Help with PID tuning (logs inside)

mj6234

Corporal
Nov 25, 2020
162
59
0
carabuser knows a lot more than me, but this is my take.

I 100% agree using the INA0S PID tables without change is the right approach. I did the same, and it wasnt too hard to get the boost on target. IMHO, messing too much with the PID tables is asking to make this way harder than it should be.

Per V8Bait's guide, having PID upwardly adjusting WGDC is far more preferable than having PID downwardly adjusting it. The latter can lead to overboot, throttle closures, and other DME interventions to try and dial things back. I have had the best luck trying to keep WGDC base 3% or so below WGDC after PID. So in the cells that WGDC is over WGDC after PID (or too close), need to pull back base. If you can get base in the ballpark, the INA0S PID tables will do the rest.

Adder does matter too, as well as some other tables in the boost control section. For adder, what worked best for me was a kind of general pattern where adder increases as MAF increases. You will find that to hit the same boost target at higher MAFs takes more WGDC, and doing it all in the base table doesnt really work great. It is a global adder to WGDC based on MAF.

I look at requested MAF + boost setpoint to find the cell to adjust based on the log. For example, in the below you can see you need to remove some base.

You can get a 'flat' boost target, but there are a lot of different tables that influence the shape of the curve. I wouldnt get fixated on an absolute boost value as that pretty much discards the load logic built into the DME (which strives to keep engine output consistent under different conditions). 18psi at 0*F isnt the same as 18psi at 100*F, and IAT isnt the only factor that affects target.

1653154341972.png
 
  • Like
Reactions: Sava666 and wheela

Sava666

Corporal
Dec 14, 2020
170
36
0
So I plan to stick with INA0S PID. Raised Boost Ceiling by 10%. Lowered some WGDC Base areas. And now I have this. Still much of Load Actual 3400 and 4000 rpms. And now I have this which is not bad. car feels great. But targeting 21ish and I still want 18ish.

BTW I have Stage1 CHRAs. 9 blade turbine/11GTX type compressor + inlets.
I believe I still have room to make it flat 18 till 6500 )))



Attached is this bin.
 

Attachments

  • 18psi_v1.1_city_PIDinaos2.bin
    2 MB · Views: 22

Sava666

Corporal
Dec 14, 2020
170
36
0
For example, in the below you can see you need to remove some base.
As your quote about 3 points below PID, you mean that I need to see Base of 27ish against 31 PID on this point, and so on for example?
My initial question was can we get PID higher? > to eliminate overlaping of PID vs Base?

Can you look through my bin please?
 

mj6234

Corporal
Nov 25, 2020
162
59
0
So I plan to stick with INA0S PID. Raised Boost Ceiling by 10%. Lowered some WGDC Base areas. And now I have this. Still much of Load Actual 3400 and 4000 rpms. And now I have this which is not bad. car feels great. But targeting 21ish and I still want 18ish.

BTW I have Stage1 CHRAs. 9 blade turbine/11GTX type compressor + inlets.
I believe I still have room to make it flat 18 till 6500 )))



Attached is this bin.

If you want to target 18psi, you need to lower requested load some where the boost target is too high. If you lower load target, it should hit a different setpoint, and call for less WGDC.

Either way I would lower base 1.5% at this data point and try again. I always blend the cells around the cell I am changing as well so no wild changes. It took me a lot of tries to get my base table right, but this is the process I followed.

1653170037764.png
 

Attachments

  • 1653169978589.png
    1653169978589.png
    64.4 KB · Views: 9
  • Like
Reactions: Sava666

Sava666

Corporal
Dec 14, 2020
170
36
0
One more thing is BLM doing nothing. I made it 2.8 all the way and nothing is changing at all. BLM ceiling is 3. It is Load limiting factor but anyway I see my Load Actual is crawling pretty close to Req. I see 178 vs 179 (previous tries).
 
Last edited:

Sava666

Corporal
Dec 14, 2020
170
36
0
Either way I would lower base 1.5% at this data point and try again. I always blend the cells around the cell I am changing as well so no wild changes. It took me a lot of tries to get my base table right, but this is the process I followed.
Exactly what I plan to do. Thanks for hints. I see I'm no the right way.

So the reason I'm keeping my Load Target high is Carabuser told me that I should raise it to keep it away and setting max boost by ceiling and this is what i'm trying to do, and adjust things. For me it's quite hard to understand mystical explanations of Carabuser. I'll try to pull down Load target.
 

Maroon_e90

Specialist
Sep 30, 2021
59
17
0
24
Northwest Indiana
Ride
2009 335i - xdrive 6MT
I believe I use stock P and I and cut D in half then the left in half again. This was a great starting point from the Tuning spreadsheet.

After that I don't really touch PID as it works as intended and you really need to control your boost with the WGDC base or commanded WGDC. If you look at your PID WGDC and WGDC base are really far off on spool. I would start there and then from 3600 -4000 rpm you need to adjust the base down as your car is trying to prevent overshoot.
 
  • Like
Reactions: Sava666

Sava666

Corporal
Dec 14, 2020
170
36
0
I believe I use stock P and I and cut D in half then the left in half again. This was a great starting point from the Tuning spreadsheet.

After that I don't really touch PID as it works as intended and you really need to control your boost with the WGDC base or commanded WGDC. If you look at your PID WGDC and WGDC base are really far off on spool. I would start there and then from 3600 -4000 rpm you need to adjust the base down as your car is trying to prevent overshoot.
Great info. Thanks.
 

Sava666

Corporal
Dec 14, 2020
170
36
0
Guys, good night! Please look again. I see good progress compared to previous ones. Apart from the fine-tuning of WGDC Base, what else is there to add? At least I was able to get rid of TCs. I made 3 logs, they look the same.

