[Diy_efi] Timing Calculations

Amd CPU amdcpu
Fri Jun 17 05:40:42 UTC 2005 

In system without tooth-wheel, where is only 1 trigger per ignition event,
calculation of dwell is an issue (not big).

You must have very good signal to make desicions from last tooth period. In
my system cycle-period error is upto 10%/180deg. Luckyly or not i have to
wait only upto 20deg after trigger to ignite, so my worst case deviation is
2deg (this happens about once in a day :D ).

Injection angle can't be as important as duration. I don't believe you are
closing injector because engine has right angle, although you haven't
injected enough.

I haven't played with seq.inj. but i've heard that the best angle to start
injection is after closing of intake valve, opposite to popular opinion - to
warm up the mixture. With that theory, the closing point doesn't matter. Is
that so?
Also have heard that could win when not injecting during overlap.
Which one is correct? How big is the difference, can we see +5% of
efficiency?

In theory you can use 1 timer and 4x16bit of memory to time 4injectors. You
must not hold them open with timer pin. Haven't seen it, but should be
possible.




----- Original Message ----- 
From: "Mark Higgins" <nospam at higginstribe.com>
To: <diy_efi at diy-efi.org>
Sent: Friday, June 17, 2005 7:41 AM
Subject: Re: [Diy_efi] Timing Calculations


> That doesn't close or redefine anything.
>
> Dwell is important, but dwell is only a small amount of time and wont
> overlap other ignition events.  I can sacrifice one timer for all
> ignition events and not need to worry about overlap.  Even if I plot
> 'dwell on' and 'dwell off' angles, deviation will be minimal because of
> the short dwell time.
>
> With ignition I can always use the last trigger tooth delay to calculate
> rpm, so I can calculate rpm as accurately as the last trigger before the
> ignition 'zone' for that cylinder.  An injection event can last up to a
> full engine cycle, so my rpm reading will deviate much more in that time.
>
> As I have been saying, the angle that injection ends is the important
> part, not the starting angle.  This is why most aftermarket EFI let you
> specify end of injection angle, which is the point when injection should
> be finished.  I have more 16bit timers than usual on my uC, which is 4.
> I can't sacrifice a timer to each injector.  Even if I did, incorrectly
> guessing the start angle that will result in the proper pulse width
> before the ending angle will cause overshoots.  Then I might as well
> just go to batch injection.
>
> Mark Higgins
>
> Astrona wrote:
>
> >Injection is time-based, but shooting-angle is also important.
> >Ignition is angle-based, but dwell-period is also important.
> >
> >If you have big enough injectors to avoid 100% duty cycle in every case,
> >you can just open the injector at desired angle and set the timer (no
> >degrees!) to shut it.
> >
> >Should close the issue or redefine it.
> >
> >
> >
> >----- Original Message ----- 
> >From: "Mark Higgins" <nospam at higginstribe.com>
> >To: <diy_efi at diy-efi.org>
> >Sent: Thursday, June 16, 2005 10:26 PM
> >Subject: Re: [Diy_efi] Timing Calculations
> >
> >
> >
> >
> >>Has anyone actually made their own EFI on this list, cuz this is kinda
> >>annoying.
> >>
> >>You are basing your idea of injector timing in sequential mode, that it
> >>is a single event, when it is more accurately defined as two events.
> >>One event opens the injector and one event closes the injector.  Both
> >>the open and close points of the injector need to be calculated, unlike
> >>in batch mode where you can just modulate the PWM and injection is not
> >>an event at all,
> >>
> >>The ignition event is simple, since it's an instantaneous event that
> >>only cares about it's own position.  The injector ON event doesn't care
> >>about it's position, it only cares about trying to accurately determine
> >>when to turn on so that the injector will be open for a set amount of
> >>time before it reaches the injector off position.  The injector off
> >>position cares about two things, it's position and whether the injector
> >>has been on for the specified amount of time.
> >>
> >>Because of the acceleration I have to deal with the on and off event
> >>coming closer together time wise, which will ultimately effect the
> >>amount of fuel.  So, the acceleration/deceleration problem will
> >>ultimately effect the amount of fuel , starving it on high acceleration,
> >>which, yes, I'm more worried about that than 1 or 2 degrees of ignition
> >>timing.  But, if I fix one, it will work towards fixing the other.  Yes,
> >>I know about the importance of ignition timing, can we get over that
now?
> >>
> >>Mark Higgins
> >>
> >>Adam Wade wrote:
> >>
> >>
> >>
> >>>--- Mark Higgins <nospam at higginstribe.com> wrote:
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>>Of course for batch fire it doesn't mean much since
> >>>>there's no individual injector timing.
> >>>>
> >>>>
> >>>>
> >>>>
> >>>That was the whole point.  Batch fire vehicles run
> >>>just fine, and make excellent power; in fact, I doubt
> >>>you could tell me whether a vehicle was batch or
> >>>sequential fire, even looking at dyno data with
> >>>four-gas analysis.
> >>>
> >>>Injector timing makes very little difference once you
> >>>get very far above idle speed.
> >>>
> >>>
> >>>| 82 Honda CX500 Turbo (Cassandra)  90 Kwak Zephyr 550 (Daphne) |
> >>>| "It was like an emergency ward after a great catastrophe; it  |
> >>>|   didn't matter what race or class the victims belonged to.   |
> >>>|  They were all given the same miracle drug, which was coffee. |
> >>>|   The catastrophe in this case, of course, was that the sun   |
> >>>|     had come up again."                    -Kurt Vonnegut     |
> >>>| M/C Fuel Inj. Hndbk. @ Amazon.com -  http://tinyurl.com/6o3ze |
> >>>
> >>>__________________________________________________
> >>>Do You Yahoo!?
> >>>Tired of spam?  Yahoo! Mail has the best spam protection around
> >>>http://mail.yahoo.com
> >>>_______________________________________________
> >>>Diy_efi mailing list
> >>>Diy_efi at diy-efi.org
> >>>http://lists.diy-efi.org/mailman/listinfo/diy_efi
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>_______________________________________________
> >>Diy_efi mailing list
> >>Diy_efi at diy-efi.org
> >>http://lists.diy-efi.org/mailman/listinfo/diy_efi
> >>
> >>
> >>-- 
> >>No virus found in this incoming message.
> >>Checked by AVG Anti-Virus.
> >>Version: 7.0.323 / Virus Database: 267.7.3/15 - Release Date: 6/14/2005
> >>
> >>
> >>
> >>
> >
> >
> >
> >
> >
> _______________________________________________
> Diy_efi mailing list
> Diy_efi at diy-efi.org
> http://lists.diy-efi.org/mailman/listinfo/diy_efi
>
>
> -- 
> No virus found in this incoming message.
> Checked by AVG Anti-Virus.
> Version: 7.0.323 / Virus Database: 267.7.3/15 - Release Date: 6/14/2005
>
>



-- 
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.323 / Virus Database: 267.7.3/15 - Release Date: 6/14/2005

More information about the Diy_efi mailing list