Could this be another possible controller. Certainly the programming
 language will have a shorter learning curve.
 C. Smith

 Possibly but I'll bet it is not $90 :)

 Paul Lamar

 Hey I happen to be an old PLC programmer and controls engineer. It
depends
 on what you want to control if you can tell me. If not just give me
 examples. The PLC was designed for electricians to use and trouble
shoot
 equipment with. It uses a hibrid program that mimics the ladder
diagrams
of
 old that are an industry standard. You occassioinally see them in
some
 washing machines or appliances etc. They provide high speed universal
 troubleshoot unlike written cryptic programming code like C or C+
CPP and
 so
 on. Also the nice thing about what is refered to as Ladder logic is
that
it
 can usually be changed on the fly in realtime without shutting down,
 compiling and downloading. Depending on what your using it for,
this can
be
 good. Also ladder logic is intuitive and once you understand it you
 understand most PLC's and many DCS systems. You could essentially
make
 changes while flying to logic or just parameters, values, setpoints
and so
 on. I can help you if you decide to go this route and don't be
shocked at
 the price, some are very reasonable, some are big bucks. Allen
Bradley
 www.ab.com/ is a top of the line PLC that is very common in
industry if
 you
 want a high end reference.

 Also do you want an input screen called a MMI or HMI. The MMI is the
 Man-Machine Interface of old. Then women came into industry and they
 changed
 the name to HMI, Human-machine interface. It is essentially a
monitor with
 on/off buttons and so on, touch screen type of thing that is
programmable
 to
 your specific operation.

 Hell I can flood you with free info to fill your library if you
want. I
 love
 this stuff and can talk forever about state of the art controls.

 This particular PLC is made by Galil Motor Controls - (there are
millions
 trying to get in the market)

 http://www.galilmc.com/learning/tutorials.php

 http://www.galilmc.com/support/catalog.php  Order their free hardcopy
 catalog here or get the ecat. I'm attaching one for you.

 http://www.galilmc.com/

 How many I/O's do you need and of what types. What type of
operations do
 you
 have going on? AI, AO, DI, DO and so on. I do process control and
 automation, robotics etc if I can help you in any way. I haven't
heard any
 discussions about controls until I seen this post.

 Nevada Bruce

 We have up to ten zero to five volts analog signals to record and
display
 on a CRT. A minimum of ten bit resolution.

 We have four injectors to control with pulse width modulation.

 And we have one RPM to measure in real time and record and display.

 We have four spark plugs to fire. Two of which can be fired with
 the same signal.

 All this must be done in 8 milliseconds.

 Paul Lamar

Thank you, that is a start. What I didn't see was an encoder of
sorts for
precise timing of these devices, I suspect this is for a rotary engine
correct, so in other words if you want these outputs to happen at a
precise
time relevant to TDC (in recips), I think we would want a pulse train
coming
in referenced to TDC or what ever that is called on rotaries ( I get
to
learn too) Am I correct or is there something I don't know about?

Q: Is your intent to build an ECU where you can control/modify it?
Q: Is this in addition to a ECU or stand alone?
Q: Have you considered redundancy or triple redundancy for safety or
is
there some type of manual backup?
Q: How many processor units do you want in the end? One or more? In
other
words, will you be making more for other engines? I ask this because
possibly a PLC wouldn't be the best way to go. I don't have a feel
for it
yet.
Q: Have you explored the world of micro controllers yet, with these,
once
you get your prototype perfected, you can black box your code and
produce a
zillion via china or do it yourself like copying DVD's (what MIcroSoft
should be doing with their operating systems if they wanted to improve
their
product - silly statement - of course not)

We have up to ten zero to five volts analog signals to record and
display
on a CRT. A minimum of ten bit resolution.

This can be done simple and with an operator screen once everything
else
is resolved.

We have four injectors to control with pulse width modulation.

Q: can we define this a little more? I assume by pulse width, we
need to
control duration of a high signal like a DO (can we use a DO?) at
the same
amplitude, ie: the longer the duration of the DI, the more fuel, is
this a
correct assumption? This also needs to be timed to a specific crank
position
reference signal is that correct?

And we have one RPM to measure in real time and record and display.

Q: I think here we would use the Crank Position and location (help me
with the proper description of what I'm trying to say here please)
we can
use this for many other computations relating to start stop time to
position.

We have four spark plugs to fire. Two of which can be fired with
the same signal.

