Does anyone know the PCM algorithm that calculates the logged parameter "PFI Injector Maximum Pulse"? I know it's not the traditional calculation of 2*60*1000/rpm, but it appears to be related to it. I also know it's not the traditional calc plus some constant like injector delay.
I looked into the math behind it. That variable basically holds the max pulsewidth for the port injection, taking into account the min and max crank angle it can spray (hard coded values), and also throws in some pulsewidth offset.Originally Posted by engineermike
Eric Brooks
HP Tuners, LLC
Thaks for taking the time to look that up, Eric.
So the pulsewidth offsets, I'm assuming this is something like calculated (or specified) time to open and close the injector? Is that hard-coded or tunable? I found these times for the DI side but not the port side. Any idea why there is a no-fly-zone of crank angles for port injection?
It sounds like they don't want them going static open/100% duty cycle but I'm also losing 1/3 of the capacity of my port injectors.
What is your injector duty cycle showing? I've not seen a case where someone can't get all of their injector when dealing with the port side.... DI side there is some hard coded crank angle start and end injection windows, and we did play around with patching the code on the Explorer ST to give a bigger window.
Eric Brooks
HP Tuners, LLC
At 7500 rpm, for instance, I’m consistently logging a pfi maximum value of 12 ms. The time it takes for 2 revolutions is 16 ms so I’m missing out on a solid 25%.
Yes, I found that it basically limits PFI to ~75% duty cycle no matter what I do with "Open Valve Injection" options. Had I been able to go to 85%DC, I probably could have left stock injectors on a recent project, but we ended up stepping up to a set of slightly larger PFI injectors due to this limit.
eficalibrator, there's an OVI parameter for timing that's set to 480 deg. If you were to assume the OVI affects PFI (it shouldn't) and divide 480 by 720, you get 0.67 or 67% which is about what I'm logging as a limit. Did you happen to experiment with this number?
For what it's worth, I did log maximum PFI pulse vs actual pulsewidth and can confirm that the logic does consider this to be a limit and will not command more. Mine even modulated the throttle a bit after it hit the limit.
I have been trying to figure this out for years. At 67% the pulsewidth just stops and the car will continue to lean out. This causes us to have to run a larger injector then actually needed. I have a gen3 procharged coyote here that makes 1400 whp and it needs 2600cc IDs to keep up.
The logic behind it is to keep the min flow for DI - Ive found it a hard way when we developed 100% PI 3 gen back in the day - probably worldwide first dto do it.
It seems no matter what DI is always taken into consideration even if you theoretically run 100% PI.
v-performance.pl/
facebook.com/veee.performance/
https://www.instagram.com/v.performance.pl/
Remote tuning:
[email protected]
+48 697276579
Why do you think they are limiting it? If you can't think of a good reason for it, I'll look at patching it out.
Eric Brooks
HP Tuners, LLC
Ford very clearly spent some time working on open-valve injection, though never really used it. Is it possible they were limiting pfi to only work as cvi (thereby avoiding ovi). Or perhaps only turning on the port injector when the direct injector for that cylinder is closed to prevent any multiple-injection point weirdness that might limit the effectiveness of the di?
I made a math parameter to calculate the time during a cycle that the DI injector is not open. It was as follows: "(720-(SOI-EOI))/360/RPM*60*1000000". Then I graphed this vs the logged "maximum PFI pulse" and found that during high fuel demand they are essentially equal (minus presumably DI inj open and close delay times). At lower fuel demand, they still correlate but the logged max pulsewidth was lower than the math parameter. It appears to be calculated at least 2 different ways and a low-select applied. I have two theories based on this:
1. The calculation for pulsewidth is done on a per cycle basis. So it opens the DI injector and when it hits its limit (warm blend, EOI clip, max inj angle, etc), it then closes the DI injector and opens the PI injector for the remainder of that cycle or until fuel demand is met. By definition, that would preclude the PI injector from being allowed to use the time when the DI injector is open since they would be opened sequentially rather than simultaneously.
2. It's possible that the relay in the PCM used to trigger the injector drivers is an either/or relay and can not physically turn both on at once. When the DI hits it limit, it just switches to the PI driver for the remainder of the demanded fuel flow.
Either of the above fit with the known logic that the DI injector gets priority and the remainder goes to the PI injector.
Either of the above would also basically imply that you could never schedule the PI injector for 100% duty cycle unless you turned the DI injector off completely. But being that the DI injector is typically higher flow, you'd lose overall fueling capacity if you did this in most cases.
Last edited by engineermike; 1 Week Ago at 11:58 PM.
That sort of sequential injection would delay the PI fuel mass delivery by one cycle. I would not expect that to be the case.
The big reason for using both PFI and DI is to prevent the dreaded "soot" build up. I would guess theres a monitor for valve soot level (possibly a valve tempterature) that is limiting PFI unintentionally at high load or high RPMs .Maybe a limit at certain IVO angles.
Last edited by murfie; 1 Week Ago at 11:53 PM.
Reply With Quote