NOTORIOUS VR
Administrator
Hero Member
Karma: +58/-7
 Offline
Posts: 1056
|
 |
« on: April 01, 2012, 10:05:06 AM »
|
|
|
I feel that after many revisions, and many late nights logging and being unhappy with one thing or another it all comes right back around to one thing. ME7's TQ management. I'm not sure how other people seem to deal with it, but from what I've learned and tried... you cannot beat the factory programming when it comes to this. They know something that we don't this is clear. As for results, I've seen the best results (still not 100% acceptable to me though) by using and then massaging the stock values. Nothing seems to work better. The car is smooth, and the car has the least amount of TQ interventions. But the car is NOT agressive... it still keeps that factory feeling and while some might like that... I don't. I know it's holding my car back... TQ ramp up is slow and linear but it doesn't hit as hard as it could if we didn't have the TQ management nanny. I have just posted logs comparing masterj's latest tuner wizzard to factory stock RS4 TQ settings... between the two tunes that was the only difference... and while you really only see what happens on a 3rd gear WOT pull on the logs the car in general drives completely different. the RS4 settings make the car drive like it's suppose to in almost every setting... low load great... part load good... WOT... also good... still a few things I am unhappy about, but worlds better then trying to generate numbers on your own or with a wizzard (no offence masterj). Link to logs: http://nefariousmotorsports.com/forum/index.php?topic=617.msg16796#msg16796So, what it comes down to is this. I would like to know how you all deal with this? What are you doing to make your car drive the way you want it to? What parameters do I need to log (either in ecux, me7l, or nefmoto logger) to understand purely the TQ management part of the ECU... What are the best places to start. KFMIOP, KFMIRL and KFZWOP/2 and a few other TQ limiting maps seem to be where the magic lies... For reference the "other" thread with info to all things TQ management: http://nefariousmotorsports.com/forum/index.php?topic=970.0 |
|
|
|
« Last Edit: April 01, 2012, 10:15:44 AM by NOTORIOUS VR »
|
Logged
|
|
|
|
|
s5fourdoor
|
 |
« Reply #1 on: April 01, 2012, 05:19:52 PM »
|
|
|
NVR, my experience is exactly supportive of your findings. Masterj's spreadsheet is not relevant for the B5 RS4-like engine. There are a variety of reasons why, but for starters, I have found the car to drive the best with RS4 KFMIRL and KFMIOP values. The low-load values need to remain similar to their original values; the problem I've had with the RS4-like maps is that nasty interventions occur. I'm very interested to see if you make progress on this.
|
|
|
|
|
Logged
|
|
|
|
NOTORIOUS VR
Administrator
Hero Member
Karma: +58/-7
Offline
Posts: 1056
|
|
« Reply #2 on: April 01, 2012, 06:59:34 PM »
|
|
|
I still have a few, but what maps did you transfer? there are some TQ limiting maps, etc that you would have to copy as well.
let me know what maps you would like to see from me (posted in my tuning thread as well).
|
|
|
|
|
Logged
|
|
|
|
|
phila_dot
|
|
« Reply #3 on: April 01, 2012, 11:10:59 PM »
|
|
|
B_zwvz is getting set one of two ways.
It can be set automatically by B_mibeg. This occurs if miszul_w < mizsolv_w.
mizsolv_w is mibas_w + dmar_w.
miszul_w is maximum permitted torque from %MDZUL.
The other way is if mifa_w != misol_w and misol_w != migs_w.
Log the following:
B_mibeg
miszul_w
mifa_w
misol_w
migs_w
My money's on KFMIOP.
|
|
|
|
|
Logged
|
|
|
|
|
s5fourdoor
|
|
« Reply #4 on: April 02, 2012, 11:02:39 AM »
|
|
|
The truth is that when I tried changing over all of the associated TQ-MGMT maps, my ECU went into perma-brick mode. I think what went wrong was that I also wrote over Pedal Maps and lumped everything together into one update. Let me find the list of maps I think that need to be related to this topic. I'll update this post with that.
|
|
|
|
|
Logged
|
|
|
|
|
s5fourdoor
|
|
« Reply #5 on: April 02, 2012, 11:12:11 AM »
|
|
|
ok normally i'd update my old post, but i'm NEVER this organized, so i'm extra psyched to post this file.
on tab 1 you'll find the rs4's tables, axis values, and full information cleanly formatted.
on tab 2 you'll find the s4's "", "", and "".
on tab 3 you'll find my attempt to interpolate values from the rs4 tables but using the s4 axis.
this was done in MATLAB and should prove to be robust, unless i made a procedural / conceptual error somewhere.
having written all of these maps at one time to my tune, i stuck the car into permabrick mode, and had a nasty double-flashing epc and cel light when the ignition was turned over. however, the car would start and then insta-shutoff. at the time NVR suggested there was a checksum error. given alot of work in conjunction with Infinkc, i realized that my mistake here was writing values to the Pedal Maps.
therefore, all of the maps on the attached spreadhseet include, please assist me to eliminate / add-to this map list:
KFMIOP - Optimal Engine Torque Map
KFMIRL - Engine Load Desired (rlsol)
KFMDZOF_UM - Map: Offset tolerance depending on allowed torque
KFMI_UM - Optimal engine torque under monitoring
KFMIZUFIL - Allowable indicated torque to torque limit before filter
KFMIZUNS - Allowable torque on afterstart extension
KFMIZUOF - Allowable torque for torque limiting indexed
KFMPED_UM - Map for allowable torque from pedal position in the functions monitoring
KFMPNS_UM - Map for allowable torque from pedal position on cold motor
KFPED_0_A - Relative torque request from pedal
KFWDKMSN - Map for desired throttle plate angle
KFWDKPP - Default throttle plate angle from charge signal
KFWDKSMX - Maximum desired throttle plate angle
KFZWOP - Optimal ignition angle
KFZWOP2 - Optimal ignition angle (variant 2)
also:
KFMSNWDK - Normalized mass flow over DK
I believe I went through all m-box XDF definitions under the Torque and Pedal section, comparing between S4 and RS4.
I'll double-check to make sure none of the group definitions are missing from the above list now.
|
|
|
|
« Last Edit: April 02, 2012, 05:19:06 PM by nehalem »
|
Logged
|
|
|
|
NOTORIOUS VR
Administrator
Hero Member
Karma: +58/-7
Offline
Posts: 1056
|
|
« Reply #6 on: April 02, 2012, 01:44:55 PM »
|
|
|
well that looks mighty interesting! I will have to give that a whirl along with a few other ideas later on.. maybe tonight some time or this week...
getting to the bottom of the TQ management is probably one of the most important things IMO, it's what will unlock the real performance of bigger turbo cars.
|
|
|
|
|
Logged
|
|
|
|
|
s5fourdoor
|
|
« Reply #7 on: April 02, 2012, 04:57:03 PM »
|
|
|
A major problem I've come across while looking at the RS4 and S4 pedal maps in unison. The S4 seems to have a valid pedal range of 0-100, the RS4 on the other hand seems to have a maximal pedal value of 52, compared to 100. What this means or how this possibly works, I have zero knowledge. This is another reason I am very skeptical about using any RS4 pedal-related items. The water gets strange because some of the torque maps have DIRECT relationships to the pedal maps. AKA, this is a very difficult problem.
|
|
|
|
|
Logged
|
|
|
|
|
|
|
nyet
|
|
« Reply #9 on: April 02, 2012, 05:33:09 PM »
|
|
|
just a nit pick. its "wizard"
|
|
|
|
|
Logged
|
|
|
|
|
phila_dot
|
|
« Reply #10 on: April 02, 2012, 05:40:18 PM »
|
|
|
You are not isolating and fixing the root problem, instead you are just neutering the ECU's ability to respond to it. Nehalem, torque is directly linked to pedal request (%MDFAW iirc). Preferably, we can gain a full understanding of the torque path vice just finding something that works. NVR, please include the variables that I listed above the next time you log. That should give us an idea of the source of the problem. |
|
|
|
|
Logged
|
|
|
|
|
s5fourdoor
|
|
« Reply #11 on: April 02, 2012, 08:13:05 PM »
|
|
|
You are not isolating and fixing the root problem, instead you are just neutering the ECU's ability to respond to it.
perfectly worded |
|
|
|
|
Logged
|
|
|
|
berTTos
Full Member
Karma: +24/-0
Offline
Posts: 91
|
|
« Reply #12 on: April 02, 2012, 08:23:38 PM »
|
|
|
gentlemen,
let's recall that there are 2 levels of Torque intervention. level 1 seems related to driving comfort and refinement while level 2 seems to function as a safety mechanism to mitigate DBW throttle issues that may ultimately result in a runaway vehicle.
in my experience - it is desirable in a high HP vehicle to all but disable Level 1. Level 1 is what inhibits a 'lively' throttle response.
