PID is Adding WGDC While Overboosting

impuls

Specialist
Jan 28, 2018
92
60
0
Ride
'07 335i
i took a quick look and this might be caused by too much wgdc adder....
seems like the car is still adding boost by pid cause you havent reached your load target yet but a second or so later it realizes it added too much so throttle closure happens

What puzzles me is the fact that the same tune worked fine before. I also have never seen WGDC being added while throttle is being closed.

I also have a log where WGDC After PID is above Base for 4 seconds while oberboosting with throttle closure. Dosn't seem to be a delayed adjustment.

Hope roads are dry enough tomorrow for a pull during commute to work... temperature is back at the level they were when the tune worked fine.
 

impuls

Specialist
Jan 28, 2018
92
60
0
Ride
'07 335i
Lower temperature didn't fix the problem.
Kind of hesitant to adjust the tune that used to work close to perfect...
Last desperate attempt: Unplug battery... maybe that deletes that WGDC adaption that probably only exist in my head...
 

AzNdevil

Lieutenant
Staff member
Nov 4, 2016
566
269
0
Hong Kong
What puzzles me is the fact that the same tune worked fine before. I also have never seen WGDC being added while throttle is being closed.

I also have a log where WGDC After PID is above Base for 4 seconds while oberboosting with throttle closure. Dosn't seem to be a delayed adjustment.

Hope roads are dry enough tomorrow for a pull during commute to work... temperature is back at the level they were when the tune worked fine.

like i said, you are a bit off your load target when you reached your boost target so maybe thats why the car kept on boosting
also from what ive seen, boost target does change slightly (1-3 psi) for me even on the same stretch of road on different days... thats why i said it looks normal to me...

as @P33P33 mentioned, there are lots of parameters that affect how boost target is calculated..so it depends if you want to open another can of worms
 

impuls

Specialist
Jan 28, 2018
92
60
0
Ride
'07 335i
like i said, you are a bit off your load target when you reached your boost target so maybe thats why the car kept on boosting

According to the control diagram load should not matter here, but maybe the diagram or my understanding is incomplete.

also from what ive seen, boost target does change slightly (1-3 psi) for me even on the same stretch of road on different days... thats why i said it looks normal to me...

Limiting load based on IAT seems to keep that under control quite well for me.

as @P33P33 mentioned, there are lots of parameters that affect how boost target is calculated..so it depends if you want to open another can of worms

Actually I am interested in opening that can of worms. Just for the sake of understanding how boost target is calculated, not expecting that this will help to create better tunes.
 

AzNdevil

Lieutenant
Staff member
Nov 4, 2016
566
269
0
Hong Kong
According to the control diagram load should not matter here, but maybe the diagram or my understanding is incomplete.

Limiting load based on IAT seems to keep that under control quite well for me.

Actually I am interested in opening that can of worms. Just for the sake of understanding how boost target is calculated, not expecting that this will help to create better tunes.

my understanding is that the load target is the master governor for boost...thats why we are able to go above 20psi after scaling
but the way the dme calculates boost is based on lots of factors as you can see from the diagrams so 2-3psi off not that surprising to me

not sure what you mean by limiting load based on IAT...
 

impuls

Specialist
Jan 28, 2018
92
60
0
Ride
'07 335i
my understanding is that the load target is the master governor for boost..
Agree, as far as I understand load basically how much air mass you want to get into the cylinder.
Depending on VE (based on a lot of factors), IAT, ambient pressure that will require more or less boost.

thats why we are able to go above 20psi after scaling
Are you referring to scaling with the Map Conversion / N20 TMAP?
The reason this works is the fact that boost is not reported correctly to the DME.

not sure what you mean by limiting load based on IAT...
The boost target will increase with higher IATs to compensate for the lower air density.
I am using Load Target Limit Factor (Charge Air Temp) to reduce the Load Target and keep the boost target constant independently of IAT.
 

AzNdevil

Lieutenant
Staff member
Nov 4, 2016
566
269
0
Hong Kong
Agree, as far as I understand load basically how much air mass you want to get into the cylinder.
Depending on VE (based on a lot of factors), IAT, ambient pressure that will require more or less boost.

