Subject: Results of the poll
From: Rotary Engine
Date: 6/15/2007, 7:53 AM
To: AARotary Engine

Nine for mechanical and fourteen for PICAXE.

Several are building PICAXE systems and only Hurant Karribian is building a
mechanical system as far as I know. Since I have considerable experience
building real time computer systems both hardware and software  it is no real
challenge for me. Been there done that for 25 years. Instead I opt for the
mechanical system. More interesting from my point of view.

I have some advice for those developing PICAXE systems. KEEP IT SIMPLE.
Trade cheap computer hardware for software whenever possible. Hardware is
getting cheaper and cheaper while software is getting more and more
expensive in terms of labor cost. This cuts the cost of low volume
computer systems in terms of time to develop. This does not apply
to mass produced in huge quantity cell phones of course.

Use one PICAXE per function whenever possible. For example; use one
PICAXE to control the mixture and another to send data to a small
paper back book sized computer for the cockpit display.
The mixture control code is greatly simplified. Load the
mass air flow sensor A to D. Goto the table address indicated.
Read the table value and store it in the PWM fuel pump location.
Go back and do it again.

****** Errors and bugs in software are proportional to the number of lines
****** of code CUBED.

This also leads to hardware redundancy. If one computer fails
your entire system is not down. Another example. Use a separate $10 PICAXE
for each rotor mixture control. You only need one or two analog to digital
converters for controlling the mixture. That translates to a cheaper PICAXE with
fewer input output pins.

Don't bother measuring the intake manifold pressure. It is less sensitive
at WOT than a mass air flow sensor. That is one reason mass air flow sensors
are universally used on car EFI systems these days. Manifold pressure can be all
over the lot at wide open throttle and is a function of WHEN you measure it.
Particularly when a slide throttle is used mounted close to the rotor housings.
If an upstream plenum located pressure sensor is used to measure intake manifold
pressure with one throttle upstream of the sensor manifold pressure sensors work
better but still not good at WOT. This is what Tracy is doing on his side intake
port manifold systems.

The mass airflow sensor automatically measures intake air temperature so
a separate intake air temp sensor is not needed saving one analog to digital
converter in the computer. AND the time and code to read and process
the information for that matter.

****** REMEMBER Errors and bugs in software are proportional to the number of
****** lines of code CUBED.


I highly recommend trying pules width modulation of the fuel pump instead
of using traditional timed fuel injection. Timed fuel injection does not
work at 7000 RPM anyway so it is over kill for a rotary aircraft engine.
You must have a fuel pump in any system but you don't need expensive individual
solenoid operated fuel injectors. Control the pump to send only the amount
of fuel needed. This greatly simplifies the plumbing and eliminates the
fuel pressure regulator and return flow lines and valves.

I'll be here to answer any questions and help in any way I can.

BTW the newsletter may be erratic for the next few days as Robin, Suzzy and
I are driving up to Placerville leaving this morning. If anyone is interested
in buying our lot up there let me know. It is off "He Went That-Away" road.
It all depends if they have Internet in the hotel we stay in.

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.