"Bernd D. Ratsch" wrote:
> Excel....if you can do it.
Ooooh thats tempting. Actually Excel would be a very easy way to achieve that
goal. Problem is - I refuse to install any of Microsoft's products on my machine
unless I have to. Winblows w/Word & Excel viewers (on a seperate partition) is bad
enough. I guess I could use someone else's machine... When the idea came to mind
I actually had visions of a online java applet (that I would write of course) that
could select datafiles and allow users to select viewing criteria utilizing all the
variables avail from my scan tool. Then reality hit.....I'll explain below...
> >Shane said:
> >That jives with what I've measured also. To add more info - 29.5 in-Hg
> >w/key on
> >engine off and 1.5-8.5 in-Hg while running according to my scan tool on at
> >least 3 different trucks.
Hold the presses - I'm not sure the part about key on engine off is right.
Wouldn't this be the computer's way of measuring the ambient pressure? Or would it
always be 29.5 in-Hg due to the bias voltage applied to the 3rd pin. Or does that
just provide a vcc of sorts to the MAP itself? Reason I pose these questions is
because I could have swore mine read a much lower value today (80+ degrees compared
to the 50+ degree day that I took all those other measurements).
> >I also plan to post (to my website) 3D maps of timing advance both with and
> >without the MP computer as I have already collected the data. I'll be
> >plotting
> >rpm vs map with advance points in the z-plane just like the ones at that
> >website.
Ok now the sad part I said I'd get to above. It didn't take long for me to realize
how futile this attempt would be. Dont you just hate when you press that send
button too early? 8-) Now I know why the part about pressing the gas pedal
further causing pinging to get worse or better bothered me so much. My dang
subconscience knows SO much more than my conscience mind I guess. Here's the
skinny:
Regardless of how much data I (or anyone else) collects relating to rpm, map, and
spark advance - I doubt the "real" base timing maps will surface just by plotting
the points (without alot of hard reverse engineering anyway). Why not? Well thats
because timing is not just pulled from the "base" timing maps alone. Thats where
the term "base" comes from - the fact that during the PCM program's calculations of
what the next timing event should be - the base element is first extracted from
this map - then comes LOTS of other calculations based on LOTS of other events -
like delta-TPS, speed, & temperature for example. These other calculations
"modify" the base value. For example, if I plot a map of what I see and Bernd
plots a map of what his sees. Even if we were at the exact same temperatures - our
delta-TPS values and speed values (for each RPM and MAP points) could be
drastically different. Ok if you aren't confused enough yet 8-) try this link for
an actual implementation:
<digging thru bookmarks>...
http://www.diy-efi.org/gmecm/ecm_info/1227747/projects/747_ignition/spk747.txt
For those who are still with me. I know - that was deep 8-). Going back to the
idea of base maps vs "in-code, on-the-fly" calculations - once understood - you'll
know that the guys that make custom chips are just adjusting the base maps. Thats
all the chip industry has EVER done. Reverse-engineering the actual code is much
harder. Small adjustment tables can (and have) been found but the code is the
key. This is why it takes them awhile to get your combo just right - because all
they have to work with is the base (and maybe some adjustment tables) maps.
Furthermore, when you change some variable after paying all that money for a custom
computer - you have to send it back and they "try" again. If they/you had the code
- they/you could just program in a "self-modifying" section based on what you
planned for the future. So what am I getting at here? Unless you plan to some day
have that source code and the ability to modify it (never gonna happen) - the
better thing to do is just to junk that dang PCM and go with an aftermarket one
which WILL let you modify most of the stuff yourself (even on-the-fly). Better
still - is to develop one yourself (like I'm doing) so that you have TOTAL control
over all the code and not just the tables (base & adjustment). If you ever want to
dig in that deep - a good place to start is http://www.diy-efi.org . Be warned -
that stuff can get deep....real deep. I've been with the group (DIY-EFI and
EFI332) for over 3 years now and am getting closer to an actual working product
<crossing fingers>. The bulk of our code is in the following file:
http://www.diy-efi.org/~bbowling/efi/efibank.c <and you thought the first link
was tough 8->
I gotta stop - my fingers are hurting... Comments?
Shane
-- '96 IndyRam-HisIndy-MPI/TB/Pulleys/Coil/Wires/CompTAs '96 IndyRam-HerIndy-numbered(#142)"Track Truck" '74 Triple-Black Dodge Challenger Rallye 360 home-brew EFI R&D vehicle '68 Black Corvette Convertible 427 (For Sale)
This archive was generated by hypermail 2b29 : Fri Jun 20 2003 - 11:50:50 EDT