On my G8 LSA I cant get any PE in dynamic air its says it is but my wideband show lean compared to the commanded. In MAF only I it fine and when tuning my VVE its enriching. Any help is appreciated. Here is my tune and my log files
On my G8 LSA I cant get any PE in dynamic air its says it is but my wideband show lean compared to the commanded. In MAF only I it fine and when tuning my VVE its enriching. Any help is appreciated. Here is my tune and my log files
Try disabling De-soot mode in the fuel tab.
2016 Silverado CCSB 5.3/6L80e, not as slow but still heavy.
If you don't post your tune and logs when you have questions you aren't helping yourself.
Did that and no help at all. Its just no a tip in lean its just lean all the the time when its commanding PE in dynamic air.
I had a similar problem with my PD blower and only "solved" (or avoided?) it by lowering the Dynamic Airflow High RPM Disable/Re-enable to 1400/1300. That gave me PE when I wanted it without any undesirable side effects that I've noticed.
If you did that your just on the MAF then. Mine is fine when I just run on the MAF only.
Sounds a lot like an issue I worked on a few years back. It's a pretty gnarly algorithm, so I'll explain just the part of it that caused this problem for me.
So you've heard of transient fueling, right....even more mysterious is the transient airflow. When in dynamic airflow, it is monitoring for transient conditions, be it MAP, TPS, and most recently, cam phasing. When in this transient state, both the MAF and GMVE are expected to be somewhat inaccurate. When you're tipping into the throttle rapidly, there is a difference in pressure between the MAP sensor reading and the actual pressure in the cylinder - and the MAF might as well be a mile away. So neither one ends up being a good representation of cylinder air. GM solved this with a feed-forward prediction algorithm that uses a Kalman filter to resolve the vector of the present transient conditions into a prediction of how much air will actually reach the cylinder on the next event. Essentially, your current air flow is weighed against a set of gains (multipliers) based on the magnitude of the transient and it comes up with what should be airflow for the engine in the immediate future. For the MAP contribution to this prediction, it multiplies those gains against a maximum MAP value. For an NA vehicle, that's just baro, barring any pressure drops in the intake piping. And supercharged vehicles use an expected max boost value, which is a calibration based on the peak boost the blower can make at that RPM. Problem is, once you boost that NA vehicle, it's still expecting only as high as barometric pressure in the manifold during a transient. Even though your MAF may be dialed in perfectly and you've spent dozens of hours tweaking the GMVE, the prediction skews the airflow lower than it really is when in boost and makes it run lean. Ideally you'd calibrate the max expected boost and get rid of that problem. But I've never seen it in the HPT editor.
A quick fix, though imperfect, is to just zero out the prediction coefficients that are causing the issue. Since we can't edit the boost MAP prediction in HPT, it will always predict at or lower than baro...so you can zero those gains out. It isn't a complete fix, since those predictors can be very useful for sharp throttle response and delivering fuel to the cylinder immediately as it needs it rather than having a few cylinders fire lean before the air model stabilizes enough. But it should be much better than the constant leanness some people experience while the airflow is in a transient state. I won't touch on the correction factor or prediction stability for now, so only change what you need to in order to fix this issue.
And I doubt anyone would want more info, but just in case, here's a flow chart of the algorithm as well as a link to the GM patent for the method:
https://patents.google.com/patent/US20050060084A1/en
Last edited by smokeshow; 11-10-2020 at 01:23 PM.
I certainly appreciated the more info aspect of it - thank you!!
That may explain why I have played with the prediction coeffs with seemingly no effect. N/A gen4 with blower added. Even putting a high bias on tps seemed to do nothing. They are currently zeroed out, again.
There is a max map input in the cal, which I have adjusted, but I suspect it is just for map diagnosis, even though it never set dtcs before it was changed from stock.
PE TAB.JPG
I think you PE enable is set as 15KPA, mine is 85KPA (LSA)
On the PE Tab look at Throttle 'HOT'
Your low rpm setting is saying to enable it above 43% TPS
My suggestion is to lower this to 10% and have your PE activate off manifold pressure. I see 75-85KPA here a lot.
Don't you want pe to enable at lower MAP? Why set it that high?
i always use throttle as the main trigger and only add in map as well if i need to for any cruise mid load areas if it still gets triggered where i dont want, i see boost at about 40-50% throttle but i also use BE in the 2 bar OS, PE also help add torque so if u want a stronger mid load response u can use it just dont need as much enrichment there if ur not full load
PE isnt just for boost, its for n/a load also so should be used lower then 100kpa
factory has delays and uses TPS as the trigger the 15kpa is so its only triggered by the TPS, so by the time it gets to enable TPS with the delay its prob into the 120kpa by that time as boost can rise quickly, u always hear about n/a using it earlier as its needed, also if it wasnt needed until after 100kpa then everyone could just dial in there maf/vve at stoich and not have to complain about the shift in commanded but that is not ever recommended as it needs enrichment
You're comparing n/a tuning to pd blower tuning, and a custom OS with BE to a factory OS.
Using MAP to enable PE with TPS set low is basically making the factory OS use BE without actually having what a custom 2/3 bar OS does. I've done it this way on a couple modded LSA cars and it works well. Without having BE, you'd have to spend a LOT of time/fuel datalogging to figure out where to set TPS to trigger PE to make sure it comes in before getting into boost, it's 100x easier to just set it to trigger off MAP.
Enlighten me on how you plan to do that in a pd blower vehicle, without boost enrichment, and without burning a tank or more of fuel and a bunch of hours datalogging?
When you're done with that, explain how doing it that way makes any difference in how the vehicle drives compared to setting MAP enable to 95kPa (where I usually set stuff that gets part throttle boost), and what the customer will notice that makes doing it that way worth all the time you spent on it when you could have already been working on something else.
I'm also curious what your MAP/RPM/TPS cutoff is for where n/a stuff "needs enrichment" when not at WOT, as well as what boosted vehicle you've dealt with that you felt needed enrichment under 95kPa, and/or was below 95kPa when you hit PE enable TPS.
You'll find after you burn a tank of fuel and a few hours logging that MAP alone is no good for PD blower cars. I have found many times by my own testing of these PE settings that by the time the MAP setpoint (factory was 75kPa) is met and the PE ramps in and the fuel catches up that I am well past 120kPa.
It takes a combination of TPS and MAP to keep fuel economy and make it always in PE above say 85 - 90kPa.
Here are my settings for example. E67 470rwkw LSA
1.PNG
Last edited by hjtrbo; 02-07-2021 at 05:11 PM.
Here is a screenshot of my old factory PE settings to demonstrate what I mentioned above. You can see I have blown well into boost by the time PE lands at 0.8 Lambda. This was with MAP set at 75kPa and PE TPS at 15. Cylinder air mass spikes up early with the shot of PD boost when the throttle blade snaps open.
1.PNG
I'm not a tuner but I am of the opinion it is important to snap the throttle open at low rpms locked in gear to see how PE and transient fuelling is going on PD blower cars