Dome Control improvements needed

Testing and development of Megasquirt 3

Moderators: jsmcortina, muythaibxr

Dome Control improvements needed

Postby nvr2fst » Mon Oct 16, 2017 9:20 am

Hi,

This weekend I was racing the car and it had a lot of issues keeping boost from overshooting with closed loop dome boost control.
I think the issue is from the ram air intake for the turbos at high speeds but that is an assumption.
If you look at dome pressure in the attached log file you see it never lowers dome pressure(actually it keeps rising) even though boost clearly overshot.
This is a similar "slow" correction i see in EGO as well so maybe they have the same issue?

current tune file and log here:
https://www.dropbox.com/sh/08acqkxaz8xe ... Ehbsa?dl=0

I setup the bias table correctly on the dyno but at high speeds when the ram air is forcing more air in it overshoots boost by 2-5lbs depending on the gear. This specifically became an issue on a slick track keeping in mind my car makes 200+more HP with 5lbs of boost. When the boost overshot, it broke traction sending me sideways and off the track at 170+MPH!!! I lived to tell the tale but it wasn't a fun time.
Video here:
https://youtu.be/xHKb7rXaM_g

I like the idea of a reference bias table but I feel like it has too much authority in the final result. Closed loop should react to make minor corrections in a meaningful way.

Check it out and let me know your thoughts.
nvr2fst
Helpful MS/Extra'er
 
Posts: 44
Joined: Sat Dec 17, 2016 7:39 am

Re: Dome Control improvements needed

Postby jsmcortina » Mon Oct 16, 2017 10:03 am

Looking at your log, the dome pressure is being controlled to within say 3kPa (0.5psI). As I understand it, that's in line with competitor systems.

However, even though the dome pressure is pretty constant, your measured MAP value is varying by 12kPa (2psi).

Why the difference? I wonder if this is a mechanical issue.

You also likely want to change your tables, such that when you get out of the throttle you more quickly request a lower boost pressure. Your current tables are keeping you at full boost even when you are releasing the throttle.

James
I can supply, repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: viewtopic.php?f=104&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
jsmcortina
Site Admin
 
Posts: 33917
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK

Re: Dome Control improvements needed

Postby nvr2fst » Mon Oct 16, 2017 6:52 pm

jsmcortina wrote:Looking at your log, the dome pressure is being controlled to within say 3kPa (0.5psI). As I understand it, that's in line with competitor systems.

However, even though the dome pressure is pretty constant, your measured MAP value is varying by 12kPa (2psi).

Why the difference? I wonder if this is a mechanical issue.

You also likely want to change your tables, such that when you get out of the throttle you more quickly request a lower boost pressure. Your current tables are keeping you at full boost even when you are releasing the throttle.

James


Thanks for taking a look James but I am not sure I follow. Take a look at this print screen from the same log file I shared.

boost overshoot.png
boost overshoot.png (151.56 KiB) Viewed 208 times


My point is Map(boost actual) clearly overshoots the Boost target but dome pressure kept rising. It never tried to drop and make a correction. Shouldn't closed loop do something other than continue to rise based on the bias table? Isn't the point of closed loop to make a correction to hit the target when other variables are introduced?
nvr2fst
Helpful MS/Extra'er
 
Posts: 44
Joined: Sat Dec 17, 2016 7:39 am

Re: Dome Control improvements needed

Postby wes kiser » Tue Oct 17, 2017 6:59 am

The dome pressure keeps rising, but understand this initially references your dome bias, then adjusts from there.

Have you adjusted the closed loop boost PID settings to no avail?
86 Rx-7, swapped to 2.3 ford turbo (BW EFR 6758), ms3/ms3x sequential fuel /waste spark, ls2 coils
88 Tbird 2.3t, Microsquirt Module (PIMP), TFI ignition
wes kiser
Super MS/Extra'er
 
Posts: 1318
Joined: Tue Jan 03, 2006 2:49 pm
Location: Charlotte, NC

Re: Dome Control improvements needed

Postby nvr2fst » Wed Oct 18, 2017 11:38 am

wes kiser wrote:The dome pressure keeps rising, but understand this initially references your dome bias, then adjusts from there.

Have you adjusted the closed loop boost PID settings to no avail?


Yep, I understand that. I am pointing out it is missing the "adjust from there" part.

The point I am making is:
1) Boost overshot the entire pass. (look below at 3rd gear log from 110 to 160mph. It overshot the target the entire time but never lowered dome pressure)
2) Dome pressure NEVER dropped based on MAP actual vs MAP Target
3) Even though it overshot the entire pass, it never even stopped ADDING dome pressure because of the bias table. (again, the bias table has too much authority and is not a reference as described. It is simply the target if closed loop does not correct dome pressure relative to boost measured vs. boost target)

I can't look at the above three statements and it tell you the system reached the closed loop boost control design intent.

I hope that doesn't come across as harsh. My apologies if it does. I am just trying to keep it short and sweet to explain my POV.


Look at the selected area for 3rd gear measured values.
Boost dome control.png
Boost dome control.png (221.23 KiB) Viewed 163 times


-Josh
nvr2fst
Helpful MS/Extra'er
 
Posts: 44
Joined: Sat Dec 17, 2016 7:39 am

Re: Dome Control improvements needed

Postby wes kiser » Fri Oct 20, 2017 6:34 am

Your response does not address my question. You have adjusted the closed loop PID values with no effect (or the sensitivity slider if you are using Basic)?

Your screenshots also don't include the Bias table value, so we can't tell if it is modifying them at all.
86 Rx-7, swapped to 2.3 ford turbo (BW EFR 6758), ms3/ms3x sequential fuel /waste spark, ls2 coils
88 Tbird 2.3t, Microsquirt Module (PIMP), TFI ignition
wes kiser
Super MS/Extra'er
 
Posts: 1318
Joined: Tue Jan 03, 2006 2:49 pm
Location: Charlotte, NC

Re: Dome Control improvements needed

Postby nvr2fst » Mon Oct 23, 2017 5:09 am

wes kiser wrote:Your response does not address my question. You have adjusted the closed loop PID values with no effect (or the sensitivity slider if you are using Basic)?

Your screenshots also don't include the Bias table value, so we can't tell if it is modifying them at all.


I have used the sensitivity slider with no improvements BUT this actually worked OK until I added the ram air intakes on the turbos.
My tune file is included in the original dropbox link above. That has the current dome bias table.
nvr2fst
Helpful MS/Extra'er
 
Posts: 44
Joined: Sat Dec 17, 2016 7:39 am


Return to MS3 Development

Who is online

Users browsing this forum: No registered users and 2 guests