Subject: Gear box help
From: "rotaryeng@earthlink.net" <rotaryeng@earthlink.net>
Date: 1/8/2015, 10:36 AM
To: AA-1-Me


  Paul,

  I am new to your news letter, I own a tailwind (w10) upon which I am
  thinking of installing a Mazda rotary engine.

  I also own a machine shop with a new 4 axis Doosan vertical mill,
  and electrical discharge machines, cnc lathe, etc.

  I would like to help with any production of parts for
  engines/airplanes/etc that comes up.

  Keep Flying!

  Randy OnMarkArms.com <http://onmarkarms.com/ Cimarron Mfg. 113A Mill
  St. Silver Cliff CO 81252 " T H A T E V E R Y M A N B E A R M E D "


  Your sure can Randy. Thanks for contacting us.

  We need a new gear box kit here in the states. The gears themselves
  are off the shelf Ford auto trannie parts that the builder can buy
  and install. That cuts the product liability exposure on your end.

  Take a look at some of these Rhino's 3D"s. and tell me if any of
  these ideas appeal to you. I also have a bunch of pictures and 3D's
  on rotary powered Tail Winds as well.


  I am also available for free Rhino help with about 12 years
  experience.

  Same for micro-controller programming on any automation projects you
     may have in mind. I have already done three. A coil winder, a round
     tube tig welder and a linear tig welder. I can also visit if you
  need me as I fly UAL standby nearly free.

  Paul Lamar


  I am hoping that someone will step up and fill the PSRU and
  electronics gap left when Tracy Crook retired.  I am currently
  building and will definitely be using a turbocharged rotary for my
  amphibious plane.  I will need a reduction unit and a controller for
  ignition and fuel injection.  My plane is a pusher so I would like
  to find a PSRU that would change the rotation so I could use a
  common 72" prop in the pusher configuration.  If the PSRU is
  reasonably priced then I would much rather buy one than try to make
  one.  There are a lot of great ideas coming out of this newsletter
  but for me it is trying to find time to work on all of the
  projects.  I would rather buy some of them and since I'm not a
  machinist the PSRU will probably be purchased if they are
  available.  Just my thoughts.


  Ron Reynolds


  The EFI is real easy to do it your self with this very easy to program
  micro controller. I can give you a lot of help to get you started.
  The big advantage is no secret source code. Open source. We can all
  work together and share our mixture maps.

  A. The Maximite comes completely assembled.

  B. Easy to learn by trail and error. You do some thing and the
  result can be or  is printed on the screen. Here is a simple
  program. Most BASICS are the same with minor variations.

  PRINT 2+2

  END

  The number 4 is printed on the screen.

  C. Buy a Maximite and play with it.

  D. You don't need to assemble it. It comes assembled.

  E. CircuitGizmo. http://www.circuitgizmos.com/

  F. Trial and error. Ask questions on here or Google.

  G. http://home.cmit.net/rwolbeck/programmingtutorial/
  <http://home.cmit.net/rwolbeck/programmingtutorial/>


  H. Hang in there Perseverance gets it done.



  In fact here is a BASIC program that is an excellent start for an
  EFI and EFIS.
  If it has a ' in front of it it is a comment.

  'SETPIN pin, cfg [,target]
  'Will configure the external I/O   pin   according to  cfg  :
  'The original Maximite has 20 I/O pins numbered 1 to 20, the Color
  Maximite adds
  'another 20 I/O pins on the Arduino connector. These are labeled D0
  to D13 and A0
  'to A5.
  '0 Not configured or inactive
  '1 Analog input (pins 1 to 10, A0 to A5)
  '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.
  'The starting line number of the interrupt routine is specified in
  the third 'parameter
  ' target  . This mode also configures the pin As a digital input so
  the value of the 'pin
  'can always be retrieved using the function PIN().
  'See also IRETURN to return from the interrupt.
  'See also:
  'IRETURN PORT P

  Mode 1
  'Mode 4

  For n = 11 To 20
  SetPin n, 9 'set pins 11 to 20 as open col 5 v output
  Next n

     Open "data.txt" For output As #1 'data file

     SetPin 8, 6, 10 'pin 8 low to high int gosub line 10
     SetPin 10, 1 'manifold press on pin 10
     SetTick 1000,20

  BakGnd:
  manpre = Pin(10) 'A to D on pin 10. may need to be scaled
  manpre2 = manpre * 10
  'Print manpre
  manpre3 = Int(manpre)
  PWM 20,manpre3,manpre3 'pulse width on primary injectors
  '''''''''''''''''''''''and secondary  20 times a second
  GoTo BakGnd


