Auto learn code

[Pat Ford]

 sorta what I'm looking at. I plan on 4 Ktype thermocouples for egt
and 2 adxl105 accellerometers. 

algorythm something like


while(1)
{
 if wot
 { 
   while (accellerating)
   {
     fuel ++;
     sample_egt();
     egt[rpm]=(last sample); // save temp at max mixture before power falls
   }
 }
 else
 {    
   // cruise mode
   fuel_to_set_egt( egt[rpm] - 50 ) // set fuel to egt -50 degrees
   sample_egt() 
 }
}

then if egt[] is in the  sram area the array would be stored.

 I'm playing with the adxl accellerometers right now, I think using 2,
2 axis one measuring fore and aft, the other forward and backwards but
at a 45 degree angle to the horizon I can factor out vehicle tilt.

 I've been told that the off axis rejection on the adxl105 is fairly good,
I hope thats the case.

Well what do you think??

Previously, you (Espen Hilde) wrote:
{ 
{ Hi Bruce !
{ I feel the same about this question, we have to duplicate in the digital
{ world the bennefits of a weber carburation, measuring and injecting fuel 
{ in each cylinder in real time!
{ If we could  measure each cylinders air consum and direct inject 
{ fuel later maybe the software and code could be simpler . Not lousy 
{ sensors /voltage resolution and advanced code to cover it up.
{ I am thinking of a combination of maf and sound /turbulence to get 
{ a wide dynamic range of the sensor system.
{ just some fantasies....
{ Espen Hilde
{   

--
Pat Ford                           email: pford at qnx.com
QNX Software Systems, Ltd.           WWW: http://www.qnx.com
(613) 591-0931      (voice)         mail: 175 Terence Matthews          
(613) 591-3579      (fax)                 Kanata, Ontario, Canada K2M 1W8