Roger, While you are kicking ideas around I have some
suggestions learned from a few weeks of Internet searching. I
have selected a ford mass air flow sensor which is a 2 wire
sensor (one heated one not) and is compensated for humidity and
temperature giving an output of 0 to 5 volts. I chose a 5 liter
mustang 90MM version with a die cast aluminum body and have a
late model version with ABS housing as an alternative. For input
into a computer chip if the voltage is converted to digital then
cubed plus a fixed number it is extremely close to linear over
more than the flow range used by a rotary engine.
Second discovery is the
Ford EEC series of computer. The earliest were speed density,
not the best choice. The rest are all mass air flow since 1987.
The EEC-IV can be reprogrammed by installing a chip on the test
port which overrides the internal maps. The EEC-V is
reprogrammed by reflashing the memory maps.
The EEC computers are used in racing and that includes some F!
engines, formula ford and drag racing. No other OEM computers
are this widely applied. The EEC has two methods of fuel management
either by mapping tables which are used when not at wide open
throttle and include the oxygen sensor input as feedback in closed
loop operation. The tables are in a matrix with each group of inputs
outputting a value to a higher level map. When the wide open
throttle contact is closed the system ignores all tables and
calculates the required fuel only requiring the mass air flow sensor
and if one is present a manifold absolute sensor and a barometric
pressure sensor. For our uses on wide open throttle mode an
adjustment in the 0-5 volt MAF output can alter mixture.
I paid $27 for the MAF sensor and $41 for an EEC-V from an F250
truck on E-Bay. For this same test engine I will be using the EDIS
ignition also from Ford at a fixed advance as a stand alone
system. The truck EEC-V was chosen for the eight sequential
injector outputs which is required on a four rotor Mazda. The
four cylinder computers are better suited for a two rotor.
There are at least two rotary engine aircraft using the stock
RX-7 engine computer. I discounted these solely on the basis of
cost. They were all more than $300 at salvage yards and I would
have required two of them for a four rotor. I bought three RX-8
engines and none of them came with the computer and the wiring
harness was cut off at the connectors. The harness is going for
$180 which is also discounted on the basis of price.
I drive a truck I bought new in 1987 which is not garage kept
and have never had any issues with the engine control systems.
I have also had two engine failures the same month renting
certified aircraft. I am therefore more comfortable with a
simple, smooth running robust rotary core engine with
automotive systems reliability for control.
george grimes
George,
You are touching on a subject that is all to common in the aircraft
community.
Why is your 1987 outside kept truck reliable, when the
...several times more expensive aircraft is very maintenance
heavy, and have fuel, ignition and mechanical related failures
more often than then anyone like to.
This is an issue I debated on another forum not too long ago,
and I think it is mainly from many aircraft owners non
understanding of how things deteriorate.
An engine that is run often, like your 1987 truck, will always
be coated with oil, but I am sure you have gone down the line of
aircrafts. many parked outside and I am sure you as well as I,
and all of us, have seen aircrafts that have been parked at the
same spot for months, and never moved.
None of them are pickled, at the very best there is a cover
over the cabin glass, but not over the engine, and they sit
there, sit and sit and sit.
Once in a blue moon the owner comes out and start the aircraft
and go around the pattern.
Every time he starts that engine it is a demolition derby going
on inside. There is absolutely no oil , nothing what so ever in the
engine, everything has gone to the bottom long time ago, and as
any four stroke can not have any propeller position ( 4 cyl and
up) where all the valves are closed, you can bet on rust in the
bore, and valves as well.
The aircraft has been sitting through a couple of three days
rain storms, and in between we have the fog.
The exhaust is very simple and straight compared with cars, and
can almost be considered a straight exhaust, ( as many in the
experimental field have) , and internal weather systems WILL
develop inside that engine, bringing moist to any and all
exposed part.
At least by doing a semi pickled engine, it would protect the engine
for a long time, simple things, like getting a spray can of engine
fogger, turn on the starter, ignition off, and spray into the intake
for a couple of revolutions. That will leave the engine with oil
that is coated, that will not leave the cylinders or valves, it
will stick there, and sit and protect the engine.
There is a lot of engine oil fogger products both in marine and
car parts stores.
It is funny, in the marine community this is clear as a crystal
ball, while the same practice in the flying community, it is
clear as mud.
