[Diy_efi] Re: Microcontrollers For use on DIY_EFI Systems

The Punisher punisher454 at hotmail.com
Wed Jul 17 17:04:07 GMT 2002


You are using assembly? IF you DO use assembly you are alot more of a 
programmer than me, BUT I have had FEW problems doing lookup tables, linear 
interpolation, and scaling using compilers with PIC's (and even better on 
AVR's).
How 'fast' are you trying to run? On pic's I was using 20Mhz, and with AVR 
Mega series I have been using 16Mhz.
  I output lookup table'd, average filtered and re-scaled values to LCD's 
using custom character style bargraphs, and 2 different numerical readings, 
along with 3x4 keypad scanning and ADC input all at once.
I dont have any illusions about My setups being efficient or pretty, but it 
usualy seems to work out okay. And with PIC's and AVR's(and I'm sure 
motorolla's too) I can find code examples of similar things to whatever I am 
doing on people's websites.


From: Kenny W <watson at iamfun.com>
>Reply-To: diy_efi at diy-efi.org
>To: diy_efi at diy-efi.org
>Subject: Re: [Diy_efi] Re: Microcontrollers For use on DIY_EFI Systems
>Date: Wed, 17 Jul 2002 08:45:41 -0700 (PDT)
>
>Quoting "Tomas Sokorai Sch." <tsokorai at xperts.cl>:
>
> > Hmmm, have you ever tried to multiply or divide
> > 2 integers?
> > If you use Microchip's math library, have you
> > seen how many cycles does a
> > divide take?
>
>I'm pretty sure idivide on an Hc11 takes 32 cycles,
>it's ugly. If you use your own routines you can do it
>in less SOMETIMES, but you're a little bit at the mercy
>what 2 numbers you are working with which is not fun if
>there's any timing involved since you can't predict how
>long it will take.
>
> > Calculations are out of the question for this
> > kind of uC.
>
>Seems like tons and tons of hard coded lookup tables
>are about the only option if you go the route of such
>basic mcu's,but with somehting like the example if
>converting an AMM system to MAP+IAT like discussed
>earlier, really you need a lookup table AND a
>multiplier to generate a voltage function equivalnt to
>the AMM output, and that's ugly since now things aren't
>even integers any more. Either that or an almost
>equally ugly (and extremely memory intensive) 3d lookup
>table that references each possible combination of MAP
>and IAT to a resultant "AMM" voltage. Now that i think
>of it that's probably easier than having that poor
>little PIC try to do math. It just won't happen unless
>it's all in multiples of 2. ;P
>
>Actually, that's a fairly reasonable approach. Connect
>the map and IAT, but run the car w/ the AMM. Record a
>suitable number of data points and plug in. That should
>actually be reasonably doable! Again though you either
>need a ton of points or some some of interpolating
>algothrithm which again is simply out of a 16-18 series
>PIC's league.
>
>-Kenny
>AKA Captain Bondo
>
>_______________________________________________
>Diy_efi mailing list
>Diy_efi at diy-efi.org
>http://www.diy-efi.org/mailman/listinfo/diy_efi




_________________________________________________________________
Join the world’s largest e-mail service with MSN Hotmail. 
http://www.hotmail.com


_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi



More information about the Diy_efi mailing list