MS3 EBC Issues
#21
Does the sensitivity slider work alongside the PID? How do you work with the two in conjunction? After doing some work on it the past two days I don't feel like I am any closer to figuring it out than I was when I started.
One thing I noticed was on fast spool, the valve oscillates between 100% and single digits. Is this normal?
I'd work on it and pull some logs on the way home but it appears my laptop finally died. Perhaps it is only the hard drive, we will see.
One thing I noticed was on fast spool, the valve oscillates between 100% and single digits. Is this normal?
I'd work on it and pull some logs on the way home but it appears my laptop finally died. Perhaps it is only the hard drive, we will see.
#22
Once you get CL working of course... Until then it's frustrating as heck. The problem I think is that there are a bunch of variables that can cause you problems. If you just happened to not have those issues in your system everything goes together easy peasy. If not well...
I have yet to get a truly reliable closed loop system working. That said, I am knocking on the door now, I hope.
I have yet to get a truly reliable closed loop system working. That said, I am knocking on the door now, I hope.
1) First leave Boost Control off completely and run vacuum from from wastegate to the cold side just before the throttle body.
2) If internally gated set the actuator preload. (look on youtube for guides **** is easy)
3) Drive the car and note how much psi you max at. Try and play with preload so the wastegae it building up enough psi to be about 3-5psi under your end goal. You may need a stiffer/less stiff spring to achieve this.
4) Next hook up the solenoid ant test it by running open loop ebc and setting the duty table to 50 all the way across and set your solenoid fq. to 19hz. You should here clicking. If no clicking check your wiring and or settings. Make sure you have the correct port selected.
5) Next switch to closed loop and set your table to 0 @ 0kpa to rev limit and whatever your target is in all the cells above it
6) go into cl settings and set you solenoid fq. to 39hz. (this will get rid of the clicking) and begin tuning CL by setting the slider under basic settings
7) Set the slider till you get oscillations then back off a little. Now set the PID.
8) In 2nd gear ( I use 2nd and 4th) do some pulls and slowly lower P to slightly go over your target.
9) Once you are hitting your target you may begin to have some slight oscillations DO NOT TOUCH THE SLIDER. This is where you will add I. Add I till the oscillation are removed.
10) Now that you have removed the oscillations you may have fallen bellow your target. No worries, Just decrease P 1 number at a time till you get back on target.
11) You may notice now your get some oscillation again no need to fear add a little more I GO SLOWLY.
12) By this time you should be close to being done. Now is the time where I go back into my boost duty table and change my targets to fit my driving style.
13) As for D I have never had to use it on both the 2554r and 2560R. But if used to dampen it does not require much.
At this point you may need to go change clothes and wash yourself after the big mess you made.
#23
5) Next switch to closed loop and set your table to 0 @ 0kpa to rev limit and whatever your target is in all the cells above it
7) Set the slider till you get oscillations then back off a little. Now set the PID.
8) In 2nd gear ( I use 2nd and 4th) do some pulls and slowly lower P to slightly go over your target.
9) Once you are hitting your target you may begin to have some slight oscillations DO NOT TOUCH THE SLIDER. This is where you will add I. Add I till the oscillation are removed.
10) Now that you have removed the oscillations you may have fallen bellow your target. No worries, Just decrease P 1 number at a time till you get back on target.
11) You may notice now your get some oscillation again no need to fear add a little more I GO SLOWLY.
12) By this time you should be close to being done. Now is the time where I go back into my boost duty table and change my targets to fit my driving style.
7) Set the slider till you get oscillations then back off a little. Now set the PID.
8) In 2nd gear ( I use 2nd and 4th) do some pulls and slowly lower P to slightly go over your target.
9) Once you are hitting your target you may begin to have some slight oscillations DO NOT TOUCH THE SLIDER. This is where you will add I. Add I till the oscillation are removed.
10) Now that you have removed the oscillations you may have fallen bellow your target. No worries, Just decrease P 1 number at a time till you get back on target.
