Why not just flatline MAF at 5.5 ms with?

[GHuggins]

So it's correcting torque and everything then? That's actually really nice to know. Sounds like you're basically going back to port injection and using DI for light throttle and cruise? Any chance of being able to dial in injection timing based on rpm and everything like that for it?

Possibility of a demonstration vid before it hits the market?

[smokeshow]

Has nothing to do with torque. It's a fuel injection controller. And there will be knobs to tweak where necessary.

[DSteck]

Yea I’m not real sure why you keep mentioning torque. It’s got nothing to do with this. The work we’re doing on torque modeling is entirely separate.

[GHuggins]

Was just wondering if it was correcting for the alternative fuel source was all. Basically it sounds like tuning of the base system is still required. Im sure all will be made clear in time either way.

[TriPinTaZ]

Sounds like the Airmass values won't be skewed like the current Port Injection kits. Which would then keep the calculated torque, from airmass, far more accurate. This would negate the need for having to adjust the torque model like you do for current port systems.

I don't care how it works. As long as I can still turn knobs and dials. If it gets wideband feedback for closed loop operation, that's just icing on the cake. Let me know when I can buy one, Dave. I want to test it on my personal 2023 ZL1.

[smokeshow]

Sounds like the Airmass values won't be skewed like the current Port Injection kits. Which would then keep the calculated torque, from airmass, far more accurate. This would negate the need for having to adjust the torque model like you do for current port systems.

I don't care how it works. As long as I can still turn knobs and dials. If it gets wideband feedback for closed loop operation, that's just icing on the cake. Let me know when I can buy one, Dave. I want to test it on my personal 2023 ZL1.

Additional knobs and dials will be available here:

Capture.jpg

[GHuggins]

Sounds like the Airmass values won't be skewed like the current Port Injection kits. Which would then keep the calculated torque, from airmass, far more accurate. This would negate the need for having to adjust the torque model like you do for current port systems.

This is what I was asking about. Figured that's what the whole bi-directional comm patch was for? Kinda figured with it going into the ECM that it was looking at and possibly even using the ECM's MAF? Would be fantastic if it did then did a correction on it's own from that and would even show you what it's adding so you could then correct the air models from it. Even leaving the stock air tables in place and shutting the DI system off or dialing it back would be a big plus. You would still have to adjust the torque model to a degree, just nowhere near as much.

[TriPinTaZ]

Additional knobs and dials will be available here:

Capture.jpg

lol nice

This is what I was asking about. Figured that's what the whole bi-directional comm patch was for? Kinda figured with it going into the ECM that it was looking at and possibly even using the ECM's MAF? Would be fantastic if it did then did a correction on it's own from that and would even show you what it's adding so you could then correct the air models from it. Even leaving the stock air tables in place and shutting the DI system off or dialing it back would be a big plus. You would still have to adjust the torque model to a degree, just nowhere near as much.

Guess we will find out eventually. I understand the secrecy behind it. Pretty cool they got HPTuners to participate in the effort.

[DSteck]

You will not have to touch the torque model. The torque model has zero to do with any of it.

There will be zero need to fiddle with crap. You will tune the stock ECM as if nothing is wrong.

[STAGEUP]

Is there a subscribe button on this thread? lol

Dave, we appreciate the great lengths that you have taken to provide the community with the innovation and development of your products. Great work sir!

[SiriusC1024]

This is what I was asking about. Figured that's what the whole bi-directional comm patch was for? Kinda figured with it going into the ECM that it was looking at and possibly even using the ECM's MAF? Would be fantastic if it did then did a correction on it's own from that and would even show you what it's adding so you could then correct the air models from it. Even leaving the stock air tables in place and shutting the DI system off or dialing it back would be a big plus. You would still have to adjust the torque model to a degree, just nowhere near as much.

Idk why some feel the need to be vague to the point of condescending about it instead of just explaining. Since it's a bi-directional controller the fueling will ultimately be commanded and tracked by the ECU. Tune the airflow and torque models like you normally would. Blending between DI and PI is handled for you.

[DSteck]

Idk why some feel the need to be vague to the point of condescending about it instead of just explaining. Since it's a bi-directional controller the fueling will ultimately be commanded and tracked by the ECU. Tune the airflow and torque models like you normally would. Blending between DI and PI is handled for you.

It’s not bi-directional. We keep it vague so the process isn’t copied, and I keep watching assumptions made (not just here) about how it works. It’d be best to just wait and see when we release it because formulating an idea of how it works/should work/you think it should work can severely influence whether or not it’s treated correctly once released. There is zero need to adjust the torque model from the fueling side of things. Port injection ultimately does change a torque profile, but it’s not going to be this “cheat the system” crap everything else is.

At the end of the day, you will have very limited options in the port controller: injector characterization, toggle wideband correction, and then some boost actuator control (SmoothBoost) which is a low priority to deal with at this time.

[SiriusC1024]

I saw no reason for it to be bi-directional other than wideband informing blend proportion, but I see the feedback loop is confined to the add-on board.

[SiriusC1024]

That also explains why there won't be much to adjust tune-side. If PI contribution was reported back to the ECU then you could have a PI spark adder table similar to ethanol content. Technically, spark will have to be adjusted for the difference in combustion behavior between DI and PI. This tells me that the PI system isn't full-time, or at least makes a very small contribution during low load. Probably only comes online during PE, removing a lot of headache with low PW injector behavior and narrowband CL control.

The controller is probably fed Eq Commanded and DynAir-DynAir_DI. DynAir_DI is capped by a flow threshold.

I'll wait and see. Just having fun with it right now.

[DSteck]

When it comes out of the wash, widebands will need to be logged via CAN using a Prolink. We didn't want to have to adhere to polling requirements on the main bus or identifying as a module; too much headache when it's easier to just broadcast and let it be. No, it is not full time, but will have an automated valve wash that is state driven and not configurable for what can trigger it aside from a master enable/diasble. It is based off of port injector minimum achievable fuel mass relative to other things. Right now, we're taking around 50 parameters from the ECM while the only hard wired inputs are the cam and crank signals which will come from the replicated outputs of the ECM (so if something happens, we don't introduce noise or error into the ECM's primary inputs). The E92 does not calculate its own DI supported airflow. E99s do, though, only after the ramp factor is in play.

For what it's worth, there is no spark modification in a C7 ZR1 based on port injection. GM didn't seem to think it mattered.

[SiriusC1024]

Ah ok I get the concept. Thanks for explaining. Sounds like a really good implementation. Valve wash is a plus.

[SiriusC1024]

Just one thing. If you don't want to answer then all good. Replicated inputs off cam and crank instead of optocoupled?

[DSteck]

Replicated. Cheaper implementation with no chance of interrupting ECM operation.