Are you referring to scaling with the Map Conversion / N20 TMAP?
The reason this works is the fact that boost is not reported correctly to the DME.

agree for the second part but the way i look at it is, the load target is treated as the max limiter to what boost pressure the car can see
this limiter is factored based on VE, IAT, and all that stuff (what you just said)

correct me if i am wrong but its the load target that is scaled since other parts of the map need to be rescaled


The boost target will increase with higher IATs to compensate for the lower air density.
I am using Load Target Limit Factor (Charge Air Temp) to reduce the Load Target and keep the boost target constant independently of IAT.

i must say... i never noticed the existence of this table until now.............thanks :)
 

impuls

Specialist
Jan 28, 2018
92
60
0
Ride
'07 335i
agree for the second part but the way i look at it is, the load target is treated as the max limiter to what boost pressure the car can see
this limiter is factored based on VE, IAT, and all that stuff (what you just said)

You are right it's a limiter and it can be viewed as limiter of the boost target, as that is what the DME is controlling based on the load besides throttle position.

Technically the DME is trying to limit the amount of air in the cylinder. The German names of the load tables are maximum filling/stuffing (literal translation) with air relative to normalized mass of cylinder volume to reach requested torque (that's what I read between the lines).

correct me if i am wrong but its the load target that is scaled since other parts of the map need to be rescaled

We are scaling how the DME translates TMAP voltage to boost pressure. Load is just scaled indirectly (and inaccurately due to the other involved variables).
 

AzNdevil

Lieutenant
Staff member
Nov 4, 2016
566
269
0
Hong Kong
You are right it's a limiter and it can be viewed as limiter of the boost target, as that is what the DME is controlling based on the load besides throttle position.

Technically the DME is trying to limit the amount of air in the cylinder. The German names of the load tables are maximum filling/stuffing (literal translation) with air relative to normalized mass of cylinder volume to reach requested torque (that's what I read between the lines).

We are scaling how the DME translates TMAP voltage to boost pressure. Load is just scaled indirectly (and inaccurately due to the other involved variables).

yep i completely agree with this which comes back to the point why i think your car is still adding boost past boost target as the load hasnt reached the target

but at the same time, this behavior can also be a byproduct of scaling throwing off the calculations.....
 

impuls

Specialist
Jan 28, 2018
92
60
0
Ride
'07 335i
yep i completely agree with this which comes back to the point why i think your car is still adding boost past boost target as the load hasnt reached the target

I don't see Rf* being used in full throttle mode in the diagram. I only see Pld* factors going into the PID. Where do you think load is factored into WGDC control besides calculation of boost target?

but at the same time, this behavior can also be a byproduct of scaling throwing off the calculations.....

Actually my tune is not scaled, I am using the MBoost option to target ~21 PSI.
 

impuls

Specialist
Jan 28, 2018
92
60
0
Ride
'07 335i
Just wanted share latest findings on this issue:

There is a diagram describing how the factor applied after WGDC Base before PID is calculated (based on I-Factor) and persisted. That means if you are asking for too much boost in your tune you can end up with the DME adjusting your WGDC PID up permanently.

In the log below PID tables are zeroed and the amount of WGDC added is matching the 1.2 factor I have seen in INPA pretty well.

Triggering a boost error should clear the factor. Actually I was hoping for underboost without PID...

Driving long enough with a proper tune should also bring the factor down, but recalculation will only happen if a lot of preconditions (I don't understand) are met.

 
  • Informative
Reactions: AzNdevil

impuls

Specialist
Jan 28, 2018
92
60
0
Ride
'07 335i
In case somebody is trying to trigger an under boost code to clear this adpation...
Be smarter than me and make sure the 30FE code is not disabled (it seems it is in most public tunes).
 

impuls

Specialist
Jan 28, 2018
92
60
0
Ride
'07 335i
Underboost doesn't reset the adaption... not sure what does, but I will just tune accounting for the 20% added to WGDC Base now.
 
  • Informative
Reactions: houtan