[Diy_efi] New guy on board - and my personal DIY project

Steven P. Donegan steve
Thu Sep 7 02:41:52 UTC 2006


Well - at present to really reprogram the board (at a code level) it
would require a PIC programmer of some type. I use a USB attached one I
bought a long time ago :-) You would also need a compiler that costs a
few hundred $ - for the moment I'll keep my prototype code to myself -
once it flies 100% I'll open source it.

To tell it to seek a goal - that will be done via either the RS232
channel or the SPI or I2C channels from my core brain for this project -
which is a 60+ MIPS ARM microcontroller (how can you call something
micro when it's 32 bits, and has vastly more CPU/RAM/etc than the
computer on the Space Shuttle?) For the basis of my first prototypes
I'll be putting together a simple RS232 protocol - commands to the board
and data back from the board - should make it usable as a data logger
even if it is not playing games with the stock ECM :-)

A note on my background - I was born in 1956. I have been doing
electronics, cars, programming and bit-level device drivers and network
code since I was 15 - ie Haight Ashbury/Vietnam days. 

I intend each of my little boards to be stand-alone (ie use your laptop
to tell it what to do or log it's data) and integrate with my ECM when I
get that done - it currently is an ARM 32 bit unit running a real time
OS called FreeRTOS - the Olimec ARM prototype board cost a whopping 70$.
That guy's job will be to take the inputs that are already
pre-conditioned and issue orders back to the peripheral boards - being a
'dinosaur' I would say I modeled this system on the original IBM 360
mainframe - a core brain with peripheral brains (channel attached
controllers in that era) to distribute the work.


On Wed, 2006-09-06 at 21:22 -0500, Joe Boucher wrote:
> This is cool.  What does it take to reprogram the board?
> 
> -----Original Message-----
> From: diy_efi-bounces at diy-efi.org [mailto:diy_efi-bounces at diy-efi.org]On
> Behalf Of Steven P. Donegan
> Sent: Wednesday, September 06, 2006 5:30 PM
> To: diy_efi at diy-efi.org
> Subject: [Diy_efi] New guy on board - and my personal DIY project
> 
> 
> I just joined the list today. I am assisting a friend in building an EFI
> setup for an old V4 SAAB, am building one from scratch for my
> supercharged Mustang (1985 vintage) and am designing one from scratch to
> 'fool' the stock ECM in my 1998 LS1 Camaro to do what I want it to do.
> 
> The last project is one that may interest you. As of this AM I sent a
> PCB fab order for 20 prototypes for my first piece of the design. The
> little 3 inch by 1.5 inch board collects wideband O2 data, GM MAF
> (frequency based) data, and 4 other optional Analog inputs, has a little
> 20 MIPS microcontroller on it, and if you tell it a goal it reads the
> MAF/O2, determines how much to adjust the signal, and lies to the stock
> ECM - i.e. if you set 15.7:1 AFR as a goal (economy) the unit reads the
> MAF gms/sec frequency and the O2 sensor and adjusts it's MAF output
> frequency to the ECM to fool it into leaning the mixture appropriately.
> 
> It will shortly read the MAP sensor and do the equivalent lying to the
> ECM - as the stock ECM cross checks sensor data to detect errors.
> 
> This is just the first little board in a distributed computing ECM
> design that can be used as a stand-alone (like a Megasquirt) or to make
> a stock ECM do what you want it to do.
> 
> Very fun to be able to enjoy my motorhead, electronics, engineering and
> programming skills/hobbies all at the same time.
-- 
"Each new law makes only a single guarantee. It will create new
criminals."-- John Tandervold





More information about the Diy_efi mailing list