Why does turbo boost 'Peak' then level off?
18 May 2001, 12:10 AM
#1
Scooby Senior
Thread Starter
Join Date: Oct 2000
Location: Zurich, Switzerland
Posts: 3,105
Likes: 0
Received 0 Likes
on
0 Posts
I'm no expert (there was excellent info from Moray, Pat and Quattro in a thread I started a year or so ago about the noises you hear with better induction), but;
Boost control is provided by a feedback loop; plant your right foot, boost goes up, the pressure sensor causes the solenoid to close...opening the wastegate...reducing the pressure....pressure sensor (well, ecu) tells solenoid to open, bleeding pressure away from wastegate..raising boost, etc.
The solenoid actually opens/closes at a very high frequency - I guess the duty is varied by the ecu to control boost.
The peak duration time is the time taken for the feedback loop to respond to your go pedal input.
The peak duration (hopefully
) increases every time you modify the induction or exhaust systems - the biggest change for me was the decat downpipe; nearly doubled peak duration in the cold. Why? Because the turbo can spin up quicker, but boost control system still reacts at same speed.
Of course, I've been wrong before
Richard
Boost control is provided by a feedback loop; plant your right foot, boost goes up, the pressure sensor causes the solenoid to close...opening the wastegate...reducing the pressure....pressure sensor (well, ecu) tells solenoid to open, bleeding pressure away from wastegate..raising boost, etc.
The solenoid actually opens/closes at a very high frequency - I guess the duty is varied by the ecu to control boost.
The peak duration time is the time taken for the feedback loop to respond to your go pedal input.
The peak duration (hopefully
) increases every time you modify the induction or exhaust systems - the biggest change for me was the decat downpipe; nearly doubled peak duration in the cold. Why? Because the turbo can spin up quicker, but boost control system still reacts at same speed.Of course, I've been wrong before
Richard
18 May 2001, 12:57 AM
#2
Scooby Regular
Join Date: Aug 1999
Posts: 7,957
Likes: 0
Received 0 Likes
on
0 Posts
common computer control of things such as this relies on a PID controller. This refers to proprtional integral and derivative contyrol of the boost.
Basically the ecu asks for a level of boost and the wastegate opens. The actual level produced then misses to target and creates an error. The PID controller uses the error and calculates how it needs to modify the wastegate control to make the error zero.
A perfectly produced pid control will over shoot its target slightly then undershopot it by an even smaller amount, then rest perfectly on the desired level.
I am not sure if the system here would react fast enough to use pid control, but if it does, it would certainly exhibit these characteristics.
Basically the ecu asks for a level of boost and the wastegate opens. The actual level produced then misses to target and creates an error. The PID controller uses the error and calculates how it needs to modify the wastegate control to make the error zero.
A perfectly produced pid control will over shoot its target slightly then undershopot it by an even smaller amount, then rest perfectly on the desired level.
I am not sure if the system here would react fast enough to use pid control, but if it does, it would certainly exhibit these characteristics.
18 May 2001, 03:02 PM
#4
Scooby Regular
Join Date: Apr 2001
Posts: 41
Likes: 0
Received 0 Likes
on
0 Posts
Ok Adam , Put the Instrument control systems hand book away , either that or you are an instrument tech !! lol
Thought you were going away to start about 4:1 decay ratio's on your PID controller to achieve zero error.
Only kidding.
I am getting a de-cat downpipe fitted tomorrow so hopefully i can put a serious comment on this board on monday.
Regards, Kevin.
Thought you were going away to start about 4:1 decay ratio's on your PID controller to achieve zero error.
Only kidding.
I am getting a de-cat downpipe fitted tomorrow so hopefully i can put a serious comment on this board on monday.
Regards, Kevin.
20 May 2001, 07:50 PM
#5
Scooby Regular
Join Date: Nov 1999
Posts: 449
Likes: 0
Received 0 Likes
on
0 Posts
Adam
How do you know about PID controllers?,I work on a chemical plant and most of the instrument department don't understand this subject LOL
Kevin
What type of kit do you work with?.We have Kent DCS systems as well as electronic TCS's and a load of Foxborough and Taylor pneumatic stuff.The pneumatic controllers are about 30 years old and are more reliable than the modern stuff.They are also more fun to work on as you can actually fix them instead of just fitting a new one.
As for the boost control on the scooby I don't think it runs only closed loop control.
I have measured the duty cycle of the wastegate solenoid and below 10 psi the ECU seems to work open loop as the duty cycle follows a curve related to the revs of the engine (not the boost).