Can someone explain if we have "Boost Mean", why do we need to log "Boost"? Why we need to log calculated intake boost?

Regards!

 

Sava666

Corporal
Dec 14, 2020
170
36
0
4th revision

 

Maroon_e90

Specialist
Sep 30, 2021
59
17
0
24
Northwest Indiana
Ride
2009 335i - xdrive 6MT
Getting closer, I would increase WGDC base from 3000-3400 RPM alot then lower wgdc base at 3600 a smidge (adjust breakpoints if need be). You also need more WGDC from 4500+ since you have made PID less reactive it won't add as much WGDC. Your 3D fuel scaler needs some small adjustment I try to keep this at +/- 3-5%. What fuel are you running? And you should definitly be logging every cyl for timing corrections. You don't need to be logging boost error anymore.
 
  • Like
Reactions: Sava666

Sava666

Corporal
Dec 14, 2020
170
36
0
Getting closer, I would increase WGDC base from 3000-3400 RPM alot then lower wgdc base at 3600 a smidge (adjust breakpoints if need be). You also need more WGDC from 4500+ since you have made PID less reactive it won't add as much WGDC. Your 3D fuel scaler needs some small adjustment I try to keep this at +/- 3-5%. What fuel are you running? And you should definitly be logging every cyl for timing corrections. You don't need to be logging boost error anymore.
Hello, Maroon. So rn I did multigear and things went crazy again. I've adjusted lil bit my scalars.

 

Sava666

Corporal
Dec 14, 2020
170
36
0
Getting closer, I would increase WGDC base from 3000-3400 RPM alot then lower wgdc base at 3600 a smidge (adjust breakpoints if need be). You also need more WGDC from 4500+ since you have made PID less reactive it won't add as much WGDC. Your 3D fuel scaler needs some small adjustment I try to keep this at +/- 3-5%. What fuel are you running? And you should definitly be logging every cyl for timing corrections. You don't need to be logging boost error anymore.
2nd run. How can you comment on sudden boost fall down after 5500? Is there any room?

 

Maroon_e90

Specialist
Sep 30, 2021
59
17
0
24
Northwest Indiana
Ride
2009 335i - xdrive 6MT
When you push into 4th gear you cause the car to make more boost due to being at lower RPM's for longer. There is a bit of a balancing game when adjusting the WGDC to not over boost and cause throttle closure. The easiest way is to decrease WGDC and you will not make as much boost in lower gears but you will not have throttle closure in higher gears. You can manage boost by gear, or you can set boost target higher and understand that in higher gears you will make more boost.

As for boost after 5500 thats just where stockers start to decline. You can definitly increase WGDC up there but you will get diminshing returns and probably wanna get inlets soon.
 

Sava666

Corporal
Dec 14, 2020
170
36
0
When you push into 4th gear you cause the car to make more boost due to being at lower RPM's for longer. There is a bit of a balancing game when adjusting the WGDC to not over boost and cause throttle closure. The easiest way is to decrease WGDC and you will not make as much boost in lower gears but you will not have throttle closure in higher gears. You can manage boost by gear, or you can set boost target higher and understand that in higher gears you will make more boost.

As for boost after 5500 thats just where stockers start to decline. You can definitly increase WGDC up there but you will get diminshing returns and probably wanna get inlets soon.
Already have inlets. Stage1 CHRAs. That's why I'm looking for limits, but we'll see. Will try your hints and report tomorrow. Thanks.
 

Maroon_e90

Specialist
Sep 30, 2021
59
17
0
24
Northwest Indiana
Ride
2009 335i - xdrive 6MT
I didn't know that, you should be able to push to 21-22psi then at redline with 85% WGDC. Try 70% or so after 5500 RPM. That should get you 2-3 more PSI up there
 

Sava666

Corporal
Dec 14, 2020
170
36
0
I didn't know that, you should be able to push to 21-22psi then at redline with 85% WGDC. Try 70% or so after 5500 RPM. That should get you 2-3 more PSI up there
So strange, like I hit some limiter after 5500 and it can push anymore, but bank 1 is around 60%. Can you look into my bin maybe you will find anything strange?
 

Attachments

  • PIDinaos_v5.bin
    2 MB · Views: 22
Last edited:

Sava666

Corporal
Dec 14, 2020
170
36
0
What can you add guys? Guess it is the limit after 5500rpm? Why it's falling down so rapidly?
Maybe any solution to supply more air? Vanos adjustments etc?

 

Maroon_e90

Specialist
Sep 30, 2021
59
17
0
24
Northwest Indiana
Ride
2009 335i - xdrive 6MT
Checked your bin out and you are actually running I8A0S. Also Im not seeing any nannies or problems. The only issue I see is that you need to increase WGDC base if you want to hit those boost levels. The y-Axis is boost setpoint and the x-Axis is MAF g/s. So you need to edit the map where your boost setpoint is and your MAF is AKA 5500+ RPM. I have high lighted where you should start, remember the adders also affect total WGDC base so make sure to take that into account when changing these numbers.

WGDC base.png
 
  • Like
Reactions: tej98 and Sava666

Sava666

Corporal
Dec 14, 2020
170
36
0
Checked your bin out and you are actually running I8A0S. Also Im not seeing any nannies or problems. The only issue I see is that you need to increase WGDC base if you want to hit those boost levels. The y-Axis is boost setpoint and the x-Axis is MAF g/s. So you need to edit the map where your boost setpoint is and your MAF is AKA 5500+ RPM. I have high lighted where you should start, remember the adders also affect total WGDC base so make sure to take that into account when changing these numbers.

View attachment 68839
So, just had a chat with Silent11, who has the fastest stockers n54 and he shared some wisdom with me to put more attention till 5500rpms and forget about 6500. So it be. Thanks for your help mate!