Q: can these be fired with DO's Digital Outputs right? But the
timing of
these needs to be variable as do the injector signals in
relationship to
the
crank shaft position (or rotary cycle timing) I'm new and ignorant
of the
Rotary engine, I understand the rotor cycle to crank shaft is 3:1 is
that
correct? We need a way of identifying crank shaft position from rotor
position so I need the proper words to describe each including a short
acronym of your choice. This is to communicate between each other to
avoid
confusion. Building and sharing a glossary would be good to start
with for
someone.

All this must be done in 8 milliseconds.

Q: for my benefit, can you help define the complete rotary cycle
posible a
hand drawn scetch to begin with, when the events will ocur in the
cycle,
the
max RPM, the min cycle width (cycle frequency at the max RPM, cycle
freq at
minimum RPM)

I am seeing some DO's (don't have a handle of the quantities yet,
probably
one for each injector and one for each spark plug.)

Probably a digital encoder (pulse train) for timing purposes and fine
control and a DI for TDC crank position. The RPM can be derived from
this
DI.

Q: can you elaborate on the .008 seconds in this discussion? Is that
the
maximum cycle width at max rpm? (I haven't done any of the math yet
until I
define the full scope of what is happening. I don't want to assume
things
and get them in my mind.) Also what is the best resolution we are
striving
for, how many degrees, or parts of a degrees do we need to be able to
control to. These are the things that affect the type and cost of the
processor, and what type processor to use. (processor speed and so on)

Q: Communications: What method do we want to describe a partial
degree in.
(like degrees, minutes, seconds or decimals. I prefer using decimal
but
it's
your call. I just want to make sure we are always talking apples and
apples.
Example: does 1.50 mean 1 1/2 degrees or does it mean 1 degree and 50
minutes?)

Example: turning on and off room lights are the low end cost, where
the operation of a high speed I/O is more in the high end. When that
room
light is required to turn on and off in a specific time frame at an
absolute
time, it becomes more expensive. Where a $90 processor may be able to
 handle the analog inputs just fine, it may not be able to handle
the more
intricate high speed DI's, DO's and pulse input's as frugally.

Q: do you want data logging and data storage capacity? You can be
thinking
about it. We don't have to know up until it's time to spec out and
purchase
something.

Q: How many people are involved that can be collecting sale
information? I
can tell you that this market is vast and changing. There are new
things
coming out every minute unlike some industries like automotive where
it's
about 75 years behind the technology curve.

Q: Are you considering monitoring/logging various temperatures,
exhaust gas
oxygen? These are simple and cheap but need to be defined in the
process.

I hope I didn't overwhelm you with stupid questions. I don't know
you well,
I know you are very smart however I don't know your strengths and
weakness
in particular fields so please take no offense at my dumb questions.

Also is there a test engine on a test stand somewhere that can be
used?
If we can define our exact operation, it does not have to be said
what it
will be used on, we can let the vendors to the majority of the
research on
what is best for this application. They have applications engineers
that do
nothing but that. My suggestion is to spend 90% of our time/energy
brainstorming and defining a good spec to give to multiple vendors
and let
them tell us what they have, the cost and why we should use their
product.
This is a vast area. It's possible that with research there is one
already
out there that can suit our purpose so we need to decide too, do we
want to
invent this wheel from scratch for some reason or will we accept
someone
else's work? Many things to think about. I'm only now got in this
discussion
so possibly these things have already been discussed and resolved.
I'm only
thinking of the many possibilities of how to skin this cat . . . I
love
cats
so I don't mean that literally. I should use the word politician,
everyone
hates them :-)

Bruce

http://www.parallax.com/

http://www.c-stamp.com/index.htm?gclid=CKeLoK3XwKACFQMhDQodwCoLUg



http://www.google.com/search?source=ig&hl=en&rlz=1R2ACAW_enUS370&q=microcont
rollers&aq=0s&aqi=g-s1g-sx9&aql=&oq=micro+controllers
Also see:
Programmable logic device
Field-programmable gate array (FPGA)


Brief overview of a micro controller v. micro processor.


A microcontroller is a specialized form of microprocessor that is
designed
to be self-sufficient and cost-effective, where a microprocessor is
typically designed to be general purpose (the kind used in a PC).
Microcontrollers are frequently found in automobiles, office machines,
toys,
and appliances.
The microcontroller is the integration of a number of useful
functions into
a single IC package. These functions are:

The ability to execute a stored set of instructions to carry out user
defined tasks.
The ability to be able to access external memory chips to both read
and
write data from and to the memory.