As soon as the boost goes over 10 psi the ECU seems to switch to closed loop as the duty cycle no longer follows the rev related curve but changes with the boost level.
Andy
How do you know about PID controllers?,I work on a chemical plant and most of the instrument department don't understand this subject LOL
Kevin
What type of kit do you work with?.We have Kent DCS systems as well as electronic TCS's and a load of Foxborough and Taylor pneumatic stuff.The pneumatic controllers are about 30 years old and are more reliable than the modern stuff.They are also more fun to work on as you can actually fix them instead of just fitting a new one.
As for the boost control on the scooby I don't think it runs only closed loop control.
I have measured the duty cycle of the wastegate solenoid and below 10 psi the ECU seems to work open loop as the duty cycle follows a curve related to the revs of the engine (not the boost).
As soon as the boost goes over 10 psi the ECU seems to switch to closed loop as the duty cycle no longer follows the rev related curve but changes with the boost level.
Andy
20 May 2001, 10:33 PM
#6
Scooby Regular
Join Date: Jan 2000
Posts: 3,055
Likes: 0
Received 0 Likes
on
0 Posts
That's it!!!
Adam has now proved beyond any reasonable doubt that all of his recent postings displaying a knowledge far greater than that of any tuner I know, have in fact been taken straight out of a book!!!!
Nice one chap!!
Stef.
Adam has now proved beyond any reasonable doubt that all of his recent postings displaying a knowledge far greater than that of any tuner I know, have in fact been taken straight out of a book!!!!
Nice one chap!!
Stef.
21 May 2001, 01:38 PM
#7
Scooby Senior
Join Date: Jun 1999
Posts: 3,410
Likes: 0
Received 0 Likes
on
0 Posts
Well the book he copied it out of was not proof checked properly... 
This: "Basically the ecu asks for a level of boost and the wastegate opens. The actual level produced then misses to target and creates an error." doesn't make as much sense as it could do, does it?
Surely the ecu's boost control subsytem (a simple feedback system as found in basic cybernetics textbooks) will have to _close_ the wastegate in order to make the engine create boost?
Moray
This: "Basically the ecu asks for a level of boost and the wastegate opens. The actual level produced then misses to target and creates an error." doesn't make as much sense as it could do, does it?
Surely the ecu's boost control subsytem (a simple feedback system as found in basic cybernetics textbooks) will have to _close_ the wastegate in order to make the engine create boost?
Moray
Trending Topics
23 May 2001, 02:57 AM
#8
Scooby Regular
Join Date: Jul 2000
Posts: 277
Likes: 0
Received 0 Likes
on
0 Posts
AndyMc
Wow I thought my company were the only firm running 30 yr old Taylor valves, makes me feel better as I thought I was getting left behind with the times!!
Mark (instrument engineer)
Wow I thought my company were the only firm running 30 yr old Taylor valves, makes me feel better as I thought I was getting left behind with the times!!
Mark (instrument engineer)
23 May 2001, 10:17 AM
#9
Scooby Regular
Join Date: Aug 1999
Posts: 7,957
Likes: 0
Received 0 Likes
on
0 Posts
deleted as I think the personal attack of moray that was there (and it was a great one) and the blowing of my proverbial trumpet were too much.
[This message has been edited by Adam M (edited 23 May 2001).]
[This message has been edited by Adam M (edited 23 May 2001).]
23 May 2001, 10:23 AM
#10
Scooby Regular
Join Date: Aug 1999
Posts: 7,957
Likes: 0
Received 0 Likes
on
0 Posts
Moray further to my last post, your limited understanding has been highlighted.
The error is fedback to the ecu (hence a feedback system) and it is the error which would be operated on to get the correct output.
The way the ecu alters the boost is by opening and closing the wastegate solenoid I believe. I said, open, you say close, we are not disagreeing. in order to open the wastegate it must first have been closed???
allowing boost to rise, until a target is reached, or believed to be reached, the wastegate is then opened.
I actually wrote, the ecu asks for a level of boost and the waste gate is openend, I assumed you had enough knowledge to interpret that as, the wastegate is opened when that level is reached. It was a bad assumption, I should have realised you are pedantic man.
I was just suggesting that the closed loop feedback control of the wastegate could be operated using a pid controller, but no you had to assume I was wrong!
I may have been, I can admit it. Pat is more likely to know this one, and also a lot more about PID controllers too.
You may have to check with your girlfriend to see if you are allowed to admit you are wrong, I will make my own decisions
PS. from our phone conversation today, I got the impression that you thought I was being serious, when I am not. You haven't offended me, you should know I am thick skinned and I assume the same of you.
