Pages: [1] 2
Author Topic: Mk4 1,8t AUQ Boost Pumping  (Read 5864 times)
Burnman6666
Jr. Member
**

Karma: +1/-1
Offline Offline

Posts: 33


« on: April 09, 2022, 08:50:09 AM »

Hello Again,

I have just make some trys to get more boost in the car of my Girlfiernd and have some Problems.
The Car:

2001 Mk 4 Gti Aum/Auq
K04001
S3 Injectors
3inch Exhaus from Turbo
Smic Upgrade Intercooler

My Sofware runs good but only the boost doesnt wants to stay you feel some Pumping it goes up and down you even can hear it. but only under 4000rpm  above its getting better.

Yesterday i tried to mod KFLDHBN but it doesnt fix the Problem.

I attached a log in 3rd Gear and the .Bin and .XDF

Software is modded in Injectors, Sec Pump, 2nd Lambda and Boost maps. Even some less Ignition.

May you can help me and have a solution?

Greetings from rainy Germany

Logged
nyet
Administrator
Hero Member
*****

Karma: +608/-168
Offline Offline

Posts: 12271


WWW
« Reply #1 on: April 09, 2022, 01:24:59 PM »

Log the PID vars. has nothing to do with HBN
Logged

ME7.1 tuning guide
ECUx Plot
ME7Sum checksum
Trim heatmap tool

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your ex
Burnman6666
Jr. Member
**

Karma: +1/-1
Offline Offline

Posts: 33


« Reply #2 on: April 09, 2022, 03:01:41 PM »

Thanks for your Answer nyet.

What variables do you mean exactly?
Boost Pid D Result
Boost Pid Max
Boost Pid Min
Boost Pid I Result
Boost Pid P Result?

Sorry but what means HBN?

And may is anywhere a guide for using the CSV in the EcuXPLot?
Found the FAQ but i didnt be able to use this CSV in ECuXplot , when i load it in to it i cant choose the different Variables.

Thank you very much.
Logged
Burnman6666
Jr. Member
**

Karma: +1/-1
Offline Offline

Posts: 33


« Reply #3 on: April 10, 2022, 05:27:00 AM »

Thaught about the HBN and realised you mean the HBN of KFLDHBN  Cheesy Sorry for my stupidness.
Will log the Pids i wrote in a few Hours.
Logged
Burnman6666
Jr. Member
**

Karma: +1/-1
Offline Offline

Posts: 33


« Reply #4 on: April 10, 2022, 10:18:51 AM »

Hopefully with all Data you need.. .

Greetings
Logged
nyet
Administrator
Hero Member
*****

Karma: +608/-168
Offline Offline

Posts: 12271


WWW
« Reply #5 on: April 10, 2022, 02:42:24 PM »

Your IMAX is doing some very strange things, which is causing massive integrator windup and boost overshoot.
Logged

ME7.1 tuning guide
ECUx Plot
ME7Sum checksum
Trim heatmap tool

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your ex
nyet
Administrator
Hero Member
*****

Karma: +608/-168
Offline Offline

Posts: 12271


WWW
« Reply #6 on: April 10, 2022, 02:43:27 PM »


Found the FAQ but i didnt be able to use this CSV in ECuXplot , when i load it in to it i cant choose the different Variables.

I'm not having any issues loading any of your CSVs in ECUxPlot, they look fine
Logged

ME7.1 tuning guide
ECUx Plot
ME7Sum checksum
Trim heatmap tool

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your ex
Burnman6666
Jr. Member
**

Karma: +1/-1
Offline Offline

Posts: 33


« Reply #7 on: April 10, 2022, 02:49:10 PM »

Your IMAX is doing some very strange things, which is causing massive integrator windup and boost overshoot.