Basically, a microcontroller is a device which integrates a number
of the
components of a microprocessor system onto a single microchip.

So a microcontroller combines onto the same microchip :

The CPU core (microprocessor)
Memory (both ROM and RAM)
Some parallel digital I/O
Also, a microcontroller is part of an embedded system, which is
essentially
the whole circuit board. Look up "embedded system" on Wikipedia.
The difference is that microcontroller incorporates features of
microprocessor(CPU,ALU,Registers)along with the presence of added
features
like presence of RAM,ROM,I\O ports,counter etc.Here microcontroller
control
the operation of machine using fixed programme stored in Rom that
doesn't
change with lifetime.


http://wiki.answers.com/Q/What_is_the_difference_between_a_microprocessor_an
d_a_microcontroller

 Nevada Bruce

Most of the answers are here.
http://www.rotaryeng.net/fuel-injection.html
And here.

http://www.rotaryeng.net/mech-fuel.html

Paul Lamar


Bruce
I looked into PLC engine control a while back. Here are some of the
things
you will need in a micro PLC

Multiple high speed counters with appropriate output impedance. Many
are
capable of 20 kHz.
12 volt cpu unless you want to add a dc converter
Thermocouple inputs if you want to monitor temps.
A handful of 12 volt dc I/O
Most have 10 or even 12 bit analog but may be limited with above
options.

If you can't meet all those conditions then you end up adding
additional
hardware and the PLC quickly looses its advantage. I noticed also
that some
units have PID capability- Paul would PID loop make the tuning logic
any
simpler?

It is hard to find all those things in a fist sized low cost box but
keep
looking-I ran out of time. The stepper motor motion control stuff
usually
has a bunch of high speed I/O -so maybe something in that niche.

Rick Pickett




"A Proportional Integral Derivative controller (PID controller) is a
generic control loop feedback mechanism (controller) widely used in
industrial control systems. A PID controller calculates an "error"
value as the difference between a measured process variable and a
desired setpoint. The controller attempts to minimize the error by
adjusting the process control inputs."

That is widely used in low RPM low power automotive EFI systems but
so far not used in aircraft EFI systems. I was thinking of trying it
and
use EGT as the controlling input. The tricky part is there are
two sides to the curve. The EGT first increases and than starts to
decrease past peak EGT.

Paul Lamar


Thanks Rick. Here is where I would start. A spec needs to be written
first
defining all functions, all I/O and what is needed to be achieved.
Then it
can be sent out by many people to many vendors and allow them to do
the
application work and tell you whether their product will meet your
needs.
This is power steering as I call it, 90% thinking v. 10% work. For
those who
do not understand controls, I can recommend some good basic books
where a
person can learn as they go. It's a very interesting and rewarding
profession. It also parallels our brains and minds. What you learn
will
apply to understanding how our brains and minds work. Our brain
being the
hardware, our minds being the firmware and software.

By the way, what are you thinking of stepper motor control for? Out of
curiousity. This is usually used in robotics but I want to hear.

We should start by defining the process, analyzing it, and writing a
spec
and then agreeing on it. Once that is done there is a super market of
vendors. Not only industrial processors but auto and many others.

There is another big factor, what are the end results you want. Are
you just
experimenting with a one of a kind or do you want to create a new
market for
many aircraft or engine owners?

Do you want to wind up with an inuitive adjustable kit type program
or a
propritory black box afair?

For instance Ladder Logic is a high level program designed for high
speed
trouble shooting with the qualified electrician in mind. He can call
up a
piece of industrial equipment from his shop and immediately see the
problem,
order the part to be picked up whereupon he can have the machine
going again
in literally minutes as opposed to hours. That really isn't what I see
needed in this case but I stand to be corrected if I'm not aware of
something.

Hard time programmers, coders, alway argue that their code can easily
replace the ladder logic. I agree but then only the programmer can
adjust
it. The ladder logic is a common language to electricians and
instrument
technicians who actually do the work in industry.

Bruce


Bruce, please send all messages at the end in plain text. Thanks.

Paul Lamar


Bruce,

    I would be very interested in a 'good basic book where I can
learn as I go'
on controllers.   I work in health care, but took a few years of
engineering
for fun in the 1990s [2 yrs electrical engineering w digital logic, 2
years
material science, and 2 years optical science, and other math, physics
etc].
I remember I got as far as programmable logic arrays when I decided I
was losing interest in engineering as a career field instead of a hobby.
Thank you for your kind offer and help.

Gene Kahn