I was thinking about this [GPH] circuit, and it occurred to me the perhaps
it might be a better approach to go the digital rout to measure the on time
for the injector, rather than design an integrator circuit, scale it and
feed it to a meter? You already have discussed using a 6502 for ignition
timing, so why not add on the fuel monitor task. Analog design can be time
consuming, in that while a design looks great on paper, and the math is
correct, there is usually some fiddling and testing of actual circuit values
until you end up with the desired result. The digital route, I think would
be much less work, in terms of both design and implementation. You could
also easily program both fuel totalizer and real time flow functions, and
more.
Your results may vary.
C. Smith

Well it all depends on your programming expertise. I don't think the 6502 is
the way to go any more as it has almost disappeared off the radar :) However
a PIC or a 68xx depending on what you like to program best will work for
that. If you don't have that expertise the learning curve is steep.

As far as the ignition is concerned I have found a guy at Aero in Germany
that is willing to build us a multi-strike unit that fires the lower plugs 4
or 5 times 1 ms apart at 6000 RPM. I don't think he is going to use a
processor. If he comes through with a unit and it reduces the BSFC the
rotary world will beat a path to his door.

BTW Mario, please call this guy and ask him how it is coming.
His name is Toni Kurpa. +49 02303-257070 He does not do email.
www.federbandschelle.de He is very sharp on CD MSD ignition systems.

I will test it on a dyno and let everybody know if it reduces the fuel burn
or not. If he needs money let me know. All the rotary guys over there should
get together and sponsor this promising project. Mistral, Austro and
SuperTec are also setup to test it for lower fuel burn on their dynos.
High rotary fuel burn is a perception here in the states that must be
addressed if the rotary engine is going to succeed.

This data is good evidence that a such a multi-strike unit will reduce the
fuel burn. So far people have been using piston engine ignition systems and
what is need is a ignition system that is optimized for the rotary's 270
degree expansion cycle. IMHO this should be the highest priority.

Paul Lamar

Well, that's all great news. Never the less, there is still a solid argument
for a PIC/microcontroller type device for some engine monitoring functions.
I purchased one of the DATAQ DI-148U, while it is limited in very low signal
level capability, it will be a great tool for analysis of the engine
systems, when I reach that point. Presently finishing up stripping the 13b,
and welding up an adapter for the engine stand. (Plans of which I will post
for the group, when completed). The data can be ported directly to an excel
spreadsheet (with the right software).

Strangely enough this is exactly the method that our Brush/Dynamotive dynos
use. Analog data is collected, A to D conversions are made, all real time
data and variables are stored in tables, then the operations are computed,
the results get ported to another table, with some visual basic creating a
display of the desired information.

My argument is that all injectors have some variation in flow. This is born
out in actual flow bench testing. This variation (K factor?) can be
compensated for quite easily with a digital monitoring system, variables are
just entered into a table. Conversely, an analog system would require some
change in component values. Pressure fluctuation could easily be compensated
for as well, using a fuel pressure sensor. Getting all this to work in a
pure analog circuit would be quite a puzzle indeed, and each installation
would have to go through the compensation effort themselves. It may be that
the accuracy or precision of the pure analog doesn't warrant compensation at
all, but I don't think we've reached the conclusion of the EFI design yet.

There are quite a few options out there for micro controllers, they have
large user groups, as well as numerous tutorials to help one get up to
speed. I just saw this http://www.omega.com/pptst/ELC_PLC.html inexpensive
controller, although a lot of the stamp/micro controllers are cheaper. It's
just one of many options.
At any rate, I just thought I'd present a train of thought, and a possible
solution.
C. Smith