Forced Induction Superchargers | Turbochargers | Intercoolers

Is this erratic boost curve caused by PID settings?

Thread Tools
 
Search this Thread
 
Old 09-08-2018, 06:51 PM
  #1  
8 Second Club
Thread Starter
iTrader: (82)
 
Ratical's Avatar
 
Join Date: Nov 2004
Location: Rotterdam, New York
Posts: 998
Received 18 Likes on 12 Posts

Default Is this erratic boost curve caused by PID settings?

I am trying to flatten out my boost curve to stay on target. Boost is all over the place on the target boost line. Is this caused by improper PID settings in the Holley system?

I am running 60psi C02 pressure commanding 15psi on the dome in the dotted line and 20psi on the solid line.



?
Old 09-08-2018, 07:04 PM
  #2  
TECH Addict
iTrader: (1)
 
3 window's Avatar
 
Join Date: Aug 2014
Posts: 2,051
Received 186 Likes on 136 Posts
Default

I’m guessing you’re using the Dome+Map strategy? How well is the dome VS boost table sorted out? The closer is it, the less compensation you’ll need using the PIDs. Right now it’s making large adjustments to the dome pressure in an attempt to keep your target boost on track. Did you set the Dome PIDs using shop air prior to testing? You can get the Dome PIDs extremely close doing this.
Old 09-08-2018, 08:04 PM
  #3  
TECH Regular
iTrader: (7)
 
badass68's Avatar
 
Join Date: Dec 2006
Location: Houston
Posts: 498
Likes: 0
Received 86 Likes on 67 Posts

Default

Yep. Lower pid settings.
Old 09-09-2018, 08:35 AM
  #4  
8 Second Club
Thread Starter
iTrader: (82)
 
Ratical's Avatar
 
Join Date: Nov 2004
Location: Rotterdam, New York
Posts: 998
Received 18 Likes on 12 Posts

Default

Originally Posted by 3 window
I’m guessing you’re using the Dome+Map strategy? How well is the dome VS boost table sorted out? The closer is it, the less compensation you’ll need using the PIDs. Right now it’s making large adjustments to the dome pressure in an attempt to keep your target boost on track. Did you set the Dome PIDs using shop air prior to testing? You can get the Dome PIDs extremely close doing this.
I am using dome control only right now. I did not try to use shop air but I will try that. I am looking up how to do that now. The PID settings are just the base 10/30/0 at the moment. So much reading to do with the Holley system and trying to skim over the parts that don't apply to my situation to find the information I need to know. Every time I look something up it seems I find a new,(different) pdf file I hadn't seen before. It would be GREAT if there was just one manual with a good index.


Old 09-09-2018, 11:49 AM
  #5  
TECH Regular
iTrader: (7)
 
badass68's Avatar
 
Join Date: Dec 2006
Location: Houston
Posts: 498
Likes: 0
Received 86 Likes on 67 Posts

Default

Usually 10/30/0 will work. Sounds like your dome map isn’t populated property.
Old 09-09-2018, 01:01 PM
  #6  
TECH Addict
iTrader: (1)
 
3 window's Avatar
 
Join Date: Aug 2014
Posts: 2,051
Received 186 Likes on 136 Posts
Default

Shop air testing is easy. Use your shop air to a regulator to the fill solenoid. Set the regulator to whatever pressure you run your CO2 or onboard air to (say 60psi). Now set the strategy to boost VS time and and make the dome pressure table 30psi dome pressure (just for testing sake)Make your transbrake/launch 10 (again, just to test). Now as you toggle your transbrake switch on and off, you should see your target as well as your dome pressure changing. Watch how closely the actual dome pressure follows the target and adjust your PIDs from there. As you mess with the PIDs, you’ll see how they each affect the overall results. I believe mine are 7/22/0, but I don’t have my laptop in front of me to verify.
The following users liked this post:
n2xlr8n66 (05-17-2022)
Old 09-10-2018, 08:33 AM
  #7  
8 Second Club
Thread Starter
iTrader: (82)
 
Ratical's Avatar
 
Join Date: Nov 2004
Location: Rotterdam, New York
Posts: 998
Received 18 Likes on 12 Posts

Default

Originally Posted by badass68
Usually 10/30/0 will work. Sounds like your dome map isn’t populated property.
Cool. I will check into that.
Old 09-10-2018, 08:35 AM
  #8  
8 Second Club
Thread Starter
iTrader: (82)
 
Ratical's Avatar
 
Join Date: Nov 2004
Location: Rotterdam, New York
Posts: 998
Received 18 Likes on 12 Posts

Default

Originally Posted by 3 window
Shop air testing is easy. Use your shop air to a regulator to the fill solenoid. Set the regulator to whatever pressure you run your CO2 or onboard air to (say 60psi). Now set the strategy to boost VS time and and make the dome pressure table 30psi dome pressure (just for testing sake)Make your transbrake/launch 10 (again, just to test). Now as you toggle your transbrake switch on and off, you should see your target as well as your dome pressure changing. Watch how closely the actual dome pressure follows the target and adjust your PIDs from there. As you mess with the PIDs, you’ll see how they each affect the overall results. I believe mine are 7/22/0, but I don’t have my laptop in front of me to verify.
Awesome. I will do that this week and try it again Saturday at the Valley to see how much more I can get this E.T whittled down!

