On The Fly Tuning (OTFT) - was Re: 8192 baud 1227808 project.

[Peter Gargano <peter@techedge.com.au>]

Bruce Plecan wrote:
> 
> > Peter Gargano wrote:
> > Approach b. also has the possibility to add in a Kalmaker (sorry Bruce for
> > using that word) style tuning-on-the-fly approach. And this only requires
> > a modified memcal to work too!
>
> Would any of this be transferable to like a 730?.
> The idea of OTF tuning is always nice.  Especially if priced right.
> $600 to be able to do one and only one ecm is way over the top in my book.

This is applicable to any ECM that uses tables and can be made to
run out of RAM, or for an ECM where the tables can be placed in RAM.
GM, Ford, or whatever ECM. If this technique could be patented, whoever 
thought of it first (and patented it) would make a packet. It's a bit
late now though if it's not already patented.

The software tools are no big deal. All that's required is for some
code to be added to (say) WinBin to send updated tables to a memcal 
that has been modified to be able to be written to. If a write signal
is taken from the ECU main board, and piped up to the memcal, and some
RAM (battery backed) is used in place of the EPROM, and some boot
and new ALDL code is added to the memcal, then it all works.

I know there are a few ANDs in the last paragraph, but each of them 
are small changes to what currently exists. The biggest change would be
to make a memcal PCB that had a battery *and* SRAM (Static RAM) and also
has sockets for the calpacks (=NetRes, or resistor networks) and a
knock sensor thick film circuit (if fitted). Anyone want to volunteer 
some Protel time to do this? (Ross, you a Protel expert?)

Most of the rest of the information is also readily available from 
Ludis's site - eg. to work out the existing P4 memcal PCB. As for WinBin,
I thought there was already some plan to do the code for this (Eric?).
The changes to the memcal's code are quite minor (IMHO), and the only
reason I haven't done them already is simply lack of time - in fact,
when I get some help with creating an ECU file for ASBX then I'll be
doing the HC11 code. Anyone want to volunteer some help on getting
an ECU file set up easily (see previous plea in another posting) given 
I have all (most?) of the information on table locations already?

If anyone wants to contact me this coming week then you could try
mailto:turboalfa@excite.com as I'll be on the road.

BTW, I think one of the reasons Kalmaker (sorry, that word again) have
not made a big impression, is that it's so difficult to actually make
a product that is secure enough to sell into a market that is known
to have a lot of users who are happy to copy, rather than pay for, the
software. perhaps there are other issues I'm not aware of, but I think
this OTFT concept has been totally under appreciated. And I also think
that if you look at the price of other software packages - like Diacom,
where people are happy to pay some hundreds of dollars, then even at 
$600, Kalmaker isn't a bad proposition (and Bruce, I understand your
experiences [multiple] with Kalmaker, but I think that you under-rate
the underlying technology and, even if some of the original stuff was
"stolen" [hmm, maybe that's allegedly stolen], there is still a lot
of work that went into the Kalmaker software). Anyway, JM$02W

Peter

PS. The $600 figure for *one* ECM may not be quite correct, but I'll
stand corrected if someone finds an (on-line?) price list they can quote?
----------------------------------------------------------------------------
To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
in the body of a message (not the subject) to majordomo@lists.diy-efi.org