Stamp EFI

[Matthew Harding]

At 05:01  25/06/98 +0000, you wrote:
>
>> >I have read a thread today that looks like ths start of an 
>> >possible EFI project using a basic stamp.
>
>> can a few people have a look at this and tell me if it's worth trying?
>
>Anything is worth trying if you get some enjoyment out of doing it.
>
>The biggest problem with the Stamp is that it can only do one thing 
>at a time (and fairly slowly). An engine isn't that forgiving. You've 
>off-loaded some of the work from the Stamp on the input side but I 
>would suggest you try and remove the 'real time' aspect from the 
>outputs.
>
>If you take a mixture of your drawing and Ray's analog suggestions, 
>you are probably a long way there. Using an ignition event as a 
>trigger and a ramped input to a comparator to determine the on time 
>of the injector means that it can look after itself without taking 
>any resources from the Stamp. You can then use the Stamp to do the 
>thinking and apply mixture adjustments to the analog part with a D/A.

I was thinking along the lines of, if you let the Stamp turn the injector
on with a PULSOUT command, then PAUSE for say 1.5ms you are ensuring that
the injectors never go over 80% duty cycle....

This would also mean that the EFI would run simultanious on all ignition
pulses up to around 3500 rpm, then would fire the injectors on every second
pulse... because it would miss the pulse whilst pausing.

other than that there would only be half a dozen commands to look up the
MAP and RPM, then look up these values in a 2D array.  Any other additional
operations, such as cold start enrichment and accelerator pump are only
going to be 1 line of code each......

Does anyone know what a standard 4 cylinder injection system actually does,
do they fire on every ignition pulse, or every second or what?
Matthew Harding
---------------

reply to any of these....

mharding at qonline.com.au
bernie at ronscomputers.com.au
bernie at browse-cafe.com.au