Thank you for the tips!
Old 09-12-2018, 08:39 AM
  #9  
TECH Addict
iTrader: (1)
 
3 window's Avatar
 
Join Date: Aug 2014
Posts: 2,051
Received 186 Likes on 136 Posts
Default

Here’s a shot of a boost vs speed log. Yes, I short shifted, but you can see the boost response in relation to the dome pressure changes.


Old 09-12-2018, 10:15 AM
  #10  
8 Second Club
Thread Starter
iTrader: (82)
 
Ratical's Avatar
 
Join Date: Nov 2004
Location: Rotterdam, New York
Posts: 998
Received 18 Likes on 12 Posts

Default

Wow, that definitely looks nothing like mine. I can't wait to see how the car runs when it is seeing all the boost all the way through the run!
Old 09-12-2018, 10:32 AM
  #11  
TECH Addict
iTrader: (1)
 
3 window's Avatar
 
Join Date: Aug 2014
Posts: 2,051
Received 186 Likes on 136 Posts
Default

Are you filtering the datalogs? That’ll straighten out the lines a lot.
Old 09-12-2018, 10:56 AM
  #12  
8 Second Club
Thread Starter
iTrader: (82)
 
Ratical's Avatar
 
Join Date: Nov 2004
Location: Rotterdam, New York
Posts: 998
Received 18 Likes on 12 Posts

Default

Originally Posted by 3 window
Are you filtering the datalogs? That’ll straighten out the lines a lot.
No sir. How do I do that if you don't mind me asking?
Old 09-12-2018, 11:13 AM
  #13  
TECH Addict
iTrader: (1)
 
3 window's Avatar
 
Join Date: Aug 2014
Posts: 2,051
Received 186 Likes on 136 Posts
Default


Put the cursor on the vertical word of what you want to filter (in this case “boost”) and left click.

This screen will pop up then hit the “smoothing” drop down and choose a filtering average. The more filter, the straighter the traces become. Tow things, filtering can become an issue if you’re looking for a problem (sometimes the trace is too smooth) and secondly, if you smooth one trace, do them all or they won’t line up chronologically.
Old 09-12-2018, 11:44 AM
  #14  
8 Second Club
Thread Starter
iTrader: (82)
 
Ratical's Avatar
 
Join Date: Nov 2004
Location: Rotterdam, New York
Posts: 998
Received 18 Likes on 12 Posts

Default

Awesome! Thank you for the help.
Old 09-12-2018, 12:07 PM
  #15  
TECH Addict
iTrader: (1)
 
3 window's Avatar
 
Join Date: Aug 2014
Posts: 2,051
Received 186 Likes on 136 Posts
Default

Any better? Keep us posted.
Old 09-12-2018, 07:04 PM
  #16  
8 Second Club
Thread Starter
iTrader: (82)
 
Ratical's Avatar
 
Join Date: Nov 2004
Location: Rotterdam, New York
Posts: 998
Received 18 Likes on 12 Posts

Default


Originally Posted by 3 window
Any better? Keep us posted.
So I smoothed all of them out to a factor of 11 just to see how it looked and tightened up the range and here is what it shows now.
Old 09-12-2018, 07:58 PM
  #17  
8 Second Club
Thread Starter
iTrader: (82)
 
Ratical's Avatar
 
Join Date: Nov 2004
Location: Rotterdam, New York
Posts: 998
Received 18 Likes on 12 Posts

Default


And here is the same log set at 7 point smoothing.
Old 09-12-2018, 08:04 PM
  #18  
TECH Regular
iTrader: (7)
 
badass68's Avatar
 
Join Date: Dec 2006
Location: Houston
Posts: 498
Likes: 0
Received 86 Likes on 67 Posts

Default

I don’t filter and mine damn near lay on top of each other. I run map control though, which you can after your dome is close. With Humphrey solenoids my dome is 14/30/0.5 and map is 0.7/2/0.
Old 09-12-2018, 08:04 PM
  #19  
TECH Senior Member
iTrader: (25)
 
truckdoug's Avatar
 
Join Date: Nov 2013
Location: Portlandia
Posts: 6,331
Received 527 Likes on 357 Posts

Default

damn I gotta get a holley system
Old 09-12-2018, 08:09 PM
  #20  
TECH Regular
iTrader: (7)
 
badass68's Avatar
 
Join Date: Dec 2006
Location: Houston
Posts: 498
Likes: 0
Received 86 Likes on 67 Posts

Default

This is using boost vs speed 2-3 pull on the street. So every time it spun the boost jumps.


Quick Reply: Is this erratic boost curve caused by PID settings?



All times are GMT -5. The time now is 11:24 PM.