simple analog fuel injection comments from CSH, HQ

[Bruce Plecan]

Thank you Thank you Thank You
This gives a little insight into what folks have been talking about.
Now with some lurking on the PIC list might be able to get an idea
    Yes, I'm sure there is alot of PIC reading material out there, I'm
also sure that some around here has done alot of reading about 
this device.  So would that someone have any real recommendation
of which book lines up most with the needs of EFI.
Cheers
Bruce
 <goflo at pacbell.net>
Subject: Re: simple analog fuel injection comments from CSH, HQ


>>  I wish someone would start a thread on what the hell to do with
>>  a PIC. 
>
>>> For an intro to the subject you might take a look
>>> at the BASIC STAMP 1, made by Parallax. 
>>> A PIC-based microcomputer programmable in a flavor of BASIC. I'll bet 
>>> a sparkling beverage-of-choice that a Stamp-based EFI that'll run is 
>>> do-able.
> 
>> Again, would be nice to get a distilled version.  Also if you have some
>> insight to how to execute this it would be great to hear.
>
>I like the distilled stuff too, but try not to...
>For distilled Stamp stuff there is a ton of info on Parallax' site, and
>the Stamp list. Strong suite of device is simplicity and ease of use,
>though.
>A little hands-on will obviate a lot of palaver. Recommend Version D,
>which can be snapped onto a 9 volt battery, and has a little prototype
>area.
>
>Per insight into Stamp EFI: Let's assume one cylinder for openers. For a
>speed-density lash-up we need to know rpm so let's drill a hole in the 
>flywheel and press in a small cylindrical magnet. We'll mount a pick-up
>coil on the crankcase adjacent, clean up the coil output with a schmitt 
>trigger, and feed it to a stamp i/o pin. When the stamp needs to know
>rpm 
>it executes the PULSIN instruction, which waits for the rising edge and
>then counts off 10 usec increments till the falling edge. PULSIN value
>is
>inversely proportional to rpm. MAP could be acquired in a similar
>fashion,
>by using the output of the MAP sensor, with suitable filtering, to drive
>a voltage-controlled oscillator and then measuring the pulse width with
>the
>PULSIN instruction. Temperature can be read with a thermistor RC network
>and The POT instruction.
>At this point we know what we need to know to calculate injector-on time
>with the speed-density algorithm. Foo factor is arranging terms so as
>not
>to exceed the 16-bit math of the stamp. Forgive me if I don't attempt it
>this moment - Way past bed-time.
>
>I don't suggest for a moment that the stamp is a suitable platform for
>such 
>an application. Simply that the minimal programming overhead of the
>device 
>allows one to concentrate on solving the problem at hand. 
>
>Jack
>
>BTW, did I mention all the Parallax stock I own? I wish :)
>
>