A nice retaliatory post would quite fun, if you are up to it.
But to my knowledge, you are mostly incapable of of it, mostly.
[This message has been edited by Adam M (edited 23 May 2001).]
The error is fedback to the ecu (hence a feedback system) and it is the error which would be operated on to get the correct output.
The way the ecu alters the boost is by opening and closing the wastegate solenoid I believe. I said, open, you say close, we are not disagreeing. in order to open the wastegate it must first have been closed???
allowing boost to rise, until a target is reached, or believed to be reached, the wastegate is then opened.
I actually wrote, the ecu asks for a level of boost and the waste gate is openend, I assumed you had enough knowledge to interpret that as, the wastegate is opened when that level is reached. It was a bad assumption, I should have realised you are pedantic man
.I was just suggesting that the closed loop feedback control of the wastegate could be operated using a pid controller, but no you had to assume I was wrong!
I may have been, I can admit it. Pat is more likely to know this one, and also a lot more about PID controllers too.
You may have to check with your girlfriend to see if you are allowed to admit you are wrong, I will make my own decisions
PS. from our phone conversation today, I got the impression that you thought I was being serious, when I am not. You haven't offended me, you should know I am thick skinned and I assume the same of you.
A nice retaliatory post would quite fun, if you are up to it.
But to my knowledge, you are mostly incapable of of it, mostly.
[This message has been edited by Adam M (edited 23 May 2001).]
23 May 2001, 01:19 PM
#11
Scooby Senior
Join Date: Jun 1999
Posts: 3,410
Likes: 0
Received 0 Likes
on
0 Posts
Adam,
You actually wrote _exactly_ :
"Basically the ecu asks for a level of boost and the wastegate opens. The actual level produced then misses to target and creates an error. The PID controller uses the error and calculates how it needs to modify the wastegate control to make the error zero.
"
Let us break this paragraph down into component sentences:
1: "Basically the ecu asks for a level of boost and the wastegate opens."
2: "The actual level produced then misses to target and creates an error."
3: "The PID controller uses the error and calculates how it needs to modify the wastegate control to make the error zero."
Now look closely at (1). You say "Basically the ecu asks for a level of boost and the wastegate opens.".
You don't say "Basically the ecu asks for a level of boost and _THEN_ the wastegate opens.".
You don't say "Basically the ecu asks for a level of boost and the wastegate opens _WHEN THIS TARGET LEVEL IS ACHIEVED OR INDEED EXCEEDED_"
When you say:
"I actually wrote, the ecu asks for a level of boost and the waste gate is openend, I assumed you had enough knowledge to interpret that as, the wastegate is opened when that level is reached."...
does this mean:
"What I actually wrote is wrong but I expect the technically competent reader will be able to get the general idea of what I actually should have written because they probably know more about it than me... or in the case of less technically knowledgeable readers... what I've written sounds plausible enough to be taken as gospel, thus giving said reader the impression that I am a techical whizz."?
Kindest regards
Moray Pedantic's my middle name Mackenzie
PS: Sorry about the call... my phone went out of reception coverage and dropped the line... Didn't mean to hang up on you. Well, not at that point anyway...
You actually wrote _exactly_ :
"Basically the ecu asks for a level of boost and the wastegate opens. The actual level produced then misses to target and creates an error. The PID controller uses the error and calculates how it needs to modify the wastegate control to make the error zero.
"
Let us break this paragraph down into component sentences:
1: "Basically the ecu asks for a level of boost and the wastegate opens."
2: "The actual level produced then misses to target and creates an error."
3: "The PID controller uses the error and calculates how it needs to modify the wastegate control to make the error zero."
Now look closely at (1). You say "Basically the ecu asks for a level of boost and the wastegate opens.".
You don't say "Basically the ecu asks for a level of boost and _THEN_ the wastegate opens.".
You don't say "Basically the ecu asks for a level of boost and the wastegate opens _WHEN THIS TARGET LEVEL IS ACHIEVED OR INDEED EXCEEDED_"
When you say:
"I actually wrote, the ecu asks for a level of boost and the waste gate is openend, I assumed you had enough knowledge to interpret that as, the wastegate is opened when that level is reached."...
does this mean:
"What I actually wrote is wrong but I expect the technically competent reader will be able to get the general idea of what I actually should have written because they probably know more about it than me... or in the case of less technically knowledgeable readers... what I've written sounds plausible enough to be taken as gospel, thus giving said reader the impression that I am a techical whizz."?
