Pages: 1 2 [3]
Author Topic: WGDC @ 95% in spite of high overboost (S3 8L APY)  (Read 3979 times)
prj
Hero Member
*****

Karma: +1072/-480
Offline Offline

Posts: 6034


« Reply #30 on: March 10, 2024, 04:23:25 AM »

From your log you don't have any boost control.
It's riding IMAX the entire time and is unable to do anything.

Set your entire LDRPID module to stock and then calibrate from scratch based on guides here.
Making random changes to IMX or LDRL is a waste of time.

Before you do fix your fueling.
Logged

PM's will not be answered, so don't even try.
Log your car properly - WinOLS database - Tools/patches
vitali1411
Newbie
*

Karma: +0/-5
Offline Offline

Posts: 19


« Reply #31 on: March 13, 2024, 01:06:04 PM »

Hi,

today I got my injectors. It´s Bosch 0280158123 (630ccm).

I flashed an ori .bin-file and adjusted before:
KRKTE to 0.0576 --> calculated like S4 Wiki says.
TVUB to the values I got from this forum.
I left TEMIN alone.
FKKVS all values to 1

Started the car, waited some minutes in idle and drove some KMs. Feels good so far.

@prj
I´ve a question regarding your post before:
Where can you see that there is no boost control? Boost is following LDRXN/IMX like programmed. KFLDHBM was maxed out by the tuner.

Hmm okay, yes, it´s following IMX. But where can you see that it´s following IMX and not LDRXN? But this is normal condition under WOT or? See S4 Wiki excerpt below.

In my log rlmax is following rlmx, means boost is restricted by LDRXN?! --> as programmed

LDRL wasn´t changed by me or the tuner.
I only changed IMX in the areas where the log reported over/underboost.

I refer to this part of S4 Wiki/Did I understand wrong?:
"If your actual boost is not meeting requested boost, you may have to increase the PID I limit for 850 and 1000mBar. In general, you want KFLDIMX to follow what you expect your WGDC to be in the steady state, so after peak boost, you should set this to where you want the WGDC to settle. Note that PID trims (I-Regulation adaptation) may alter this limit. It should also roughly follow the profile of requested boost when it is tracking LDRXN. That is to say, if you have a flat max (WOT) boost request, you should have a flat KFLDIMX. The reason IMX is so important in ME7 is that in the steady state, P result is near zero and D is scheduled out with B_lddy. This means that I dominates.

On a WOT pull, I is maxed out due to integrator windup (accumulated error) so it is simply following IMX."
« Last Edit: March 13, 2024, 01:34:41 PM by vitali1411 » Logged
prj
Hero Member
*****

Karma: +1072/-480
Offline Offline

Posts: 6034


« Reply #32 on: March 13, 2024, 01:54:53 PM »

IMX should not be used as boost control.
PID I should not be maxed and riding limiter

http://nefariousmotorsports.com/forum/index.php?topic=12352.0title=

Now you have effectively no boost control, just a DC limiter, as is visible from your log.
There's nearly no reaction to overboost by the controller or underboost, it's just running fixed dc.
Logged

PM's will not be answered, so don't even try.
Log your car properly - WinOLS database - Tools/patches
vitali1411
Newbie
*

Karma: +0/-5
Offline Offline

Posts: 19


« Reply #33 on: March 17, 2024, 06:00:55 AM »

Hi again,

after I had adjusted KRKTE + KFZW + KFLBTS, the car feels better. I could improve the ign retardation.
Requested load peaks @ 220%/4000upm, falling to 170%/6800upm.

KFLBTS seems to be in line with my AFR gauge. Is it to rich?

The next step is to make an effort to check the boost control and understand the linked thread.

The car managed ~230g/s for the first time. The charge air temperature rose by approx. 10 degrees during a pull.
The duty cycle is 72% with the original K04-022 WG actuator.

Can I draw conclusions about the performance from the air mass? In some forums you can read for 1.8T with K03/K04 --> Airmass*1,3= ~PS.
How do I recognize the limit of the turbo?

Do I have to worry about the connecting rod? - I assume so?!?!

Edit: As prj already said.. There is no boost control. By changing LDRXN I only aligned the load curve to the actual boost. --> I should really put some effort in boost regulator Smiley
« Last Edit: March 17, 2024, 06:44:41 AM by vitali1411 » Logged
vitali1411
Newbie
*

Karma: +0/-5
Offline Offline

Posts: 19


« Reply #34 on: March 18, 2024, 03:53:31 AM »

Hello,

I hope posting a screenshot is okay in this case.
I read S4 Wiki and the linked thread several times..

What I understood is:

KFLDIMX:
- We assume that a certain boost pressure value is assigned a percentage, regardless of the engine speed.
-- For example of my screenshot: We assume that 884 hpa is 51%

KFLDRL:
- The boost in % is the input for the y-axis
- The output is then the values in the table as WGDC (%) depending on the engine speed

Now I have to tune KFLDRL by logging, an example:

- Tune KFLDRL for 50%
- Calculated boost for 50% = 859hpa
- Set entire KFLDHBN to 1.8594
- Ensure that LDRXN is above KFLDHBN
- Log from 1000 upm to redline
- do this for all percentages
- Adjust KFLDRL or use ME7Tuner

What we do is logging to predict the WGDC for a certain boost and RPM. That's called feed forward an shall improve the reaction and accuracy of the PID.

Am I on the right track?
It´s not that easy to get this together out of all the posts and threads.
« Last Edit: March 18, 2024, 05:07:40 AM by vitali1411 » Logged
ratosluaf
Jr. Member
**

Karma: +1/-3
Offline Offline

Posts: 37


« Reply #35 on: March 19, 2024, 12:28:50 AM »

Quite not.
- Set entire KFLDHBN to 1.8594
KFLDHBN is max pressure to ambient ratio, so with this you're setting your boost limit to 0,85 bar in "normal" (1 + 0,85) circumstances. You can limit your boost with this map or with ldrxn.

For logging purposes, you need to set your boost request higher than your boost target in order to not get any interventions.

Make your kfldimx linear with maximum cell higher than boost target, then do cwmdapp = 8, and make runs with kfldrapp forcing wgdc from 10 to 95% or max  you can safely run. Then use ldrpidtool and just paste your kfldrl output, thats all

Just do not get any boost interventions during logging, this tool makes it pretty damn easy and accurate.
Logged
vitali1411
Newbie
*

Karma: +0/-5
Offline Offline

Posts: 19


« Reply #36 on: March 19, 2024, 01:57:05 AM »

Thanks for your answer @ratosluaf,

- "Make your kfldimx linear with maximum cell higher than boost target"

I did, or is my KFLDIMX wrong too?

In general I misundestood the 10 to 95% --> I thought limit boost via KFLDHBN and adjust KFLDRL according to the logs.
That way seems to work too, but of course I will try the way you mentioned.
Logged
Pages: 1 2 [3]
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.016 seconds with 16 queries. (Pretty URLs adds 0s, 0q)