gameover, do you have time for an in-depth on changing Dynamic Airflow Filtering?
Thread Starter
TECH Addict
Joined: Feb 2004
Posts: 2,826
Likes: 0
gameover, do you have time for an in-depth on changing Dynamic Airflow Filtering? I have been looking over your posts, which have some good stuff, but there really isn't much about the affects of changing the Dynamic Airflow calculation and filtering values. Maybe even a comparison of Z06 screenless MAF values, and screened MAF values, with an idea of what the differences do? I understand if you are too busy...it's not a simple subject. I am just having trouble determining how the values are used in filtering and calculation so that i can make proper changes.
I'm with you on this one. While your there, if you could explain the dynamic airflow zoning a little more, it would be appreciated.
TECH Fanatic
Joined: Mar 2003
Posts: 1,839
Likes: 0
From: La Porte, TX
I kind of suspect that the boundries are used to serperate how much maf vs ve blending is going on. But the way the are set from the factory does not leave much resolution. I was thinking about setting them to create more resolution, but I'm not too sure about the co-efficients, and filters, and how they effect each particular boundry. I have read the help file for the dynamic airflow section at least 10 times, and I still don't get it.
Thread Starter
TECH Addict
Joined: Feb 2004
Posts: 2,826
Likes: 0
Originally Posted by WS6snake-eater
I kind of suspect that the boundries are used to serperate how much maf vs ve blending is going on. But the way the are set from the factory does not leave much resolution. I was thinking about setting them to create more resolution, but I'm not too sure about the co-efficients, and filters, and how they effect each particular boundry. I have read the help file for the dynamic airflow section at least 10 times, and I still don't get it.
Trending Topics
LS1 Tech Stories
The Best V8 Stories One Small Block at Time
Gas Monkey Built a 6-Wheel Ferrari Testarossa With a Corvette LT4 Engine
Verdad Gallardo
7 Most Reliable High-Performance Engines GM Has Ever Built
Verdad Gallardo
Amazing '71 Camaro Restomod Is Modern Muscle Car Under the Skin
Verdad Gallardo
6 Common C5 Corvette Failures and What's Involved In Repairing Them
Pouria Savadkouei
Retro Modern Bandit Pontiac Trans AM Comes With Burt Reynolds' Autograph
Verdad Gallardo
Top 10 Greatest Cadillac V Series Performance Models Ever, Ranked
Pouria Savadkouei
Top 10 Most Powerful Chevy Trucks Ever Made!
Hennessey's New Supercharged Silverado ZR2 Has 700 HP
Verdad Gallardo
Coachbuilt N2A Anteros Is an LS2-Powered C6 Corvette In Italian Clothes
Verdad Gallardo
FormerVendor
Joined: Jan 2003
Posts: 525
Likes: 0
From: Melbourne, Australia
time is not on my side at the moment 
From the help file:
"As engine operating conditions change the current zone changes and this selects the appropriate filter equation to be used in the dynamic airmass calculation"
The thing here is that we are still missing the filter equation coefficient table from the editor. At 1.6 we couldn't work out an easy way to display it in the editor as it is a little funky.
When we add it, you will still need a degree in linear predictive filtering and have access to something like MatLab to see the transfer functions and understand them. Although i will do my best to explain it...
Cheers,
Chris...
From the help file:
"As engine operating conditions change the current zone changes and this selects the appropriate filter equation to be used in the dynamic airmass calculation"
The thing here is that we are still missing the filter equation coefficient table from the editor. At 1.6 we couldn't work out an easy way to display it in the editor as it is a little funky.
When we add it, you will still need a degree in linear predictive filtering and have access to something like MatLab to see the transfer functions and understand them. Although i will do my best to explain it...
Cheers,
Chris...
Thread Starter
TECH Addict
Joined: Feb 2004
Posts: 2,826
Likes: 0
I have been pondering that for awhile. I *think* that the zones determine how much filtering is done. More in some, less in others. I think it may affect how smooth the engine runs by more effectively using filtering based on engine characteristics and how well the engine transitions through different running conditions. Of course this is just speculation...the HPTuners guys are holding the keys to the car.
Uhm...if it's not too much trouble...I don't suppose we could get some more info on O2 calibration...I am seeing some interesting possibilities there as well...
Anyways...I'm still waiting to pick out my J-body...
Uhm...if it's not too much trouble...I don't suppose we could get some more info on O2 calibration...I am seeing some interesting possibilities there as well...
Anyways...I'm still waiting to pick out my J-body...
When we add it, you will still need a degree in linear predictive filtering and have access to something like MatLab to see the transfer functions and understand them. Although i will do my best to explain it...
Filtering can be complicated, but most of the time it's a series of common known filtering methods. I don't know your background Chris (although I know you guys are great programmers and hackers), but I'm a PHD student working in experimental fluid mechanics and work with filtering my experimental data; although I don't do much predictive filtering. If it is with in my scope I wouldn't mind spending some time to help figure it all out. I'll have some time on my hands after finals in a few weeks.
Originally Posted by gameover
When we add it, you will still need a degree in linear predictive filtering and have access to something like MatLab to see the transfer functions and understand them. Although i will do my best to explain it...
Cheers,
Chris...
FormerVendor
Joined: Jan 2003
Posts: 525
Likes: 0
From: Melbourne, Australia
i think i described it in another post around here somewhere but basically the filter is a predictive thing that predicts the new airflow value something like:
Airflow(t) = K1.Airflow(t-1) + K2.MAFair(t) + K3.MAFair(t-1) + K4.MAP(t) + K5.MAP(t-1) + + K6.MAP(t-2) + K7.TPS(t) + K8.TPS(t-1) + K9.TPS(t-2)
The K1-K9 values are selected from a multi-state table (18 states from memory) so you have 18 sets of coefficients dependant on engine operating conditions (zone). The thresholds and limits determine the zones and the coefficient matrix determines the airflow prediction. This table is not in the editor at present.
What you see from the filters is varying "favor" between the VE, MAF, TPS and MAP transients depending on operating conditions. eg. at steady state cruising the airflow weights towards the MAF heavily, but during other conditions terms from VE, MAP and TPS come into play to correct for the MAFs inability to accurately reflect cylinder loading during transients and situations where you may have large manifold/intake leaks.
It's an interesting thing, but the code is a twisty mess of 18 way state machine mixed with steady state detection routines and component failure bypasses. And LS2 is even more fun...
Chris...
Airflow(t) = K1.Airflow(t-1) + K2.MAFair(t) + K3.MAFair(t-1) + K4.MAP(t) + K5.MAP(t-1) + + K6.MAP(t-2) + K7.TPS(t) + K8.TPS(t-1) + K9.TPS(t-2)
The K1-K9 values are selected from a multi-state table (18 states from memory) so you have 18 sets of coefficients dependant on engine operating conditions (zone). The thresholds and limits determine the zones and the coefficient matrix determines the airflow prediction. This table is not in the editor at present.
What you see from the filters is varying "favor" between the VE, MAF, TPS and MAP transients depending on operating conditions. eg. at steady state cruising the airflow weights towards the MAF heavily, but during other conditions terms from VE, MAP and TPS come into play to correct for the MAFs inability to accurately reflect cylinder loading during transients and situations where you may have large manifold/intake leaks.
It's an interesting thing, but the code is a twisty mess of 18 way state machine mixed with steady state detection routines and component failure bypasses. And LS2 is even more fun...
Chris...
