with a needle valve no telling what happens. It will probably overheat
    the pump and the fuel.

    In mine, Ellison, Aerocarb, Revmaster, etc., etc., fuel delivery is
    tied to
    throttle opening and intake manifold vacuum.

    The diaphragm pump won't work as they are designed to work off a
    cam shaft. A mechanical pump driven off the e-shaft will work
    with my system as well. All it need do is fill a small header tank
    equipped
    with a stand pipe. If it pumps too much fuel it just goes back to
    the fuel tank through the stand pipe. No back pressure to speak of
    on a positive displacement mechanical pump.


    Paul Lamar ...No rotor no motor.

    Paul,

    The mechanical injection will idle just fine.

    Fuel Delivery is tied to throttle opening -- it is not just a
    function of pump output with rpm.

    The barrel valve controls how much fuel goes to the nozzles. It is
    linked directly to the throttle. As the throttle is opened the
    barrel valve turns to allow more fuel to go to the nozzles. You can
    see the linkage in the photo.

    There are several other features of the system that allow very
    precise fuel delivery at all engine operating conditions.

    Have a look at the basic schematic. See where it says jet can?
    That's where we put the needle valve for leaning. The more we open
    the leaning valve the greater the return to the tank and the less
    fuel flowing to the nozzles.

    So we are not leaning by restricting flow -- we are leaning by
    increasing flow to the return. This is the key concept.

    There are also other subleties to the system. The barrel valve
    diverts some additional flow at low throttle opening. This is the
    secondary bypass valve. Have a look at the photo and the explanation.

    This system will work like a charm for an aircraft engine. We will
    have precise fuel metering throughout the operating range, very
    precise leaning -- and best of all great reliability. As long as the
    engine is turning we will have fuel delivery.


    Regards,

    Gordon

    What does all this cost? Most people installing the
    rotary are driven by cost.

    The Kinsler pump puts out 1000 pounds
    per hour at 2000 RPM. We are running our engines
    at 7000 RPM and we need 30 pounds per hour max.
    10 gallons an hour at cruise.

    We would then have to run the pump at 30/1000 or 60 RPM.
    If we had a two inch pulley on the e-shaft the pulley
    on the pump would have to be 7000/60 x 2 or 233 inches
    in diameter!!!! :)

    These low RPM 800 HP mechanical fuel injection systems seems
    impractical for a high RPM 250 HP rotary.

    Paul Lamar ...No rotor no motor.

    Paul,

    Here is the flow for the various size Kinsler pumps.

    The smallest pump, the KW-200 will work just fine. It can flow 570
    pounds per hour at 50 psi and 3500 pump rpm, which is 7000 engine rpm.

    If you run a lower pressure flow will be lower. Even with a
    turbonormalized engine 25 psi fuel pressure will be plenty.

    So flow would be:

    New Flow = Old Flow * (New Pressure / Old Pressue )^0.5

    New Flow = 579 * (25 / 50 )^0.5 = 403 pph

    At 5 psi it would flow only 180 pph. At 2 psi it's only 112 pph.

    This is maximum flow at 7000 engine rpm. Don't forget you've also
    got a return flow to the tank. I don't see a problem with this
    amount of flow. If you think this is too much you can run the pump
    at a lower speed, just change the pulley.

    Btw, what speed does the unneeded oil pump turn on the rotary? If it
    turns 1/2 speed and the gear is strong enough, you can just run the
    pump right off this gear.

    As far as cost, I don't anticipate it is that much. The costliest
    item is the fuel pump. You also need the barrel valve. The single
    port barel valve is probably the cheapest and best for our use.

    You also need a couple of nozzles and the Parker valve for leaning.
    You need a couple of AN Tee fittings, and some fuel line. That's it.

    An Ellison carb costs more than $800 for the smallest one. An
    aftermarket EFI controller that lets you lean the engine from the
    cockpit costs $1000 or more. And that's not including fuel pumps and
    fittings and lines and all the other little pieces you need.

    Btw, the Kinsler system is designed for leaning. You can use either
    their jet selector, which you can turn from the cockpit to ratchet
    between eight different jet orifice sizes, or an electric leaning
    valve. See pictures.

    Using our Parker valve isntead, will work just dandy. The system is
    designed around this method.


    Regards,

    Gordon.

    PS: The sprint cars have a lumpy idle because they use big cams with
    lots of overlap, not because of any shortcomings with the fuel
    injection. I would guess that when you add P-ports you effectively
    lengthen the overlap too.

----------------------------------------------------------
You are pumping a lot of fuel back to the tank.
You now need a two level fuel selector valve with dual fuel
lines everywhere just like an EFI system or you
will pump one tank dry in about 20 minutes or less.
One of the things we want to avoid with a low pressure
fuel injection system.
Race cars have only one fuel tank.

570 pounds per hour at 3500 RPM is still 20 TIMES more than we need.
That is a four inch pulley driven by a two inch pulley.
To halve the pump RPM again and reduce it to only 1750 RPM
and ten times more than we need or 285 pounds per hour
would require an eight inch pulley on the pump and a two inch
pulley on the e-shaft. Still impractical and space consuming
from an aircraft installation point of view.

I have some more info on the Kinsler but I won't upload it
here. Too many attachments already. I'll upload it next message.
Cost is approximately $1500 which is way more than Tracy EFI system.
More than most people spend on their engine.

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.