Two more things I did that helped a lot were making the cam events the same at MP14 and MP21 and also disabled the IMRC for best driveability (both drop down menus set to Open) and also set to zero any IMRC open loads for Fuel Economy that were less than 1.00.
It seems like most of the surging (if you have IMRC) is centered around the 14/21 combo and right on the verge of the IMRC transition.
Will definitely bookmark this thread. Thanks Higgs!
Today I am experimenting with zeroing the Snap To Line 14/21 and just letting it run up and down from 14-22 instead.
Changing 14s to 26s, I found tipin to have brief but noticeable lean spots. Just a heads up and be careful. I'm not sure if this is because it threw the speed density off or the load/torque off it could be both. The MAF curve is used to calculate the engine load. If it is off because it was tuned by fuel trims instead of properly calculated on a flow bench your load values will be off. Seeing as I'm still on the stock setup my load values should be accurate. The ecu tries to use 26, but moves to a different mapped point when the lean spots occur.
Fuel trims should be corrected in the load/torque tables or speed density after a change in intake flow. Correcting them in the maf curve just throws the load values off which effects every thing else. The initial maf transfer values should be obtained on a flow bench. Using a ratio of the new maf housing vs the old should give a fairly accurate maf curve if a flow bench is inaccessible. I've tried a few different ways to calculate the new flow between stock and aftermarket intakes. Most come out to be with in .01 of each other. I'm still working out an accurate way to calculate the flow differences it my come down to fabricating a home made flow bench.
How far off would you say the load would be by tuning the MAF with fuel trims(+ or - 5%, 10%)? I calculated a base correction for my car then dialed it in with fuel trims and my peak load is like 104%(3.7 V6 Mustang) and the idle load is like somewhere in the teens I think. Either way my car pulls like 99.98% of the spark from Mapped Point OP at WOT. I definitely agree though, an incorrect MAF Curve has far reaching implications in the tune. Let us know if you find a good way to calculate flow data on a MAF, that would be really helpful. Do companies like JLT, Steeda Airraid even provide this on request? I don't wanna get too off topic, just curious.
So if the fuel system does not change but the airflow system does then the fueling will indicate the discrepancy through fuel trims (or lambda error).
If there is a different amount of airflow entering the engine, you must recalibrate the airflow model, MAF and/or VE.
It also stands to reason that more air through an engine means more torque produced.
Recalibrating the MAF to reflect the correct airflow is not wrong (it IS wrong if you change injectors and bake that error into the MAF). You must recalibrate the MAF if you change the airflow (intake, IM, cam event changes, etc).
If you do not change airflow modifications, there is really no reason to change the MAF.
One thing to note is headers....many times there is not need to change the MAF at part throttle because the exhaust was not the main restriction, but there will be a difference at WOT.
The period and air mass are both exponential curves so it really depends where in the curve the dial in corrections were applied.
I'm looking into this calculator for a more accurate base correction.
http://www.pipeflowcalculations.com
The faster the air is moving the more accurate the maf is. The new cars are much more accurate than they used to be at slow speeds as well. I didn't use any base correction when I went to jlt and gt350, although the maf was off about 60%, it dialed in quickly, curve looks just like stock.Originally Posted by murfie
The torque tables are not far off stock, in fact, really only 14 and 21 need correction and it's fairly small.....but I did try something pretty cool that is working great! I'll post about it tomorrow.....
Just to clarify, youre copying and pasting the logged values from the tables into the engine torque 0-14, OP tables, and then use the speadsheet to calculate the inverse. Note: when i first did that i had to edit the spreadsheet to match my axis for load values.
they are tools, you may need to set them up for your car. if you had to change the axis in the spreadsheet, the scanner axis' probably do not match your car's either, so change those too.
I remember catching that at some point. i dont remember what i did, ill have to try it again.
So to make sure I am tracking what y'all are saying, in order to properly correct fuel trims with the stock MAF is to apply fuel trims to all the torque tables? Example: correcting the fuel trims with a stock intake but using LU47s
definitely not that.
fuel trims or lambda error gets applied to the MAF curve if you have changed airflow parts, like intake, intake manifold, headers, cams/events, etc.
if you change injectors then you change the injector data.
after air and fuel is correct, you adjust the torque model.....you adjust it with Engine Indicated Torque logged into the Configs.
Higgs Boson thank you for sharing your thoughts, efforts, and results here!
We have been working on refining the tune in my business partners VMP '15 GT and made drivability progress by introducing multiple TQ and Spark tables (it started out with just one mapped point enabled). Anyways, overall drivability was improved but it introduced a noticeable and highly irritating surge just before boost (2,000-4,000 RPMS and 75-105kPa). Anyways, I just modified your tables to work with the Absolute Load (SAE) and Engine RPM PIDs from a drive earlier today.... and we are getting significant wheel torque errors in this range. I expect this is the key to our surging issue. I will report my findings tomorrow.
Side note: for anyone who is making lots of redundant entries or modifications to the list of graphs... it goes much quicker by opening the graph.xml file in a word editor, or better a XML Editor. I needed to change the load PID, load units, and RPM PID in Higg's example, and it went VERY quick with the XML Editor.
How much error are you seeing? i did this a bit like OP by maxing out IPC error and was seeing values like 800 and such.
Seeing averaged values in the 200+ range, with individual hits of 400+
I did some testing and did not get the surging squared away, but will try Higg's procedure again.
In the meantime: can anyone provide a quick rundown regarding "IPC Wheel Torque Error"? Are you using this to dial TQ and TQ-inverse tables?
No use Indicated Engine Torque
so i went out again and gave this a try. ran only OP point to keep it simple. didnt do squat for my throttle issue but made the car peppier.
if you do it for OP but the car surges in another MP then it wouldn't affect anything, right?
are you sure your issue isn't mechanical?
post your tune.
Reply With Quote