NefMoto

Overboosting, not following Desired Boost

Hi Nefmoto,

I've changed LAMFA, LDRXN as shown in attachments.
Running a full throttle run in 2nd gear, giving 100% pedal at once.
It looks like it's having a huge boost peak, and then tempers down below desired.

During the pull you can feel a * hik * which was not at the default map.

Hope someone can help me out.

Also a image with wastegate

2nd gear is dynamic.  Use 3rd.  Also post CSV.

Below the CSV file, i'll hope something wrong can be found while pulling in 2nd gear.
Tomorrow i will log again in 3th

Too much IMX.

You didn't log any of the PID variables, that would help.

Thanks Nyet, will checkout the IMX ( KFLDIMX right )

Will log boost PID's tomorrow thx!

Now having 2 spikes.

hard to tell w/o pid variables.

Pull more IMX

Okay, Thanks for reply nyet.
Currently having the following logs.

I've changed IMX and i think its getting close?
Should i finetune IMX more?

Please log the PID variables.

Yes, pull more IMX

BoostPID-I-Result","BoostPID-P-Result","BoostPID-D-Result
Those above are not the PID's you need?

Logged again, and think i'm getting close.
Should i pull more IMX at 3500RPM?

Log file contains all PID values i could find

Less KFLDIMX, and likely less Q2

Also, instead of an increasing KFLDIMX, i'd consider a flat KFLDIMX and tweaking KFLDRL to compensate

Finally, you're maxed out on your turbos up top, id reduce redline boost request. You're out of injector anyway.

Ok, so have changed IMX with flat line, and lowere LDRXN.
Having the following logs now.

Should i change IMX more?

You brought it down too low, and you didn't compensate via KFLDLR

Also, log pedal position and throttle plate angle or you won't be able to use ecuxplot properly.

Ok, so few days ago i bricked my ecu flashing non checksummed bin.
Shorted out the ECU while hanging on bench, great O_O

Bought a spare ECU, and immo off + flashed original bin.
Finally can continue fixing the overboosting.

Currently i'm having this in logs, i don't really understand what to change to compensate spikes.

having what , there is no pic or log attached.

Sorry, here are the logs :)

Try pulling a bit of DIMX

Can i apply this to gain better control?
http://nefariousmotorsports.com/forum/index.php?topic=12352.0;topicseen

Absolutely. Just takes a bit more work to set up.

Also I really think you should flatten DIMX quite a bit. You have falling boost request anyway.

What happens if i replace the imx table with the one on the post.
Is there chance i blowup my engine? Haha  ???

Lol, you will definitly not blow up the engine :D the ecu has many foolproof methods programmed to prevent this unless ofcourse you disable it like the Boeing guys did :D

That is literally the last thing you want to do. Not because you'll blow up your engine, but because its pointless unless you do all of the work described in the thread.

You mean flatten something like this? sorry my english is not that well.

no, flatten so it doesn't increase with RPM at all

Ok, have flatten IMX now, check image attachment.
Should i not compensate with KFLDRL?

Having the following logs now

Much better. but keep extend the flat imx down to 2500rpm or so.

Also, there is something very wrong with your DRL. it should not spike to 95 at input of 65, that is what is causing your oscillation.

Can you log wgdc prelin?

Thats ldtvr_w right?

Changed IMX and logged with requested variable

MUCH better.

I think you just need to remove more of DIMX, add Q2 back at around 2500-2750, and bring up LDRL a bit from 4000 up and you should be close!

Alternately, you could probably add some boost request <3500 RPM, you have decent enough fueling and timing isn't horrible.

IMX is a bit lower, Q2 added a bit and DRL up a bit.

Sorry i think you mean i should lower q2

Hard to say, the initial overshoot is caused by insufficient Q2 (D).

So have changed some graphs, IMX is now at 51 flat ( isn't this to low?
Peak boost is gone, still shows underboost spikes.

yea, you were right, pull Q2. Not sure why it is so crazy after boost onset but you dont have any on ramp.

I changed IMX to 55 flat, car was not starting on 51 flat.
Small overboost is not a problem right(peak)?

PS could ot be machenical?

No I don't think it is mechanical

I dont get it anymore, i've flatten out 56 imx from 2500.
Engine response is slow, when i let the engine idle for 2 min~ it stalls or not responding to gas. Orginal no problems.

you problem not imx or q2
never need very low KFLDRL or KFLDIMX auq engine

and plz write actual modification the car and exhaust ?
removed original downpipe ?

and faster upload actual file

Hi, no original downpipe, 200cells racekat now

So, everything stock, only 200 cells cat

ok

plz upload file

Here you go, hope you find something.

Disagree. If the wg is set very hard, it is needed.

He could also try less WG pressure.

is it stock wg spring? if yes then you had a mechanical problem. check n75, lines and wg
also ldrxnzk def is wrong. you changed map axis size right after it @1F2AA

I think its stock.

I don't see anything mechanically wrong with the WG or the n75.

The PID just isn't tuned right for this set up.

Any details for what table data i should change?

1st of all get back to stock 0x1F2AA

i sent for you file
check your pm

Ok, lowered LDRXN at the point it was reaching maximum wastegate dutycycle.
How is this looking ^^?

Better but you are out of injector up top and you have too much DIMX still.

Thanks Nyet, could you please explain why the DIMX needs to be lower?
To take away the overshoot? And also the higher actual than desired?

Yes.

Ok, thanks for the info, looks like the injector duty time goes to 100 because of the suddenly chaning AFR to 0.75 right?

Yes, but you should have enough overhead that BTS can do what it wants safely

that said, you should probably figure out what is sending your req into BTS territory.

So, TABGBTS have set this one to 750, currently its not reaching this limit.
Does interventions on timing also trigger BTS?

if the problem with WG steering is not solved , please share Your Q0 Q1 Q2 maps, some 1.8t , like AUQ have some crazy PID maps... very hard to control on higher boost levels.

Major modifications to Q0 Q1 Q2 are needed.. especially if YOu mounted bigger turbo.