UGH! need to squash SFI/batch fire argument

[Six_Shooter]

I was under the impression that SFI didn't necessarily mean the injectors fired when the valve was open, but just that they fired one after the other instead of groups.

Which is tied to when the valve is open, or rather, about to open.

[RobertISaar]

i don't see why everyone is going bonkers over injector firing phasing overlapping? it's not a problem at all and is inevitable at higher injector duty cycles.

[ericjon262]

i don't see why everyone is going bonkers over injector firing phasing overlapping? it's not a problem at all and is inevitable at higher injector duty cycles.

what he said...

[Six_Shooter]

Who's going bonkers?

[EagleMark]

to me, reverting to batch fire just sounds like more code to do the same thing. I do think the PCM's are more then capable of keeping up though, it's not moving faster then the speed of light or something. I don't know the exact numbers, but an injector firing 100 times a second doesn't seem absurd for an electro-mechanical device.

Maybe the PCM can but injectors are mechaincal. It's not like poof they are open and poof they are closed, it takes time to open and close as well as how much time to stay open. And the time at higher RPM and more cylinders is not seconds but millseconds... or is that microseconds...

I was under the impression that SFI didn't necessarily mean the injectors fired when the valve was open, but just that they fired one after the other instead of groups.

SFI just means each cylinder injector is timed to the event, so a V8 = 8 events. Batch fire is only 2 events.

Even comparing these old injectors for SFI to Direct Injection is like comparing apples to oranges...

Who's going bonkers?

Me! 20,000 RPM? I remember that video from earlier and watching the injectors fire with that cover off is awesome.

[RobertISaar]

8 cylinders SFI at 8000RPM(15mS per 2 revolutions), let's say 80% duty cycle, that's 12mS injector on-time. at 80% DC with 8 cylinders, 6 of the 8 injectors are going to be open at any one time(and for a short amount of time, 7 out of 8).

it's not a problem at all on the logic side of things... the interrupt routines for dealing with injector sequencing and timing are lean enough to deal with these kinds of situations. and i don't see why it would be a problem on the mechanical side either, since injectors are "dumb" devices... they just react to signals. they'll deliver the same amount of fuel for the same length of pulse from the injector drivers regardless of being fired in bank, batch, sequential, etc...

i personally don't care for SFI, since it's just more things to go wrong for very little to no benefit. direct injection? different story.

[1project2many]

F1 engines are a different animal, and are designed very differently than passenger car engines. They idle at a much higher RPM than a passenger car engine, and have a very different placement for the injectors as well. In an F1 engine the injectors are mounted above the intake runners, in what would be the plenum. This allows there to be an abundance on fuel in the runner and remain a suspended mixture, to aid in high RPM running.

But they're subject to even more stringent limits in terms of achieving maximum economy and power from the fuel system. Even with fuel remaining in suspension it's not as if they're pumping the plenum full of liquid fuel to be kept on tap for later. The question is still valid: Are F1 engines sequentially injected?

from this article: http://www.formula1.com/news/features/2013/4/14512.html
eight cylinder engine
1 injector per cylinder
100 bar maximum fuel pressure
"At 18,000rpm - before the limiter kicks in - the fuel injector is fired once every 6.6ms for a duration of 2.7ms at full throttle."

At 3.33 ms per revolution those numbers only work for batch fire. And that makes sense. In order to get the amount of range needed to cover 13k rpm difference from idle to WOT with sequential injection you'd need to use a higher maximum fuel pressure. To me that indicates a dynamic range issue.

Haha... so much for squashing this one! :)

[RobertISaar]

... that's kind of odd, 18,000RPM = 6.6667 mS per 4 stroke cycle. so it only fires once per 4 stroke cycle, and for a 2.7mS period? that's a 40.5% duty cycle, unless there is some miscommunication on the site, if it were "fired twice every 6.6ms for a duration of 2.7ms at full throttle" as would be done in double-fire batch(or for some reason, firing twice in SFI), then that's 81% duty cycle, which makes perfect sense.