This is on an 06A906032HS, it's a transverse 180hp wideband 1.8T w/ 5120 hack.

This looks like it's clipping (ee term) at 191.25 to me.  I know the KFMIRL/KFMIOP and timing tables end at 191.25% rl on S4s.

I usually don't run the car this hard but I'm trying to debug/wrap my head around ps_w to rl_w and aligning them better (input welcomed here)

might want to change the topic

Stock KVLAD or multyplied with 2?

Try kvlad stock and you will see some modifications in the log.
In my 032hs xdf i have 2 kvlad defined
0x10D48 1 byte 0.007813 * X
0x10D49 1 byte 0.007813 * X

AND other thing , i think you touched LDORXN.
Did you have this error : 17963 - Charge Pressure  P1555 - 35-00 - Maximum Limit Exceeded??? Is Your throttle cutted?

LDORXN is stock.  I attached my bin and xdf if it's of any use.  I have like once or twice (since turning the boost way up) experienced something that felt like a sudden, brief, drop in torque.  KFFLLDE has been modified to pull more load in the event of significant knock retard

I also have kvlad at:
0x10D48 1 byte 0.007813 * X, value is currently 0.5938 (stock is 0.2969 so this is multiplied by 2)
0x10D49 1 byte 0.007813 * X, value is currently 1.5626 (stock is 0.7813 so this is multiplied by 2)

I imagine one of these is correct, I'll look into it and report back.

I found this interesting:




I am experiencing something similar.  As the RPM drops to idle (for instance, depressing the clutch after engine braking at let's say 2000rpm) I get a brief surge up to 1200rpm, then the idle settles near 760.  This is somewhat sporadic, and I've had this issue for a long time.  I think probably as long as I've been running that 5120 file.   

Every once in a while, the engine also races to approximately 3000rpm for a few seconds, then settles into its usual warmup behavior, this got my attention right away.  It's only happened maybe twice, and the weather has gotten substantially colder here in the past two weeks.  My gut says this probably has something to do with the temperature change, but I have not properly investigated yet so I can't say for sure.


I will investigate more and report back either way.  Annotated FR picture attached for my own reference and anyone else who ends up here through searching.

It looks like 0x10C3E for the HS and 0x10BDC for the CB are the same point in memory just with an offset. Assuming that is the case, and we know KVLAD is at 0x10CE7 in the 018CB file, it's 0x10B from our reference point.  If we apply this to the 032HS file, and assume KVLAD is 0x10B from the reference point, it gives us 0x10D49. Comparing to 8N0901018CB, it looks to me like 0x10D49 is correct. 

In that case, following the same logic (which may be flawed, I'm not a professional), KUMSRL is at 0x10D48 in the 032HS file.

Hey Tao,

I cross checked KUMSRL with the TT file and I believe the location I mentioned is correct.

My car idles fine, it’s just the transition to idle from engine braking is a bit strange. It might be a hardware issue too on my end.

What do your fuel trims and MAF readings look like? Is estimated airflow following actual?


Sent from my iPhone using Tapatalk

LAMFAWKR isn't used on these cars stock, that is to be expected.  They just run lambda=1 until the stock TABGBTS of 400C is exceeded, then it uses the BTS system fueling for EGT enrichment. During knock, it is my understanding these files, as a general strategy:

1. yank timing
2. run higher calculated EGTs as a result of the degraded ignition angle efficiency (calc EGT goes to the moon real fast if you knock under significant load)
3. BTS reacts to the degraded ignition angle efficiency and subsequent increase in EGT by significantly enriching the mixture
4. If you're running really terrible fuel, and the car pulls enough timing for long enough, the "continuous knock" flag gets set, the ECU runs off LDRXNZK (on these cars it adds some load in some places to counteract the loss in power from ign retard and BTS dumping fuel)

 There's an excellent thread titled "using BTS all the time" or something like that, that thread is summarized on the s4wiki under the lamfawkr section.  I implemented it and found it very helpful when I recently got a bad tank of fuel.  It's a fundamentally different strategy to how the 180hp 1.8Ts are tuned stock.  I prefer it, but none of this is relevant to the topic at hand

I think I have a bad check valve on the brake booster line, the pedal goes hard when the car sits for over a week.  This could probably contribute to the problem, I'm not going to diagnose further until I know the hardware is functioning properly (it clearly isn't right now).

If you have any interest, maybe log rlsol and rl_w, before and after making the changes I described to KUMSRL and KVLAD (at the addresses I specified).  I will do the same this weekend or maybe tomorrow if I have time. I think it will be useful to compare notes on this.

Nyet, if you're in here, should I be posting this in the 5120 hack thread, or just PM tao until we get this resolved and then update the 5120 thread? I don't use forums much and I don't want to cause clutter.