Another thought
Turbo Mini Forums

So, I just had another concern for the charge stealing issue and EFI.
This problem that I believe all siamese port engines suffer from, even carbs. Inside the port, there is near stagnant air until the first of the two intake valves open. At this point, air is accellerated which can leave some fuel(because it's heavier) behind. Eventually the fuel will catch up, or at least mostly so, and I'm not really so concerned so much about that. However, when the next cyl opens, the air already has some momentum and ramcharges this cyl the entire time the valve is open.
In this scenario, I see the first cyl as running slightly rich, and the second running slightly lean. Scatter cams, I understand help with this situation by evening out the air flow(not just fuel) into the cyls. This seems like a good solution, but I'm wondering if anyone has any ideas on how to address this in software, or are we content that a carb would have the same problem, so we're not terribly worried?

-Tyler
DO NOT TOUCH MY HORNS OF DOOM!

This is basically different volumetric efficiency for the inside cylinders compared to the outside ones. The most obvious way to correct for this would be to run 2 different VE tables for the different cylinders. This way you provide the correct amount of fuel for the different amount of air getting in the cylinders.

I'm not sure how big the difference in VE is between cylinders and if it's worth the trouble of having to tune 2 different tables. There could also be a single value that could adjust the pulsewidth by a certain percentage or a set of adjustment values dependent of RPM.

Certain ECUs have fuel trim capability that allow each injector to be adjusted for each cylinder actual VE and some also allow AFR monitor and feedback for each cylinder.

If you're not going to spend a lot of time on a dyno to optimize all fueling and ignition parameters to get the last HP and/or MPG, I'm not sure if any of it would make much of a difference or even much sense.

If I remember correctly, I think that the router board being designed for Megasquirt will have fuel trim capability and multiple O2 sensor feedback. Of course, if you only want to differentiate between inside and outside cylinders that's overkill and the current V3 board and MS-II already have provision for 2 O2 sensor inputs and 2 VE tables. The software would have to be modified to use those in that way though.

Jean

http://www.jbperf.com/

Ah, I didn't realize there were 2 VE tables in the V3 board. If the tables already exist, tho, why does the software need to be modified to use it? Is it used for something else, currently?

-Tyler
DO NOT TOUCH MY HORNS OF DOOM!

The second VE table is used for the dual table option (duh! :)). With this option, each of the 2 injection driver has it's own VE table. So it could be used for staging with 2 injectors per port or for controlling water injection or controlling 2 banks on a V engine.

It could be used for what you suggest if you use 2 injectors per port but there would still be a need for code modifications because MS-II does not do sequential injection which is required to time the correct injector with the associated cylinder. So if you go to the trouble of doing this, it's not that much more difficult to modify the code to be able to use the 2 tables with a single injector per port.

Once I have a solid working siamese-port version of the MS-II code, I'll have a look at this. Maybe you could add this to the wish list thread in the MS section so I don't forget.

http://www.jbperf.com/

I just read something that would also call for having different VE tables for the inside and outside cylinders. Have a look at:
http://www.msefi.com/viewtopic.php?p=100121

I'm not concerned about the acceleration and deceleration aspects but more about the steady state where there is always some of the fuel on the port walls being reintroduced in the intake flow. Since there will be more of this fuel going into the inside cylinders (similar to charge stealing), this would compound the problem with the possible better VE of the outside cylinders.

That also means that the positioning of the injectors as close to the head as possible gets an additional theoretical validation for reducing wall wetting.

Jean

http://www.jbperf.com/