Pages: [1]
Author Topic: Simos 18.1 - 6mt, throttle stays opened after driver tq req goes to 0  (Read 4290 times)
jamespinger
Jr. Member
**

Karma: +1/-0
Offline Offline

Posts: 35



To clarify the title:
When driving in gear, typically at high RPM/high torque, if the gas pedal is released and the clutch is *not* depressed, the throttle stays opened for another ~.5 seconds or so.  There's usually an increase of 2-300RPM before the throttle is closed.  It's alarming.  Clutching in/shifting/etc behaves as expected.

I've been looking through the FR, thinking it may have something to do with the transition to lv_pu/lv_puc, but I can't figure it out.  Prior to that I that I thought it might have something to do with the "anti surge damper", but when I disable that, it still happens.

There *is* the following mention on p13528:

Quote
"In the case of rapidly falling load, the transition to (LV_PUC) can be accelerated in order to avoid an engine
speed run-up, which is undesirable. To trigger this function (see the following chapters: ignition, ignition angle
correction for trailing throttle) some load conditions have to be fulfilled."

Attached is a log to demonstrate - approx 18 seconds in I did it in 2nd, and 24.8 seconds in I did it in 3rd.  In both cases you can see the pedal goes to 0 with torque request to 0, and "some time later" the throttle finally closes.  The time between pedal 0 - throttle 0 seems to be related to how much torque there is.

Logged
jamespinger
Jr. Member
**

Karma: +1/-0
Offline Offline

Posts: 35



Bummer,
150+ views and no responses.  There's a ton of people that know so much about this platform and so few are willing to help.
edit: maybe if I add some keywords it'll get more views.... pops bangs flame gurgle burble popcorn.  :-/

For what it's worth, rev hang is caused by lv_pu and the cure for that lies in ip_t_min_pu_cs
  • .  The difference being clutch state obviously.

I'm relatively sure the behavior that I originally posted about has something to do with lv_pu.  Airflow targets drop quite gradually, if only I could figure out what determines that slope.
« Last Edit: May 10, 2021, 05:54:16 AM by jamespinger » Logged
michelhadid
Jr. Member
**

Karma: +9/-2
Offline Offline

Posts: 38



Bummer,
150+ views and no responses.  There's a ton of people that know so much about this platform and so few are willing to help.
edit: maybe if I add some keywords it'll get more views.... pops bangs flame gurgle burble popcorn.  :-/

For what it's worth, rev hang is caused by lv_pu and the cure for that lies in ip_t_min_pu_cs
  • .  The difference being clutch state obviously.

I'm relatively sure the behavior that I originally posted about has something to do with lv_pu.  Airflow targets drop quite gradually, if only I could figure out what determines that slope.

Did you ever find the cause?
Logged
jamespinger
Jr. Member
**

Karma: +1/-0
Offline Offline

Posts: 35



Nope, and it drives me crazy.

Sent from my Pixel 4 XL using Tapatalk

Logged
jamespinger
Jr. Member
**

Karma: +1/-0
Offline Offline

Posts: 35



Anything from someone like @littco or @k0mpressed  that has tuned thousands of these?

Sent from my Pixel 4 XL using Tapatalk

Logged
prj
Hero Member
*****

Karma: +915/-427
Offline Offline

Posts: 5840



You have full ram data logging, and you have the FR.
Why don't you just log what is happening and trace through the FR?
Logged

PM's will not be answered, so don't even try.
Log your car properly.
jamespinger
Jr. Member
**

Karma: +1/-0
Offline Offline

Posts: 35



You have full ram data logging, and you have the FR.
Why don't you just log what is happening and trace through the FR?

Yes yes, I do have full ram datalogging and the FR.  And I've read through what I believe to be the relevant sections for many hours over many nights.
I'm told the GTI doesn't have the same behavior (by someone with a DSG).  So I've spent lots of time comparing a GTI cal to my own.  But I'm not sure if a manual GTI with the same CAL would do it, so maybe there's no difference.

There's most definitely something that I'm missing and I'm hoping that someone with some experience has some insight.  I believe I said this at some point in the past on the same forum but a different topic - I'm not looking for a handout I'm genuinely asking for help...

I know most of the people on this forum are professional tuners, but I do believe the original purpose of NefMoto was to open source and share information, no? 

It seems that it switches to lv_pu almost immediately after letting off the pedal. 
Torque request drops immediately
put_sp decreases slowly
Throttle stays wide open (until it gets to a threshold).

