[Diy_efi] Timing Calculations
nospam at higginstribe.com
Wed Jun 15 22:44:40 CDT 2005
Of course for batch fire it doesn't mean much since there's no
individual injector timing. All you care about is duty cycle.
For sequential I actually have to specify an on and off time. If, due
to acceleration, the time between the calculated on and off event is
shortened, it means my injection time is shortened. Since the off time
should be static, it means I have to calculate the on time well in
advance of the off time, possibly a full engine cycle in advance. Since
ignition is an instantaneous (relatively) event I have the option of
reducing the error by basing the timing from the closest trigger point,
e.g. ignition timing is 3 degrees from the 15 tooth.
I was just thinking I might not calculate the injector off point and
just let it deviate a little in order to sacrifice wandering into the
intake/exhaust overlap period for the sake of precise fuel control.
I'll simply calculate the on event time as best I can and specify the
number of uSeconds for the injector to be open. If the off point
wanders some, so be it.
Anyway, this is just free thought. I need to spend some time, maybe
tomorrow, reworking the whole timing scheme.
For anyone who cares, my highest resolution trigger wheel is 24 tooth.
> Ignition timing is much more important than injector timing - if not,
> how do batch fire engines even run?swap the injector wires and nothing
> seems to change - swap the ignition wires and see what happens. TomS
> ----- Original Message ----- From: "Mark Higgins"
> <nospam at higginstribe.com>
> To: <diy_efi at diy-efi.org>
> Sent: Wednesday, June 15, 2005 9:57 PM
> Subject: Re: [Diy_efi] Timing Calculations
>> Before people get off on more of a tangent with 60 tooth trigger
>> wheels. Thanks for all the discussion and calculations. I also want
>> to factor that all of your calculations are based on the worst case
>> scenario, that timing will come at the farthest point from the
>> trigger wheel. Obviously the deviation from the actual and desired
>> event time will start at 0 from the trigger point and reach the
>> greatest error just before the next trigger point.
>> Anyway, it seems pretty obvious to me then that I should build in
>> some sort of rpm prediction calculations, possibly based on throttle
>> (or MAP) changes if I want them to be quick. Of course, I'm more
>> worried about injector timing then ignition timing. I can't move the
>> injector off time (or at least I shouldn't, much) which means I need
>> to fairly accurately predict the on time based on what I currently
>> know about the acceleration/deceleration, and I need to predict that
>> on time, possible most of a engine cycle ahead of the injector off time.
>> This means I'm going to have to sit down and seriously think about
>> the logic I'm going to use to reduce this problem as much as possible.
>> Thanks everyone.
>> Mark Higgins
>> Diy_efi mailing list
>> Diy_efi at diy-efi.org
> Diy_efi mailing list
> Diy_efi at diy-efi.org
More information about the Diy_efi