SpeedPro comments

[Carl Summers]

Hi Jeff,
      It's "nice" to hear from others that have had the interpolate prob
with the self tunes of aftermarket EFI....I don't necessarily want to look
at your code that helps with this...but I don't understand why the people
that write this stuff don't understand that if 80 kpa at x rpm is larger
than 90 kpa at the same x rpm...WRONG...SMOOTH...get a grip from another
area or stop the self tune....obviously we can't all sit at a particular
load cell.  I have never looked at the SpeedPro software(hint), but would
love to compare it to what I have seen...I worked at a company with a very
smart individual that could enable a good "self tune" strategy but it is so
many man hours away from becoming public it isn't funny.
    I admire your ability to tune, but I would like to say that the strategy
of pw vs. rpm for a nitrous engine isn't so bad considering nitrous
technology....I have been playing with nitrous for many years and although
there is a science to it...it is still based on flow and cannot be closed
loop at this time....I would love to hear arguments regarding
this....especially with EFI......I have not seen a wide band O2 correctly
measure AFR on a nitrous engine for many reasons I guess I am still
supposing about....and just so were not talking about theory I will state an
example of what I have tuned....mild 400 chev..hydraulic roller, AFR's,
Performer RPM,etc etc...331hp at the tires....on nitrous 651 at the
tires....no burned plugs, no bearing probs, no audible knock, nothing
abnormal when watching the piezo to monitor knock.  It is good to hear from
people like yourself and I mean no flame by what I am stating...this is how
we learn from each other. ttyl
-Carl Summers


-----Original Message-----
From: owner-diy_efi at diy-efi.org [mailto:owner-diy_efi at diy-efi.org]On
Behalf Of Jeff Bromberger
Sent: Tuesday, September 19, 2000 2:49 PM
To: diy_efi at diy-efi.org
Subject: SpeedPro comments


I'm running a wideband bank to bank speedpro on an optispark LT1 with a
supercharger. I've also tuned a wideband bank to bank unit on a 5.0 mustang
with nitrous. comments: the documentation couldn't be any worse. I bought my
system used, so I don't even get the support of a shop. The marketing says
that the systems come preloaded with a map to get your car up and running.
nope. the TFI computer we got (which was new) had complete random data in
the eeprom. 103lb/hr injectors, 94 ci motor, all maps were garbage, etc. The
enrichment tables are quite confusing at first. There is simply no
explanation of what a throttle follower is or what reasonable values should
be for IAC parameters, etc. There seem to be only 3 people at speedpro that
deal with this product and they don't want to support end users. Despite all
of this, I'm pretty happy with the computer. (no i'm not crazy). Here are
some good points: within 10 minutes i had WOT tuned and the car running
better (faster) than ever before. When we installed the 5.0 computer, as
soon as we got it to start we hit the highway for a 200 mile roadtrip to a
dragstrip in another city. In the first 10 miles I had it tuned by doing
pulls from 50 - 80mph. I put it in closed loop, set the correction to +/-
25% and made a VERY rough VE table. I made the VE table by knowing the
approx peak torque rpm - 3800. I made this column 95 VE (way too high,
erring on the safe side). I then made the 2000 rpm and 6000 rpm colums about
75% VE. I then used the interpolate function (quite handy) to smooth the VE
map between these three points. We then did pulls in closed loop monitoring
the ve, target a/f, actual a/f and o2 correction %... if the ve that the
computer was assuming (interpolated from all surrounding cells) was 80 and
the o2 corr. was +10% then I simply changed those cells to 88 VE. Then i'd
do another pull. After 3 iterations of this I had the entire rpm range at <
1% correction under WOT. I don't know how it could get much easier... well
actually i do... and i wrote a program that takes a log file and an engine
config file and does this % correction for you... why not? let the computer
do the math. well, when i posted about this on the turbobuick web board,
felpro's lawyers told the site admin to remove the post. they said they were
"looking into this issue". needless to say, i buried this project for fear
of being sued into a cardboard box.... I even offered to send them the code
if they wanted to put it into a future version.... no response. oh well.

The nitrous control is much less fortunate. It no longer uses the VE table
at all. You have to directly program the injector pulse widths vs. rpm. It's
only one dimensional - i.e. no load is considered, if NOS is enabled, lookup
the rpm and grab the pulse width. it also has a single A/F ratio that you
can select during nitrous use. But this doesn't work at all on our computer.
We put 11.0 in there and it show's us that the target A/F is 13.0 and the
actual is 12.5 (we did some prelim tests without actually injecting nitrous,
but just setting up the injector curve NA) one of the speedpro guy's kids
must have coded this part... i'm more impressed with that digital dna dog on
tv... so i guess my overall comments would be summarized as: the unit is
great for WOT on normally aspirated and boosted cars... nitrous is so so.
driveability is going to take you a while to do, mainly because of the lack
of documentation and examples and support. I would definitely suggest the
wideband option. but i'm not convinced that the SEFI is worth the extra
cost. I've actually been told that the way the bank to bank works makes it
more driveable with huge injectors than the SEFI....

i'd love to hear any other user's comments/applications/results....

jeff


----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org