Port Injection Tuning Question (33355)

[radz28]

Hi everyone,

I'm trying to put together a tune for adding PI. I've been reading many posts about many of you using the PE table to pull fuel from the DI (so as to keep TORQUE calculations happy, as I have an A10) to compensate for the fueling from the PI. But - I watched a training video that states to use table 33355 - Injector Flow Rate Fuel Pressure Mult 3, in the E92.

Has anyone else used this method? Is this a viable method? The way it was described suggests it will compensate for the PI-fueling and not effect TORQUE models and such. It sounds like you pretty much subtract about the percentage of fueling the PI is brining in, from the DI (in this table), and you're good to go.

Thanks for your time, and have a great day.
Chuck

[ZincGT]

adjust your maf curve and VVE, keep PE the same. Need to keep the injectors around 5.5ms and the tq reading will be decent. You can add tq to the tq model if need, don't go crazy though.

[radz28]

Thanks Zinc'. The VVE, too?

[LSxpwrdZ]

Or you can cut a percentage of the fuel injector flow rate to scale the fueling back but keep the airflow model the same which will keep TQ calculations accurate. Then makeup for that fuel with port injection.

[radz28]

Or you can cut a percentage of the fuel injector flow rate to scale the fueling back but keep the airflow model the same which will keep TQ calculations accurate. Then makeup for that fuel with port injection.

Thanks for commenting. I finally figured this out, sort of along the lines of your suggestion. It was a bit more involved, but believe I have it figured out. My SCANNER seems to show I have anyways.

Thanks again for sharing. It's been difficult to find any information about this way of doing it. And I wanted to do it this way for exactly the reasons you stated.

[TriPinTaZ]

Changing the injector flow rate can skew closed loop operation.

From my experience the best path is to scale back the MAF to get the fueling in the ballpark. Then scale back the PE to get it in line but don't go lower than 1.151 in the PE table or you will reference a lower Torque Coefficient at WOT. Then scale the MAF back more if you need to and adjust the Torque Model so that it is reporting at least stock values at WOT. The TCM uses this torque value to command the various values to the transmission for the shifts.

[radz28]

Changing the injector flow rate can skew closed loop operation.

From my experience the best path is to scale back the MAF to get the fueling in the ballpark. Then scale back the PE to get it in line but don't go lower than 1.151 in the PE table or you will reference a lower Torque Coefficient at WOT. Then scale the MAF back more if you need to and adjust the Torque Model so that it is reporting at least stock values at WOT. The TCM uses this torque value to command the various values to the transmission for the shifts.

It sure as SHEET did!

Agreed. I think your posts were among the few that I learned from for that method. Thank you for sharing. I didn't want to screw with the TM much though, even though I already screwed with the model for other reasons. To your credit, I learned a lot about the VTT controls. Higgs and Huggins were a BIG help, too.

I got into 17071 and started farting around and am getting good CL operation now. I was seeing trim corrections as much as 30% EITHER WAY, at times. It seems like blending and smoothing the [33355] table with [33354] pretty much corrected the fluctuation. The corrections pretty much seemed to mirror the discrepancies between these two tables, and why trims would jump around inconsistently. My scans seem to show predictable and consistent trims now, and my trims are mostly >10%, as I continue to move cells in 17071 the right direction. And when the PI comes in, trims still behave.

[smokeshow]

Which injection system is this using?

[radz28]

I'm using the Motiv Motorsport Reflex PLUS.

https://youtu.be/KpnKYcYz7kI

[GHuggins]

It sure as SHEET did!

Agreed. I think your posts were among the few that I learned from for that method. Thank you for sharing. I didn't want to screw with the TM much though, even though I already screwed with the model for other reasons. To your credit, I learned a lot about the VTT controls. Higgs and Huggins were a BIG help, too.

I got into 17071 and started farting around and am getting good CL operation now. I was seeing trim corrections as much as 30% EITHER WAY, at times. It seems like blending and smoothing the [33355] table with [33354] pretty much corrected the fluctuation. The corrections pretty much seemed to mirror the discrepancies between these two tables, and why trims would jump around inconsistently. My scans seem to show predictable and consistent trims now, and my trims are mostly >10%, as I continue to move cells in 17071 the right direction. And when the PI comes in, trims still behave.

You might want to just modify the rail pressures in those tables where PE and the port injection are solely being used. Leave the multipliers below that stock. This would allow all other operations outside of the PE area to function normally. The settings in the header are directly tied to the rail pressure settings in the density table and the multipliers are directly being used against the density tables. I use them for dialing in the over-sized injectors when the density table isn't enough. I don't know what the 3 different ones are used for. I always make the 3 for rail, offset and the open loop IVT tables the same.

[radz28]

I, essentially, figured out (I believe) some of the relationship you're talking about, and for now, only seem to have to pull fuel pressure [17071] down to the lowest pressure I altered in [33354]/[33355], so if that were 13MPa on those MULTIPLIERS, then I'd hold a lot of [17071], in areas up to 13MPa (stock), to about the same pressure. I'm holding that RAIL PRESSURE into where I have PE coming in, and CL is pretty much well below 10% correction, and OL is un-effected.

Thanks for your comments G'. Those were new to me, and are helpful