A simple made plug, into the exhaust pipe will do wonders, and a
cover that will fit tight over the intake will do the same wonders.
The majority of airplane engines does never reach the stated overhaul
hours, before it needs to be overhauled again, while engines sitting
in trainers, is known to go well beyond 3000 hours, and still
run strong.
If you KNOW, you will not run your airplane engine in three,
four ( at most five) days or more , spray it, and plug the
exhaust, and intake, if stored in a non heated steel walled
building , it gives only marginal protection, use the same routine.
I don't care if it is a piston engine, rotary engine, steam
engine, or warp drive engine, that is about the time limit you
have, three, four,.... five days at most, after that time, you
have a dry engine, and then you are at risk.
I used to say, and still make that statement...if it ever rains
at your location, ever, do it.
Ok leaving that and back to FI systems,
George, You seem to have ventured into the Ford systems, and
have luck with it, and I can see from what you are saying that
the Ford system have probably the most versatile options,
suitable for a big range of options.
I will keep that as an option, ( I really will, I will pursue
the Ford idea based on your description) while looking for a
suitable FI system.
I stumbled on the 555 system, and had an idea that it would be
a solder together type thing, but realize now that this is
highly experimental.
I don't mind to experiment, but I rather have something that I can
venture straight into, and I am feeling more and more that the 555
system still needs development.
I therefore pose a question to the whole rotary community ( or
others) that have seen the 555 system.
Are there any workable 555 systems out there, based on the
description from the rotary website, that is working, and functioning?
Similar systems work, we know that, Klaus Saviers system works
wonders, and it could be bought, so could a couple of other
systems, in kit form, but they are developed to a point where
they will work.
I really like the idea with low Amp draw , no computer, and
simplicity of the 555 system, and very low cost, thats a real
winner in my opinion. The system have all the attributes of a
winner, but I start to hesitate if I need to be a developer
also, ...I can do it to some extent, and probably have to,
because whatever I get, it has to be tuned to my specific
application, but are afraid of being drawn to heavy into the
basic development loop, ....I just want an FI system, thats all.
So, are there any 555 systems up and running now?
Roger Isaksson
http://www.team-bhp.com/forum/motorbikes/4615-fuel-injection-
india-bikes.html
However I am a bit more optimistic because a
fuel injection system is rather simple. A 555 CMOS version is a
miniature
fuel injection system in a one dollar 8 pin IC chip. One pin is
RPM input another zero to five volts input for engine load and a
variable pulse width comes out another pin for the fuel
injector. I have enough experience designing computers with IC
circuits dating back to 1975 so I know it will work. Prior to
that I was building radio control model airplanes (with tubes)
as a teenager in the early 1950's. Read the "Who is Paul Lamar"
section on my web site to see what I have done over the years.
http://www.rotaryeng.net/who.html
BTW I learned to fly Cubs and Champs at age 14 in exchange for
washing airplanes.
If you build a 555 system I have time to help by email.
Paul Lamar
Roger,
If you want something you can get up and running fairly quickly,
George is on the right track. The Ford systems are reliable
and readily available
(read inexpensive). Ford has a book available on tuning. It
is sort of directed at the Mustang crowd but covers all the
systems. It has been a few years since publication, but what
you need is in there. A system for a 4-banger is the deal.
If your AC has dual systems then you install two and can A/B
them so each can be tested at runup individually.
If you need more fuel flow while running individually, upsize
the injectors for the displacement increase from what the
ECU was set up for. That will get you close.
You could use the O2 sensor with Mogas during the tuning
process. There is a couple of places to put the
potenciometer for leaning. Then remove the O2. It can be
tuned without the O2 also so you can use avgas.
These are set up since going to sequensial for a MAS. Newer
ones are designed for coil on plug. Nothing wrong with the
ones that use the waste spark for your 0-200. A bit less
expensive than the individual coils.
Dale Davies
-------------------------------------------------------------
Dale ,
I am starting to lean towards doing the 555 system, it's a personal
challenge...I can see from the correspondence that you all have,
that in many respects, you are well over my head, and I want to
struggle through something like this, I really don't mind it,
but don't want a big research project either, so I think a 555
system will put me about where I need to be. ....but by all
means, I wouldn't mind looking at the Ford system as well,
George mentioned it earlier, and I promised to follow up on it,
where can I get info about the Ford FI system?
Roger Isaksson
In my computer experience basing anything on a sole source
device is sooner or later going to bite you. Manufacturers move on.