Logged
prj
Hero Member
*****

Karma: +915/-427
Offline Offline

Posts: 5840



Your first mistake is going off of hearsay instead of following the OEM documents.
Your second mistake is assuming that anyone of so called "pro tuners" has any clue about this. No they don't, all most of them know is how to % the torque limiters, with a few exceptions.

I seriously doubt anyone has ever looked into tweaking PU -> PUC transitions.

Instead of comparing calibrations and doing god knows what else, log the exact parameters.
Most likely you don't like the delay between normal state and PUC, this is called PU.

You can try changing to zero:
id_t_min_pu
ip_t_min_pu_cs[0]
ip_t_min_pu_cs[1]

And then see if anything changes for you. If not then I don't have any answer off the top of my head either, go into FR and log the whole diagram paths to understand what is going on.
Or if you are not capable of doing that, then just live with it. Nobody is going to spend the time to do it for you, and you assumption that everyone knows how it's done is utterly misplaced.

During tuning you almost never change such behavior of the car, because that is a recipe for disaster, as many customers are going to complain.

Also throttle is completely irrelevant. You can have the throttle as wide open as you like if the cylinders are deactivated.
Logged

PM's will not be answered, so don't even try.
Log your car properly.
jamespinger
Jr. Member
**

Karma: +1/-0
Offline Offline

Posts: 35


« Reply #8 on: August 02, 2021, 10:56:16 AM »

Your first mistake is going off of hearsay instead of following the OEM documents.
Your second mistake is assuming that anyone of so called "pro tuners" has any clue about this. No they don't, all most of them know is how to % the torque limiters, with a few exceptions.

I seriously doubt anyone has ever looked into tweaking PU -> PUC transitions.

Instead of comparing calibrations and doing god knows what else, log the exact parameters.
Most likely you don't like the delay between normal state and PUC, this is called PU.

You can try changing to zero:
id_t_min_pu
ip_t_min_pu_cs[0]
ip_t_min_pu_cs[1]


Ah, so I had ASSUMED the people here had some general hacking interest, not just adjusting torque targets.  You're suggesting that most of them don't.  Noted.

And - if I get rid of lv_pu entirely, what EVER will I do about all the flame and bang?  That's also a problem. (note the sarcasm)

Those tables you mentioned are what help with/control rev hang (the cs ones are what force the car in lv_pu for a set period of time while you're shifting).  Already 0'd those out and it makes a world of difference during *clutch in* conditions.

I do think forcing the car through lv_pu into lv_puc quicker is a fix, but I don't know if it's *the* fix.  I haven't tried it myself but 2 friends did and they said the tach does some weird things.  They set c_t_max_pu to very very low values - think .1.  Here, the throttle does close immediately (.08 seconds) after letting off the pedal.  Screenshot of the RPMs attached. 

FWIW, the throttle *DOES* close in lv_pu "long" before the car switches into lv_puc.  So there's some airflow behavior while *in* lv_pu that's causing it.  I *do* know/acknowledge what you're saying about deactivating the cylinders.  So I know the throttle isn't a good indicator.  But in the log I originally attached it stays WOT when my pedal goes to 0.  The only thing the car actually does when I let off the pedal is open the DV to start reducing airmass.

I had thought that the issue might be mechanical.  I have a hybrid turbo with a big compressor wheel using the stock DV.  The DV opens, it just takes a really long time for the put to actually come down.  It doesn't drop as quickly as put_sp does (which doesn't drop that quickly at all anyway).  My one friend has an IS38 and the other an IS20 and their cars exhibit the same behavior it just doesn't seem as pronounced as it is in my car. 

As always, I appreciate your replies @prj.
Logged
prj
Hero Member
*****

Karma: +915/-427
Offline Offline

Posts: 5840


« Reply #9 on: August 03, 2021, 12:19:23 AM »

You're just going to have to go module by module and figure it out.
I do not recall having the problem you are describing.
Logged

PM's will not be answered, so don't even try.
Log your car properly.
jamespinger
Jr. Member
**

Karma: +1/-0
Offline Offline

Posts: 35


« Reply #10 on: August 04, 2021, 12:37:37 PM »

You're just going to have to go module by module and figure it out.
I do not recall having the problem you are describing.

You're probably right.

Another screenshot and log attached if anybody else cares to look.
Here you can see that I let off the pedal right around 26.9 seconds in and I gain another ~250 RPM at WOT before it finally closes at 27.5 seconds.

Screenshot includes both airmass/sp and put/sp. The screenshot also includes lv_pu and lv_puc

Logged
Pages: [1]
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.021 seconds with 17 queries. (Pretty URLs adds 0s, 0q)