Do you have a Idea why? When i run Stock Boost or a bit more it seems like there is no Problem even above 4000 rpm.
Got a Pierburg Opel( GM) OPC N75 may there is a Problem?  Huh
Logged
Burnman6666
Jr. Member
**

Karma: +1/-1
Offline Offline

Posts: 33


« Reply #8 on: April 11, 2022, 10:05:40 AM »

Thinking about the Problem.
Wehen i understand it right its like a ping pong, the regulation is oversteering.
In the case its not the Software i would do/check following things:
-Check the Wastegate ( movement and pressure)
-Change the hoses of the n75 to the Wastegate and pressure to Valve because diameter is really big (may to big, thinner hose may soften the regulation a little bit)
-May Change back to original VW N75 instead Pierburg OPC

Or do you mean it has to be 100% a Software Problem?


Greetings


 
Logged
nyet
Administrator
Hero Member
*****

Karma: +608/-168
Offline Offline

Posts: 12271


WWW
« Reply #9 on: April 11, 2022, 11:54:49 AM »

Yes. Look at your ldimx_w. It is way too high, and you are seeing significant I windup. You're over thinking things.

You can make your PID less sensitive to integrator windup, but that is a PhD topic probably out of scope here Smiley

prj doesn't like it, but I get around it by just severely limiting imax and making it a feed forward parameter. You can also hack around it by dorking with the wg linearity map, or using prj's total feed forward method from start to finish.
« Last Edit: April 11, 2022, 11:57:21 AM by nyet » Logged

ME7.1 tuning guide
ECUx Plot
ME7Sum checksum
Trim heatmap tool

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your ex
Burnman6666
Jr. Member
**

Karma: +1/-1
Offline Offline

Posts: 33


« Reply #10 on: April 11, 2022, 03:22:46 PM »

So i have to modify KFLDIMX as the simplest solution?
How would you modify it? How do i limit it?

X is differrence between Acutal Boost and desired?
Y is RPM

Thanks and Greetings
Logged
nyet
Administrator
Hero Member
*****

Karma: +608/-168
Offline Offline

Posts: 12271


WWW
« Reply #11 on: April 12, 2022, 02:50:03 AM »

So i have to modify KFLDIMX as the simplest solution?
How would you modify it? How do i limit it?
It should be more or less what you want your WGDC to be at any giving operating point *before* KFLDRL correction.

Quote
X is differrence between Acutal Boost and desired?
Y is RPM

Consult the FR. plsolr is most definitely not lde.
Logged

ME7.1 tuning guide
ECUx Plot
ME7Sum checksum
Trim heatmap tool

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your ex
Burnman6666
Jr. Member
**

Karma: +1/-1
Offline Offline

Posts: 33


« Reply #12 on: April 12, 2022, 03:56:46 PM »

Okay thank you. Will try to understand the Text in the FR.
Even its in German i find it really difficult to understand.

But i will give my best.

Greetings from Germany
Logged
Burnman6666
Jr. Member
**

Karma: +1/-1
Offline Offline

Posts: 33


« Reply #13 on: April 28, 2022, 04:12:17 PM »

Hey , still doesnt fix the problem tried many changes but seems they havenĀ“t the effect i want.
Is there a possibility to set the WG DC Fix ?

So i can check the Turbo working right and see its capabilities.
Logged
Norwegian1.8T
Full Member
***

Karma: +1/-1
Offline Offline

Posts: 72


« Reply #14 on: April 29, 2022, 02:16:52 AM »

Your WG isnt the problem, boost is reacting propperly according to N75 DC. You problem is as Nyet said huge windup.
After ive been trying to alter the stock PID controll to a non-oem turbo many times and they've all been shit ive just gone the feed forward route and it now works perfectly. The by far easiest way to get PID to work propperly is to use the feed forward mathod that Nyet mentioned, see link: http://nefariousmotorsports.com/forum/index.php?topic=12352.0title=

There is a tool made for this that generates DRL data for you, the tool is in the post just a few pages further back.

Hope this helps Smiley
Logged
Pages: [1] 2
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.02 seconds with 17 queries. (Pretty URLs adds 0s, 0q)