Transient fuel comparisons & rich tip-in

[Throttler]

I'm struggling with seeing the effects of the transient fueling settings on both WOT transition to PE and part throttle transients.

I feel like I've done a ton of benchmarking the settings of different vehicles, as well as developing analysis techniques. I have developed some histograms to try to help guide me as well as exporting to excel. I have to say that not having a rate of change function working [ dy/dt or n - (n-1) ]is a major hindrance to making the histograms be meaningful. I'm glad to hear that v2.25 will permit this ability.

Truck is a 1999 2500 Extended Cab Long Bed 4x4 Silverado w/ cast iron headed 6.0L LQ4 & 4L80E running 87 octane E10. Only physical mods are removal of EGR and swapped the injectors w/ LQ9 injectors (stock LQ4 P/N 25320288 4 holes, 24.8lb/hr; LQ9 25317628 2 holes, 25.2lb/hr; 1.6% increase in flow).

Current tune has the VE table calibrated, MAF calibrated, TM removed completely, BK thresholds increased, KR authority slightly decreased, DFCO off, COT off, spark tables tweaked in the WOT region only (high octane = low octane), PE leaned out to 13:1 and enable triggers reduced.

I am running the GM blended scheme off both MAP and MAF, both narrow band O2's disabled, running Open Loop, monitoring with a NTK AFX wideband. I did copy over the IFR vs kpa tables from the LQ9 to my tune, which has some flow rate variation with pressure, whereas the stock LQ4 injectors have a constant flow rate. I did not copy over any of the other injector parameters like Voltage Offset, Pulse Corrections, Min Injector Pulse or Default Injector Pulse. Only thing done to the trans settings is associated with autoscaling the tire size.

Typically I experience rich on tip-in and lean on tip-out for part throttle transients. WOT transition to PE seems okay, but could be a little snappier tracking of the AFR demand.

What concerns me the most is that no matter how I change the 4 'typical' transient tables (Fuel to Wall Impact Factor, Impact Factor Gain, Boiling Time, & Fuel on Wall EXP Decay Mult) I can't even get the response to even reverse itself (lean on tip in and rich on tip out, for example). I've tried some pretty radical mods especially to the Fuel to Wall Impact Factor, and I have a hard time noticing any conclusive difference with the logs.

Now I will say that I really haven't taken several hours in a given day and methodologically changed the tune and repeated the same transient. As like most, job and family interfere with this type of commitment, and I'm not about to pay for a dyno session for this. I basically will log during a normal trip from A to B, making some changes to the tune between logging sessions. Here's a couple examples:

Here is a rolling WOT transition with the cursor placed on the max AFR error. I don't think this looks too bad. Yes I know I have some KR here. This is utilizing the 2000 Silverado transient tables. Also note the rich dip at the shift point in the middle of the trace, and the very lazy return to the non-enriched 14.68:1 commanded AFR after the tip-out.



Next up is a highway cruise to WOT transition with 2001 Silverado transient tables. This one goes lean on tip in and is a little lazy getting onto point at the PE ratio of 13:1. Also note the crazy AFR swings on tip-out.



I'll try to add some off-idle transients from a standing start soon.


Here are some comparison's I've done of the various ~Gen III engine transient calibrations. I'm considering the 1999 Silverado as the baseline table that I'm comparing the other tables to.

Generally I'm comparing the "hot" portion of the tables where my truck runs when fully warmed up, which is typically 194*F, which is pretty much right between the 176*F and 198*F columns.

Fuel to Wall Impact Factor vs Coolant Temp vs MAP

Here is a graphical depiction of the Impact Factors at the fully warmed portion of the table. The early LS1 (1998 Camaro, 2000 Corvette, 1999 Silverado) had a fairly shallow change b/w MAP values, then in the 2000 Silverado the values were opened up dramatically (however note the Impact Factor Gain changes decribed below), then the Impact Factor values were commonized across the platforms, in the 1999-2000 Camaro, and the 2001-2002 Silverado, 2002 Escalade, 2001-2002 Camaro, 2001-2002 TransAm, and 2001-2003 Corvette.





One thing to note, the 2001-2002 Silverado, 2002 Escalade, 2001-2002 Camaro, 2001-2002 TransAm, and 2001-2003 Corvette Impact Factor tables added more resolution than the 1999-2000 Camaro and the other previous tables. In other words instead of just 20, 40, 60, 80, & 100kPa MAP values, they went in increments of 10kPA from 20 to 100kPA, doubling the resolution. Also the ECT values are slightly different making it not directly comparable. I developed regression equations based on ECT to translate the later Impact Factor tables to the earlier table format.

Impact Factor Gain vs Airflow

Once again there seems to be 3 distinct generations of Impact Factor Gains. The pre-LS6 LS1 family had a moderate transition from 1.0 to 0.5, the 2000 Silverado is the anamoly with a shallow transition from 0.75 to 0.4, and the later Silverado/Corvette had a steeper, more gradual slope from 1.0 to 0.2.





Fuel Boiling Time vs Coolant Temp vs MAP

The 3 different characteristics continues with the bulk of the Gen III applications being in the 0.4 to 0.7 range, including the 1999 Silverado, 2001-2002 Silverado, 2002 Escalade, 2001-2002 Camaro / TransAm, and 2001-2003 Corvette. The 1998 Camaro seemingly has a roughly uniform increase as the MAP values increase at a fairly high level. The 2000 Silverado has a linear characteristic but operates at much lower numerical values. One thing to note, these comparisons are made at the 176*F and 198*F temps only, and the overall tables may be different at lower (warmup) and higher temps. I have a full spreadsheet with the ability to compare at other temperatures, it is just not
presented here (similar comment applies to the Impact Factor table above).

Also similar comment with respect to table resolution changes in the 2001 timeframe. Once again a series of regressions were develop to be able to have a direct comparison.





Fuel on Wall Exp Decay Mult vs Airflow

Here we find that the 1999 Silverado, 2001-2005 Silverado, 2002-2005 Escalade, and 2002-2004 Corvette have the same Airflow Multiplier transitioning from 1.0 to 0.5, while the 1998-2002 Camaro / TransAm and 2000-2001 Corvette have this table at Unity across the Airflow range. Once again the 2000 Silverado stands alone at having a much shallower transition from 0.6 to 0.4.





Can anyone explain the actual equations where these constants/gains/multipliers are used? Even having units on these values might help with physically understanding them, as long as these values are not unit-less. I find this book online that describes some equations representing droplet evaporation and wall film evaporation:

https://books.google.com/books?id=6Q...page&q&f=false

pages ~57 is the area. Didn't find much on fuel injection wall impact with a cursory search.

[Throttler]

After looking at this some more, I'm finding that the rich dips on tip in and lean on tip out occur about .5 seconds after the MAF indicated airflow ratio to Dynamic Airflow go away from Unity (1). So this leads me to think that the Dynamic Airflow filtering / VE Correction parameters could be improved upon to more accurately represent the engine configuration. I will say I'm not finding a lot of information on modifying these parameters.

I'm starting to think I'm squarely in redhardsupra territory here...

Dyn Airflow - MAF ratio vs AFR Error.jpgDyn Airflow - MAF ratio vs AFR Error run chart.jpg

[matty b]

Since you have all injector data available to you, why the hell didn't you put it all in? Not saying it will help your issue, but you are wasting time when you already know that fueling variables are incorrect and you didn't change them. Short pulse adder, offset and the rest should match the LQ9.