Kindest regards
Moray Pedantic's my middle name Mackenzie
PS: Sorry about the call... my phone went out of reception coverage and dropped the line... Didn't mean to hang up on you. Well, not at that point anyway...
23 May 2001, 01:37 PM
#13
Scooby Regular
Join Date: Mar 2001
Posts: 32
Likes: 0
Received 0 Likes
on
0 Posts
Didnt think you had hung up on me.
BTW, it should be misses the target.
Moray, we both know what I had meant.
As far as I am concerned, it is a moot point.
The reason for this is that I am not even sure if the wastegate is run by an ecu based pid controller. Whether it is or not, the fact that presumed I thought the wastegate had to be open for the boost to rise is not an issue here.
If it was infact the case that the ecu keeps the wastegate open for the boost to rise and then closes it when the target is reached the pid control would work in the same way but by closing the wastegate rather than opening to modulate teh boost pressure.
Given that I was answering a question relating to why overboost might occur, the answer was possibly due to pid control. That solution remains a viable answer regardless of exact knowledge of how our turbos work(or any turbo that I know of).
Whether my word is taken as gospel or not(and it shouldnt unless posting under the name of GOD) the suggetsion I put forward is plausible.
I think you are just pissed cos you dont know what PID controllers are.
PS. thanks for explanation of middle name,
wondered why you wrote Moray P Mackenzie as your name on the sms.
If you want to learn about PID controllers I do have some good books, but they arent much fun.
BTW, it should be misses the target.
Moray, we both know what I had meant.
As far as I am concerned, it is a moot point.
The reason for this is that I am not even sure if the wastegate is run by an ecu based pid controller. Whether it is or not, the fact that presumed I thought the wastegate had to be open for the boost to rise is not an issue here.
If it was infact the case that the ecu keeps the wastegate open for the boost to rise and then closes it when the target is reached the pid control would work in the same way but by closing the wastegate rather than opening to modulate teh boost pressure.
Given that I was answering a question relating to why overboost might occur, the answer was possibly due to pid control. That solution remains a viable answer regardless of exact knowledge of how our turbos work(or any turbo that I know of).
Whether my word is taken as gospel or not(and it shouldnt unless posting under the name of GOD) the suggetsion I put forward is plausible.
I think you are just pissed cos you dont know what PID controllers are.
PS. thanks for explanation of middle name,
wondered why you wrote Moray P Mackenzie as your name on the sms.
If you want to learn about PID controllers I do have some good books, but they arent much fun.
23 May 2001, 01:41 PM
#14
Scooby Regular
Join Date: Apr 2001
Posts: 36
Likes: 0
Received 0 Likes
on
0 Posts
Sorry, meant to post that under name of Adam M,
rest assured Adam M = GOD (in this situation anyway)
ps. stef, I deleted teh post which insulted you so I shall have to reinstate something here.
Why have you jumped on the bandwagon to hae a go at me (waiting for harj - to be honest)?
Needless to say, I am going to have have to kick your **** when I see you.
Anyone else want a free shot?
Cem
R19KET
Danny Fisher
Craig H
Firefox ?????????
Come on guys, I am feeling hard done by!
rest assured Adam M = GOD
(in this situation anyway)ps. stef, I deleted teh post which insulted you so I shall have to reinstate something here.
Why have you jumped on the bandwagon to hae a go at me (waiting for harj - to be honest)?
Needless to say, I am going to have have to kick your **** when I see you.
Anyone else want a free shot?
Cem
R19KET
Danny Fisher
Craig H
Firefox ?????????
Come on guys, I am feeling hard done by!
23 May 2001, 11:19 PM
#21
Scooby Regular
Join Date: Mar 2000
Location: Gloucestershire, home of the lawnmower.
Posts: 4,531
Likes: 0
Received 0 Likes
on
0 Posts
Adam,
I have noticed recently your (almost) sudden increase in technical knowledge, or more exactly, your technicaly natured posts.
Have you been digesting automotive theory books and Subaru workshop manuals ?
Or are you mostly spouting stuff you have picked up from others and do not exactly understand what you are saying ?
Cheers
Ian
(Mostly)
I have noticed recently your (almost) sudden increase in technical knowledge, or more exactly, your technicaly natured posts.
Have you been digesting automotive theory books and Subaru workshop manuals ?
Or are you mostly spouting stuff you have picked up from others and do not exactly understand what you are saying ?
Cheers
Ian
(Mostly)
Thread
Thread Starter
Forum
Replies
Last Post
hedgecutter
General Technical
3
25 September 2015 02:35 PM