  10 RPM:
  RPM = RPM + 1
    IReturn

  20 RPMRoutine:
  RPM=RPM*30
  Cls
  Locate 30,30
  Print "RPM = ";RPM
  RPM = 0
  Locate 30,60 'col 30 line 40
  Print "Manifold pressure = ";manpre 'EFIS function
  IReturn


  Paul Lamar


  Keep in mind that in most computer programs there is far more code
  written to deal with errors than with the actual intended function.
  In the case of the aircraft efi, if a sensor reading is needed for a
  calculation and the sensor is not reading or reading wrong,  you
  have to be able to deal with that in real time but only after you or
  the computer have figured it out.  Now you get to apply this or
  another solution to all the sensors and possibly different
  combinations of sensors.

  Kevin Alderman

  Kevin Alderman
  President Aviation and Avionics Services Inc.

  That's where the Maximite shines. The VGA display  is a built in
  EFIS and any sensor errors can
  be displayed in plain English.  In fact it would be possible to
  display sensor output continuously.
  This is head and shoulders above what Tracy offered.

  IMHO in my 46 years experience with micro computers the Maximite is
  as significant as the Apple II,
  the Radio shack TRS80, DOS, the Macintosh, Pet, Commodore 64, Amiga,
    MS Windows or Android and Linux.

  The Maximite is a revolution in micro controllers. It is what DOS 32
  bit should have been.
  It is like have a full blown desk top computer with a 64Gig hard
  drive and an operating system
  the size of a thin paper back book.


  Paul Lamar



  I think you missed my point- regardless of the CPU behind the
  system, or even the code its written in there will come a point
  where something isnt right. If your code requires RPM and MP to
  calculate fuel PWM, and one of those sensors either reports wrong or
  not at all, then the system has no information to calculate the PWM
  with or even worse incorrect information. So, how do we deal with
  this in code? We could set the MP reading at say 30 unless the
  actual reading gives us a different value, or in the event of RPM we
  could go to an alternate PWM table that uses MP as a guide with a
  manual leaning as backup. What if both sensors go out, if we had a
  TPS we could use that as a baseline value to get at least a median
  fuel PWM so that we get something, and the list goes on and on.

  If the code is written "read PORT0" "read PORT1" take these values
  go through this formula or table and match a PWM value, if one of
  those ports isnt giving an accurate value your resultant PWM will be
  wrong and your engine will stop running.

  As far as boards go, I think I have settled on the ARM derivative
  that TI has put out as the Hercules line- these have dual ARM cpus
  running in lockstep that provide self testing while running. If
  there is a error or something is wrong the system can automatically
  reboot. 2 of these with dual sensors should provide a comfortable
  margin of safety if coded right.

  I have read through the MegaSquirt II code, most of the 6000+ lines.
  It appears that more than 2/3 of it deals with potential errors.

  Kevin Alderman

  If you have a sensor failure you will know it immediately by the way
  the engine is running or
  not running.

  Here is another idea. There is so much storage in the Maximite (64
  GIG) you could go back
  for the last hour and replay all the incoming sensor data in
  graphical form. It would then be easy
  to spot  the   bad or non existing data. Don't forget you have a
  complete O.S. under
  software control.

  At that point you have the option of running a completely different EFI
  program. Say one with out the mass air flow sensor or even a
  constant RPM,
  constant HP, super simple, no sensors at all, limp home mode. Similar to
  what the Wright Brothers used.
  http://www.rotaryeng.net/simple-cheap-555.html

  It is like having an OBD II recording on line and instantly
  available for review. None of
  them have built in O.S.'s with extensive storage. In other words a
  black box.

  Patent applied for :)

  It is hard to visualize what I am talking about unless you get a
  Maximite and try it.
  It is beyond the wildest imagination of what could or can be done
  with a MegaSquirt and
  what has been done with any other EFI computer including the many
  stock car
  computers with OBD II.  It is like having OBD II connected to a note
  book computer
  monitoring and recording  the engine  performance real time.

  I am sure some airliner glass cockpits are doing something similar
  or they should
  be. I'll ask my A&P wife Robin. She got and A in Airbus computer
  diagnostics.

  Paul Lamar

  The Rotary Engine News Letter. Powered by Linux. ACRE NL web site.
  http://www.rotaryeng.net <http://www.rotaryeng.net/ You Tube
  http://tinyurl.com/beqqxas Copyright 1998-2015 All world wide rights
     reserved.