> long time no talk. this is for vw bug no advance, no injector, no retard
> no thing
> just basic fire. in stead of point uses hall effect. not instead of
> direct
> firing like i am showing in attachment coil "on time" needs to be changed
> like PWM duty cycle change but the frequency of pwm should be same as
> input and input start should be synchronized
> like the picture shows. I think using external timer as clock source and
> software pwm should solve the problem unless I am wrong.

If it just needs the coil to fire when the hall effect changes state
then you need to anticipate when to start energizing the coil ahead of
time. I assume the hall effect sensor duty cycle stays pretty much the
same throughout the RPM range.

Let's say the coil must fire on the pos->neg transition of the hall
effect. So pos->neg turns off the output transistor(or IGBT, whatever)
and the coil fires. The transistor is turned on by either the delay
expiring or by neg->pos transition(failsafe mode). 

neg->pos. 
Turn transistor on if off.
Clear "x".
Start counting "x".

pos->neg.
Turn transistor off.
Stop counting "x".
Use "x" as seed for "y".
"y" = "x" * delay multiplier.
Start decrementing "y".

if "y" == 0, turn on transistor.

Loop forever.

You can either use processor cycles and poll all the time or load a
timer to share for x and y, or use the CCP peripheral if you anticipate
adding more tasks later and want to be semi-automatic.

This may not be the ideal way to drive a coil, you might just need a
fixed "off" time. That would make this even easier.

Cheers,

Bob
 


-- 
http://www.fastmail.fm - Accessible with your email software
                          or over the web

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist