NefMoto

Looking for more experienced answers to a couple questions. 

Its a 1.8t AWM Manual Quattro A4.  2.0t coils, sai delete, cat delete, single mass 240 mm flywheel with s4 clutch, 93 octane.

Modifications have been made to:
LAMFA and LDRXN and the KAT and SAI crosswords and Euro-bytes

I'm having some issues with lambts coming in when EGT model before cat(tabgm) is around 690.  EGT for part protection (Tagbts_w) is reading 840 at the when I get the drop in AFR current(lamsoni_w).  At the same time, optimum ignition angle(zwopt) is at 25.5 but current ignition angle is at about 8.  From reading the Bosch me7.5 Function and Data Sheet I understand that a portion of lambts fueling deals with inefficient ignition angle.  So my understanding would be that by feeding more timing in through KFZW at higher loads thoughout the whole rpm range.  Am I correct?

Second, I noticed that Engine Load(rl) is up around 170%.  Corrected Engine Load(rl_max) and Requested Engine Load(rlsol_w) are around 160.  Can I push rl_max and rlsol_w higher by modifying LDRXN, KFMIRL, and KFMIOP higher to reach that 170%?

When I have an internet connection for my laptop I will upload my original, first tune, and next tune, with a xdf.(http://[img])[/img]

Make sure the constant for the BTS activation is set high enough so it wont trigger until you want it too.

I understand that I can bump the threshold for activation.  I've read through several threads that have simply raised the threshold and stopped the fueling intervention.  If bts fueling is recognizing an ineffecient timing as elevating the exhaust temperature because there is a factor for poor fuel quality, and a higher quality of fuel is being used that allows more advance, it would allow the addition of timing to reduce the temperature of the bts model and bring it back on line with the EGT model.  Im trying to gain a deeper understanding of what is going on within the system to get it to do what I want it to.  Once ive improved timing I will revisit TAGBTS

regarding the high calculated load, have you re scaled your MAF or perhaps MAF is reading high?  This will cause calculated load to be artificially high.

I will log the maf reading.  The maf is a Bosch replacement with all of 2000 miles on it.  If we think its off I can get it replaed again.  Is that 170 load reachable on the k03s?

Post logs please.

Compare ps_w to actual boost to see if your MAF is calibrated correctly.

I doubt the MAF is faulty; you might just have the wrong transfer function.

MAF may not be faulty, but could be the wrong part altogether and read higher than expected.

k03s will flow about 175g/s if you ring it's neck.

OK.  Finally have internet.

I am uploading my current xdf which Ive compiled using the work of several others on here.  A lot of what I have in the new tune goes to sb_gli, thelastleroy, and nyet on the 1.8t AK community project.

Ive made revisions for the 2.0 coils, LDRXN, LAMFA table and axes, KFZW, MSLUB, and ESKONF.  I havnt checksumed any of this.  If you all feel I could put this in and not wreck something I will checksum it and load it in tonight.  I have several other maps defined but it was getting to complicated organizing them all in TunerPro.

and the definition file

Sorry for the extra posts, my log was too big and i had to remove a portion of it to get it below the attachment size.

You modified it not good.  Don't use excel or open office to modify the file, it fucks up the format and things can't parse it.  Just modify the original with notepad.

Sigh. Excel is a fucking cancer.

Attempt 2 through notepad

No ps_w in the log.

That said, lambda control looks stable, so the MAF readings might be right, unless KRKTE is wrong and compensating for it. Seems unlikely.

You could just use BTS to drive fueling the way you want it.

Could you take a look at KZFW and tell me if I have the right idea or if im overdoing it?  I feel really good about the other changes.

Ive set up my logger to get ps_w next time.  Anything else I should log?

From your logs timing looks fine.

I disagree.  Lambda control is disabling at 3,500 (for who knows what reason), but right before that they and going < 1.  I think MAF readings are a bit high still.  Also MAF reading is > MAF @ Throttle plate, another induction that maf could be reading high.

Ignition angle is wacky because of load > specified.

As a test, modify KFKHFM in the upper load columns.  Take like 10% away and see what happens. 

At 3,500 fuel request goes pig rich.  There's more to BTS fueling than just KLFBTS.   You'll need to work on this as well.

Spool up is lazy, could get more DC for faster spool.... and your diverter valve is shot.

