Hi Paul,
I had to scan paper and put it into Word to get this out without a
lot of trouble. Numbers on the left are hex byte positions out of 7FFFF
total. Map starts at 70D6C, two byte word, 29 68 hex converts to 10600,
move decimal 4 positions left for A/F=1.06 at this position. The
numbers from 70CFC through 70D2B are 12 RPM's (horizontal scale) in
single precision floating point (990 RPM to 5500 RPM). Those from 70D2C
through 70D6B are 16 volumetric efficiencies (.125 to .9375). As you
can see, a large portion of the map is on or close to "27 10", which is
1.00. There are other Target Oxygen Maps for special conditions but
this is the main one.
Jim
Here you go!!!
Best regards,
Jim Langer
Racing Beat, Inc.
714-779-8677
Fax 714-779-2902
www.racingbeat.com
Thanks Jim. Very interesting. Perry Mick runs the stock 2nd gen
computer in his airplane and this could be very useful.
Right now Perry is working on an variable gain amp to modify
the mass air flow signal to lean his mixture. Here is a chance
to achieve this leaning by modifying the mixture map directly.
Perry, you will have to swap out your computer for an RX8 version
with CAN or OBD II input output I think.
BTW Perry uses his airplane to commute to work, weather permitting,
and has 450 hours so far.
Jim, Is this Techtom Mighty Map DOS or Ver-3 output?
http://www.technosquareinc.com/mitelite.htm
I'll have to fire up some of those old DOS computers
I have lying around.
I'll write a DOS program to convert this stuff
to decimal to make it easier for a human to read
and send you the program. Perhaps you can print the
data to a file which was a common practice in the old
DOS days as I recall. Also hitting Control P turns
the printer on and prints everything that goes to the
screen... also as I recall.
Do I have it all?
Is this a stock RX8 table or a modified table?
Paul Lamar ...No rotor no motor.
---------------------------------------------------------
OK What we have here I think is a two dimensional map of
mixture ratio verses volumetric efficiency.
In this case volumetric efficiency is the same as manifold
pressure. 29 inches of mercury is one. BTW It is interesting
that the best the RX8 intake system can do is .9375 times
29 is 27.2 inches of mercury. A P-port would be up to 1.2.
Jim, what I would like you to try is a lean mixture only at RPMs above
5000 RPM and volumetric efficiencies in the range of .5 to .8.
This I think is where the stratified effect starts to take
hold and the point where we are running the engine during cruise
in aircraft.
This is like driving down the road in fourth or fifth
gear at less than 60 MPH. Something one might do if one
were interested in immediate acceleration without down shifting.
Or sometimes people just forget to shift as the engine is so smooth
and quiet up to 6000 RPM.
Admittedly this is not as good as being in sixth gear with
a wider throttle opening at 3000 to 4000 RPM. Unfortunately
the positive effect of lower RPM and less engine friction is not
helping the high RPM stratified charge effect.
We can do an OBD II fuel consumption test by recording the
mass air flow and dividing by 14.7 to get the fuel flow
at say a steady 60 MPH in fourth and fifth gear and then
repeating the test after the map was leaned in this range.
Paul Lamar ...No rotor no motor.
-------------------------------------------------------
This is a stock 2004 RX8 main target oxygen map.
"OK What we have here I think is a two dimensional map of
mixture ratio verses volumetric efficiency."
No, it is RPM vs. Volumetric efficiency yielding A/F ratio.
Mazda defines Vol Eff (they call it "Load") as actual MAF divided
by theoretical 1.0 chamber filling MAF at that RPM.
"It is interesting that the best the RX8 intake system can do
is .9375 times 29 is 27.2 inches of mercury. A P-port would
be up to 1.2."
These target oxygen maps only go to 5500 and .9375 because
O2 feedback only occurs below this range. It shuts off at higher
RPM/Load. The system commonly reports Load numbers of 1.05 to 1.10
at about 7500 RPM where it peaks.
"Is this a stock RX8 table or a modified table?"
This is a stock 2004 RX8 main target oxygen map.
Jim
--------------------------------------------------
Thanks for the corrections Jim. I'll OCR it today and work on the
conversion
program. I took the liberty of rearranging your comments to conform
with our message format. My comments are in quotes.
So I take it... O2 feed back modifies or "trims" these numbers?
Perhaps intake air temp also "trims" these numbers. Is that correct?
Unfortunately I have no real idea what exactly is going on here
because I don't have the program source code for the CPU. For example;
what happens after 5500 RPM? Unfortunately that is just the start
of the high RPM low load range that could benefit from leaner A/F
ratios.
Unless we can obtain the program source code somehow it might
be easier to use a Megasquirt computer which the program source code
IS available for.
No doubt in my mind that leaning it out across the board
would not work.
You have certainly set out a difficult
task for yourself. If you could dump the entire program
in hex form and then disassembling it and comment it would be
a monumental effort. It may be written in "C" which makes
it double difficult to decipher or reverse engineer.
My son knows an EFI programmer that has worked for Ford in
the past. I'll try to contact him.
Paul Lamar ...No rotor no motor.
----------------------------------------------------
Hi Paul,
The O2 sensor is only functional through about 50 or so HP. It is
not functional above that level, and is not tied to any exact RPM or
load (may be tied to a pulse width). The O2 sensor has great
authority, and in addition to changing the mix, it builds an algorithm
for 3 levels of long term fuel trim - one near idle, one at light cruise
and one at high cruise. Interesting point - the top level of long term
fuel trim "sticks" when you go to open loop. This explains why "piggy
back" devices can give unexpected results.
The files we have for the PCM flashes include all programming in the
7FFFF bytes. It is compiled and not fun to decipher. I have a
disassembler, but have made no real progress.
We are now offering flashes for the PCM on the RX8 that changes the
fuel flow at higher RPM (leaner), turns the fans on earlier and
increases the red line. We have complete control of all timing, main
fuel, rev limiter, MOP, target O2, fans, etc.
Jim Mederer
------------------------------------------------------
Sounds like you have made a lot of progress in understanding
of what is going on.
http://www.racingbeat.com/FRmazda4.htm
Lots of good info in here.
Does the leaner mixture at high RPM show up in the MPG
one gets? Have you done any constant speed real time OBD II
captures in different gears at different RPMs back to back
with and without the flash? Good way to tell what is going on.
BTW see if your PCM memory dump program can output to a text or
spread sheet file. That would be great. I am having trouble OCR-ing
that data you sent in .doc form. Lacking a direct text file output,
please print it and scan it in at a higher rez (at least 400
dots per inch). Your scanner software should be able to save
the result as a jpg. That would be a better form than loading it
into MS Word. I ultimately need it in plain ASCII text file form
so I can write a program to convert to decimal from hex and format
it in the correct tabular form.
If the disassembled PCM listing is a Motorola processor instruction
set I might be able to help decipher it. There are a few tricks I
learned years ago. It involves writing programs that analyze the
listing.
BTW one more question I have been curious about. I know
you have a lot of contacts at Mazda that I don't have. Why won't
Mazda put the rotary in the MX5 Miata? Seems to me everybody
and his brother have a little two passenger sports car that
is competing with the MX5 Miata. Pontiac, Saturn, Honda etc.
A rotary Miata (or call it something else) would blow those
guys into the weeds due to lower weight if nothing else.
Especially if they did an all aluminum version of the rotary.
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.