Subject: EFI Maximite was EFI and CD ignition
From: paul lamar
Date: 9/9/2017, 5:53 AM
To: A10-Me-Earthlink



    Paul,

    I ordered 2 RX 8 crank angle sensors and am fabricating a bracket to
    mount the sensors.

    Room is tight due to modified water pump and heat shield for the
    augmenter.

    Could you explain how the Mighty Mite will use the pulse generated
    by the sensors and the one tit trigger wheel?

    The first pulse would be firing ignition by the CDI and the MM would
    know 20 degrees BTDC,

    Does it look for the second pulse to get a time frame for one rev of
    the Eshaft?

    What happens next?

    What would the code look like?

    Thanks

    Joe Berki

    There is no code for the CD ignition. The maximite does not need to
    be involved.

    The front rotor (car) fires from the first CAS at 22 degree BTDC.
    Then the second CAS 180 degrees later fires the rear rotor. That is
    all there is to it.

    That reminds me I need to check and see if I got the sample
    transformer for the CD system.

    Now the CAS can also be connected to the Maximite to fire the
    injectors after it checks the mass airflow sensor and figures out
    the correct pulse width.

    Paul Lamar

    This part is what I was trying to  understand I know the CDI is
    independent.

    I wanted to know how the Maximite fires the injectors.

    Joe Berki


    The  Maximite has 20 I/O pins numbered 1 to 20, All of these pins can
    be configured as digital out puts or something else. The pin will put
    out 5 volts.

    You choose the pin you want to control the injector.

    Then it must first be configured. The instruction is "SETPIN pin,
    cfg" "pin" is simple the pin number you choose. Let say you choose
    pin number 2

    The instruction is then written SETPIN 2

    cfg stands for configure. Here is a list of possible configurations
    for most pins.

    0 Not configured or inactive 1 Analog input (pins 1 to 10) 2 Digital
    input (all pins and 5V tolerant on pins 11 to 20) 3 Frequency input 4
    Period input (pins 11 to 14) (pins 11 to 14) 5 Counting input (pins
    11 to 14) 6 Interrupt on low to high input change (pins 1 to 20, D0
    to D8) for target 7 Interrupt on high to low input change (pins 1 to
    20, D0 to D8) for target 8 Digital output  (all pins) 9 Open
    collector digital output to 5V (pins 11 to 20, D0 to D13) In this
    mode the function PIN() will also return the output value.

    We want to make it and out put pin so we choose # 8

    The instruction is then written like this SETPIN 2, 8

    That is all  we have to do to configure pin #2 as an output pin.
    From now on it is an output pin.

    Now when we tell pin to go high the injector will be fired.

    PIN(2) = high   turns the injector pulse on 5V We wait.  The wait
    time is a function of the mass airflow sensor. PIN(2) = zero turn the
    injector pin off.

    That's it.

    You have to get a Maximite and play with it. You can use a volt meter
    to check your program. Programming is mostly trying things and see if
    it does what you want it to do.


    Paul Lamar

    Paul,

    The hardware is the easy part I wanted to know where to start with
    the software after the pins are configured. The 2 RX 8 sensors wil
    deliver a pulse when the CDI fires for both rotors. The Maximite
    looks at this and will fire the injector and keep it on based on the
    MAF sensor voltage. What computations need to take place to determine
    when to fire the injector based on ignition data? The pulses and the
    time between the first pulse and the second pulse will give unit time
    per revolution. The injection would be based on an event that already
    occurred. Does the rear rotor pulse drive the front rotor Maximize
    for injection?

    Joe Berki

    Do you have a Maximite yet?

    I had a brilliant idea yesterday to drastically simplify the code.

    You use a FOR NEXT loop to time the injection,

    The code would then look like this.

    PIN(2) = high   turns the injector pulse on 5V when the CAS occurs.
    FOR y = 1 to xxxx NEXT PIN(2) = zero  turn the injector pin off.

    xxxx becomes proportional to the mass air flow reading.

    That's it !!!!!! Your done !!!!!!!

    One Maximite per rotor.

    If I could only get a patent on this ROFL

    Paul Lamar

    Joe, the fuel system is independent from ignition. It starts up as
    soon as the mass airflow senses the cranking sequence.

    Steve Carlisle

    Google FOR NEXT LOOPs.

    There is at least a half dozen explanations. I was amazed.

    ---------------------------------------------------------------

    One of the things you must do when converting the MAF sensor data to
    the required width of the pulse is know how much fuel will be
    injected for a given pulse width.

    I found a great site for calculation the size of the injector you
    will need.


    http://www.witchhunter.com/injectorcalc1.php <http://www.witchhunter.com/injectorcalc1.php>


   This screen capture
    jpg shows that you will need 4 injectors

    420 cc/min to generate 250 HP full rich at a BSFC of .65 pounds of
    fuel per HP per hour. The RX8 yellow injectors alone won't handle
    it.