The lambts module is based around two things from what I see in the me7.5 guide.  The first is a temperature model on front of the cat.  The second is the delta in ignition between optimum and current.  When I look at optimum vs current from the logs there is a massive gap.  Then I found the delta for ignition angle efficiency and there's a spike in inefficiency before I go pig rich.  Ive added a degree and a half of timing at 1760 and 2000 rpm and 140 load and then duplicated the timing map from the ak tune. 

The diverter valve is a new 710n.  Vacuum line to controller is a good silicone vac line.  Could the n249 valve be the culprit?

I'm going to get the revised tune loaded in and get a log and see what that does.

The map showing the elevated by a timing issue

OK.  Loaded it, logged it, modified it, logged it, modified it again...

and the logs

Start with fixing boost. Not sure what is going on with your PID, but you don't have enough DC and you are liable to run into positive deviation.

Log the rest of the PID parameters.

Get boost to follow req properly and only then move on to fueling and timing. You're more or less safe for now until you get boost set up properly.

I'm in the process of tuning my AWM with a Mazdaspeed K04 (the full turbo, not a hybrid) and 630 Dekas.

I also ended up using KFLBTS (lambda for component protection) to dial in the AFR's. Unlike my B6 S4 where I just used LAMFA, I found I needed to use KFLBTS.

The one thing that you may want to consider is that in order to use KFLBTS for a fueling strategy is that you need to zero out the values in KFFDLBTS (factor lambda for component protection).

As I understand it, the values in KFFDLBTS will be applied to KFLBTS when determining the fueling strategy and as you can see, the values in KFFDLBTS are all over the board.

Also, as others mentioned, you need to make sure that the EGT intervention is low enough that the system uses KFLBTS right from the get-go.

Here's a copy of my stage 3 file if you want to have a look see at what I'm talking about.

PS: to those who know what they're doing, please be kind, I know there is some hack tuning in there, but it all seems to work lol.

PPS: I know that my requested fueling values look to be lean, however my KRKTE is a little out and I needed to compensate. Since this is a narrrowband system, the fueling values are essentially arbitrary for all intents and purposes.  

PPPS: this file was written for some 105 octane fuel I was running, hence the high timing values.

PPPPS: *** I'm also running an e-boost controller, so the boost tables could be completely fucked, I just took my best guess based on the data i had at hand. My reason for the e-boost controller is that the lines are a bitch to hook up and since home was 4 hours away when I was scrambling to get all this done, I elected to just use my e-boost controller as I had enough issues getting the car drivable as it was (took about 8 revisions on the side of the road to get the fueling dialed in :p )

Cheers :)

(https://cdn.thingiverse.com/renders/bc/7c/e7/dd/b8/81ae77dac3520ed975004263321a9568_preview_featured.jpg)

Lol, fuck yeah, TPB rulez!!!! ;)

My goal was not to use lambts as a fueling strategy.  Your afr comes from 3 maps, lamfa driver requested, lambts part protection and Lamfakr knock control.  The ecu will select the lowest lambda and use it.  I've seen several other 1.8s going stupid rich way to early due to part protection coming in.  I was trying to reduce the delta between ignition angle and optimum ignition angle to correct the egt model and make lambts come in more appropriately.  I just ended up making the threshold high enough that it wasnt interfering as I was picking up knock before I could get the delta low enough. 

uh. I hope you weren't doing this by messing around with zwopt

No.  I was adding timing through KFZW.  The efficiency is calculated using the delta between KFZW and ZWOPT.

So I did a check on the health of my turbo today and everything was good.  Shaft play was good, wastegste had good tension and no signs of cracks, Turbines looked great.  So I went into KFLDIMX and added some duty cycle at 2250 and 2500 at the 1000 and 850 mbar regions.  Will test tomorrow once the downpipe goes back on.

Can anyone explain to me how to log tans or add it to the list in me7logger gui?

Made the changes.  I try and take a log before and after to keep weather from skewing the results, and try to do it in roughly the same patches of road, we dont have much flat straight road in Tennessee.  I had earlier spool.  Still am short of what Im after.  Should I add more wastegate duty cycle?

A lot happier with spool.  Still Going to work on duty cycle beyond 3000 but really happy with it

Last log looks really good on the bts and timing stability fronts. You could probably due with a small bump in kfmirl and corresponding kfmiop in anticipation of boost yet to come.