Level 2 cannot be disabled (as far as i've been able to discover) and is there to intervene when the ecu calculates a great enough deviance between requested load and actual load - which could indicate a stuck throttle plate, an implausible pedal module signal, a gross MAF misread, etc - all of which could cause unintended acceleration.
now - the 2 most important maps relating to Level 2 interventions are KFMIRL and KFMIOP. as we have discussed, KFMIOP needs to contain values derived from KFMIRL via interpolation. someone (forgive me - i can't recall who) learned that ME7 uses linear interpolation when performing the calculations. this means that a linear interpolation for deriving KFMIOP works perfectly (even though Audi did not use exact linear values in the factory files, for some reason. Klaus forgot his TI-89 that day maybe).
so - what if you have properly interpolated KFMIOP from your desired KFMIRL and you're still experiencing Level 2 interventions (scary throttle cuts or 10% WGDC)?
check the following -
1. are you using the stock airbox? an open element without proper correction definitely has the potential to trigger Level 2 intervention. the 85mm MAF w/o flow straightener with the stock airbox suffers significant turbulence at various throttle positions and RPMs and this is only exasperated by running an open element. the turbulence is sufficient enough to result in artificially low MAF values which may deviate enough for a brief time to result in load miscalculations great enough to trigger Level 2 intervention.
solution? remove the open element or perform correction for it. (correction is really needed for stock box as well)
2. have you deviated from the factory load spread in KFMIRL? by this, i mean, have you increased particular columns or cells at differing rates? i have found that the best approach is to simply multiply the entire table by a percentage rather than attempting to tweak particular areas independently, thus maintaining the relative load increases across the various % Torque requests. the problem with raising portions of KFMIRL by differing percentages seems to be related to a lack of resolution in the Mbox. we simply need more columns. mbox wasn't made to nicely deliver 500+ HP. when we deviate from the stock load spread, even with an appropriately calculated KFMIOP, Level 2 interventions are much more likely to occur.
i wouldn't spend time trying to copy over RS4 values. RS4 maps have more columns and are too tame anyway. also, as we all know, a particular map is made to function with all of the other maps. it's like trying to copy a gene sequence from one animal to another. it never turns out the way you think bc the sequence is only fit when it is accompanied by the rest of the genetic map.
start by raising KFMIRL for all data points by a percentage - then interpolate KFMIOP and go from there.
you can slightly tweak KFPED to request more a little earlier as well.
examine my K04 file that i posted and note the axis changes and the changes to the Level 1 TM tables.
that file is perfectly smooth at all throttle inputs and has not once exhibited a Level 2 intervention in over 20,000 miles. it is also pretty aggressive and is a hoot to drive, even just down the street to get coffee.
also - don't touch the _UM maps or you will end up with a bricked ecu on your first startup (boot loop accompanied by twitching throttle plate sound familiar?). thx to phila-dot for advising me on those (used for calibration).
|
|
|
|
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #13 on: April 02, 2012, 08:58:13 PM »
|
|
|
the turbulence is sufficient enough to result in artificially low MAF values which may deviate enough for a brief time to result in load miscalculations great enough to trigger Level 2 intervention.
Are you sure about this? I found more interventions with the MAF too HIGH, not too LOW.. a lot of my throttle/wg interventions went away when I slightly underscaled my MAF. I'm still tracking down the source of my timing interventions... do these also fall into two types if intervention? |
|
|
|
|
Logged
|
|
|
|
|
jibberjive
|
|
« Reply #14 on: April 02, 2012, 10:48:30 PM »
|
|
|
I love this thread. This exact topic has been on my mind, but I wasn't going to try to address it until I got a couple of other things ironed out. I definitely think there is some good optimization that can still be done with this regarding bigger turbos. ok normally i'd update my old post
Awesome file! Did you ever get around to trying the RS4 interpolation that you did on your previous excel file where you convert RS4 to S4 and scaled the top end? How did it run if so? |
|
|
|
|
Logged
|
|
|
|
|