    You will need 41 pounds per hour from each injector. 41 pounds is
    6.56 gallons per injector.

    Now we need to have a chart for the injector chosen that shows the
    flow rate per millisecond of opening time. I am working on it.


    Paul Lamar

    Paul, and more info from th guy that made the injector chart,
    http://users.erols.com/srweiss <http://users.erols.com/srweiss> <http://users.erols.com/srweiss <http://users.erols.com/srweiss>

    Steve Carlisle


    Paul,


    I am still missing something. The CDI will work off of the 2 RX-8
    sensors mounted on the from cover using the one tit trigger wheel.
    Does the 2 Maxi Mite setup then use the 2nd gen CAS to drive it? I
    thought for some reason that that CAS could be eliminated.

    Thanks

    Joe Berki

    You mean the thing that looks like a distributor? Yes you can get
    rid of that. The 2 RX8 CAS's will trigger the Maximites as well. You
    can get a clean CAS pulse from the LM1815's on the dual CD board.

    BTW You can use the Maximite to calibrate an injector. You can set a
    pulse width using the Maximite and fire the injector once and
    measure the amount of fuel that comes out. You can trigger this with
    a push button in place of the CAS.

    Then you can set another different  pulse width and measure it
    again.

    Pretty soon you will have a chart that looks like this.

    The max rotary injector pulse width is about 7 ms. At 8000 RPM it
    takes only 8 ms to make one rev. So if you are firing the injector
    every rev you are limited to about 7 ms.  This is known as "timed
    injection".

    You need an injector that will squired enough fuel in 7 ms. What we
    did on the TTC is upped the fuel pressure so it would squirt more in
    a given pulse width.

    The video was done differently.  It is just to show the test set up.

    I did it in my back yard so I would not smell up my office :-)

    The Maximite  is very very handy device for all sorts of things.

    Paul Lamar



    Paul,

    I think I have got an understanding please review and let me know if
    I got it right.

    We fire the injector off of the RX-8 sensor 20 degrees BTDC and at
    the same time the CDI fires.

    Because of the configuration of the rotary, plugs on one side and P
    Port on the other, the ignition of lets call it #1 chamber occurs at
    the same time #3 chamber gets a load of fuel.

    Is this close to being accurate?

    Joe Berki


   This is some great work, which is very helpful to many of us CDI/EFI challenged folks!

   Questions:  since the injectors are basically spraying into the intake manifold, is the injector timing really critical - especially at high rpm?  If so, are you saying that the injector duration can be simply a function of MAF, manifold pressure or rpm, unrelated to timing?

   If the stock rx8 CAS can generate an appropriate pulse, would it make sense to mount a second CAS 180 degrees off, and modify a trigger wheel to fire the  CASs at 22 BTDC?

   Scott Gettings

   Yes super simple. You just measure the mass air flow sensor and delay that long.
   You need a chart on the injectors your using so you know how much fuel is injected
   per ms. You can make adjustment in these numbers by adjusting the fuel pressure
   regulator.

   Actually to keep the software  real simple you can use a shaft connected to the fuel
   pressure regulator to manually control the mixture ratio. Similar to what we did
   on the manually controlled TTC waste gate.
   All the EFI must do is start and keep the engine running
   using the mass airflow sensor.

   It is also typical for a carburettor equipped  aircraft engine.

   Patent applied for :-) Not really :-)

   The injection happens at the same time as the spark.
   I don't think it is critical at all as the next face is opening and will receive the fuel and air.
   The P-port intake never sleeps :-)

   BTW to probability of software  bugs is proportional to the number of lines of software cubed.
   Same goes for understanding it

   Yes on the second CAS. That is what we are designing the CD ignition based on.
   Again simplicity. Easy to understand and easy to trouble shoot.

   Now you have a bit of redundancy. If one CAS fails the engine still
   runs on one rotor.

   Paul Lamar

To vary the injector timing to test for best BSFC and power at cruise RPM, a separate plate mounted on the front cover could be installed with slotted bolt holes. This way injector timing can be changed with minimal lines of code. For automotive use the spark signal is actually used for the next cylinder in the firing order. The signal is 45°to 90° before required but the software retards from there. I think injection timing is worked this way also.
On a rotary for aircraft, the running RPM is fairly constant so finding the optimum injection timing should be fairly simple on a dyno or by experimenting with making and documenting changes and test flights. Simple to try changing injection between rotors also by switching the wiring between CAS sensors for firing the injectors.

Dale Davies

   The Rotary Engine free News Letter. Powered by Linux. If you want off
   the list PLEASE let me know and I will remove  you.  ACRE NL web
   site. http://www.rotaryeng.net You Tube http://tinyurl.com/beqqxas
   Copyright 1998-2017 All world wide rights reserved.