They MUST continue to lower the cost by redesign or go out of
biz. The 555 is made by half a dozen companies world wide.
It so ubiquitous you probably have two or three in the house
with out knowing it :) Look inside the toaster as a start :)
Paul Lamar
Roger, To research fuel injection the internet is a good start.
Search "EEC EFI pinout" or EEC modify tuning" " EEC-IV"
"Ford EFI" "Ford MAF EFI" and occasionally add wiki to the search
perameters. I also bought a book on E-bay called
how to understand, service, and modify Ford fuel injection & electronic
engine control written by Charles O. Probst, SAE
The one I have is for 1988 to 1993 This book was an easy read and
fairly detailed. This one is foe EEC-IV only so you might want a newer
edition to cover the EEC-V. After that the ignition module is
incorporated into the ECM rather than stand alone and a lot of the body
control module functions are interconnected with the engine computer.
They all have on-board-diagnostics. The -IV and V run with many of the
sensors disconnected such as MAP EGR coolant temp, oil pressure
and O2 sensors. The oil temp, and pressure and coolant temp will be
on the glass pannel so sensors will be present and can also be
input to the EEC-V. The majority of the sensors output 0 to five
volts analog and are converted to digital when input to the EEC
module. The module runs on assembly language which is many times
faster than C++ or fortran which makes sequential injection
calculation in real time. Most of the sensors we don't necessairly
use are only in use at part throttle or engine start. The MAF and
crank angle sensor and if you want the part load tables used a
throttle position sensor.
I know there is a lot of other stuff in the EEC such as EPROM
chips and RAM and precision voltage control, and then there is the
program itself in the ROM. All this stuff is integrated circuit
chips, resisters, diodes, and capacitors and has the possibility of failure
from static electricity. They are hardened in the case to a small
degree. All the ignition/EFI systems we are discssing are also made
out of integrated circuit chips, resistors, and capacitors and have the
identicle overvoltage and static electricity vulnerability. I assume
the OEM stuff has Zenir diodes or other components on the inputs to give
some degree of protection and most of the outputs are actually grounding
the various solenoids to complete a circuit so that grounded wireing
doesn't function but is less likely to damage anything.
I heard some hanger flying one stormy day that I put down to
wait for weather to pass. Soneone said he would never fly with an
automotive engine with points and a condensor to worry over. He was
oblivoius to te fact his magnetos have these parts still and cars havent
used that technology in 30 years.
I was going over the material in the link to Tracy's EC-3
and was impressed with the redundancy and ability to reposition
toggle switches to use the cold start to replace a non functioning
injector by firing the working ones with twice the pulse width. It
also seems that setting the injector flow rate correction before
working on the mapping adjustments would be the starting point.
Getting the mixture knob at the straight up or neutral point would
seem to reduce the need to offset the entire fuel mapping adjustment.
george grimes
I think I have that book up in the hangar George. As I recall there
is no source code to the EFI program in that book. Usually that
is a trade secret. Consequently there is no way to change it
and reassemble or recompile it. I had a 1994 V8 Cougar at the time
and I was having trouble with the EFI. Turned out to be a speck
of dust on the hot wire mass airflow sensor. The engine ran fine
except the check engine light would come on.
If I were taking the computer approach I would use ZBasic products.
http://www.zbasic.net/
They have a free BASIC compiler that is very VERY fast. Not fast enough
for timed injection but plenty fast enough for batch injection.
It is real easy to program through a USB port.
As computer languages go it is easy to learn and I can help
anyone with any problem that they may have using it. I have the top
of the line computer and it only cost $90 bucks. Great deal if
somebody wants a VERY powerful control computer.
Can be programed in easy to learn compiled BASIC in weeks rather
than months or years using asm or C. I am pretty proficient already.
Paul Lamar
For Rogers O-200, the controller could be any of a few as you are stating
Paul. Running with the EDIS coil. This simplifies things a little. Then the
engine uses dual plugs. A second system installed complete with a second set
of coil, fuel pump and controller gives redundancy. You can run on one and if
there is trouble switch to the other system. There are tuners available for
some of these now. It would just take contacting them to see what is available
and what you can do with it.
Dale Davies
--
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site.
http://www.rotaryeng.net
Youtube key word PaulLamar2
Copyright 1998-2010 All world wide rights reserved.