11) You may notice now your get some oscillation again no need to fear add a little more I GO SLOWLY.
12) By this time you should be close to being done. Now is the time where I go back into my boost duty table and change my targets to fit my driving style.
7) What should the slider be to start with? All the way to the right or left? What should the PID numbers be to start the process? 100:0:0?
So basically, set sensitivity first and once it's set don't touch it? I still don't understand the 0 @ 0kpa statement in item #5. You are trying to adjust sensitivity to hit target, what is your target if you set your numbers to zero?
#24
Boost Czar
iTrader: (62)
Join Date: May 2005
Location: Chantilly, VA
Posts: 79,499
Total Cats: 4,080
If you want to use the old method, you can just input your PID values.
Otherwise the slider sensitivity thing is it's own algorithm.
so, one or the other i suppose.
#25
"Boost Control Lower Limit Delta" should NOT be 100...
Basically what this setting does is prevent your PID from working until you get within X kpa of your target. X being the number you enter in that box. If you have 100kpa in there, your PID is working nearly the whole time. Change the setting to about 10kpa or less.
EDIT: Some people (Brain?) are using something like 27kpa... I am still doing research on this, but it certainly shouldn't be 100kpa.
Last edited by Chiburbian; 08-20-2014 at 07:00 PM.
#26
Boost Czar
iTrader: (62)
Join Date: May 2005
Location: Chantilly, VA
Posts: 79,499
Total Cats: 4,080
Cant remember; pretty sure I had 100% in there. Not even sure what 100% even means to be honest. If it's the kPa value that CL PID starts, then I'd use lower than 100 for sure so it keeps the solenoid closed, earlier, for faster spoolage.
here's how I tune CL EBC.
1. Make sure the solenoid is hooked it up and properly configured. This means selecting the correct output pin, freq. (39Hz for DIY solenoid), polarity, etc.
2. Populate CL targets
3. Set PID to 100 0 0
4. Log, do a quick pull to spool up the turbo.
5. If CL targets are not reached, then lower P, repeat #4. (will probably need to go to ~15 0 0)
6. Once PID CL EBC is hitting your CL target lower it slightly so it peaks over it 1-2psi, log and do redline pulls.
7. Increase I until your target is maintained through the revs. Increasing I will lower the sensitivity to P, so P must be added back in as I is increased.
8. Add D to smooth out the boost plot. Increase I and P where necessary.
I dunno why you put ******* with freq. and preload on your list. Your wastegate preload should be tight. EBC can only increase boost over the wastegate, If you're running a wastegate that's 3psi off your target, then it's almost pointless to run EBC. Knowing where your boost level without EBC is fine, but there's not reason to add/reduce preload. If you make more boost than your goal, then you have other issues and shouldn't be bothering with EBC. The solenoid most people use is designed to run at 39Hz, so use 39Hz.
I've never used the sensitivity slider so I can't talk to it.
here's how I tune CL EBC.
1. Make sure the solenoid is hooked it up and properly configured. This means selecting the correct output pin, freq. (39Hz for DIY solenoid), polarity, etc.
2. Populate CL targets
3. Set PID to 100 0 0
4. Log, do a quick pull to spool up the turbo.
5. If CL targets are not reached, then lower P, repeat #4. (will probably need to go to ~15 0 0)
6. Once PID CL EBC is hitting your CL target lower it slightly so it peaks over it 1-2psi, log and do redline pulls.
7. Increase I until your target is maintained through the revs. Increasing I will lower the sensitivity to P, so P must be added back in as I is increased.
8. Add D to smooth out the boost plot. Increase I and P where necessary.
I dunno why you put ******* with freq. and preload on your list. Your wastegate preload should be tight. EBC can only increase boost over the wastegate, If you're running a wastegate that's 3psi off your target, then it's almost pointless to run EBC. Knowing where your boost level without EBC is fine, but there's not reason to add/reduce preload. If you make more boost than your goal, then you have other issues and shouldn't be bothering with EBC. The solenoid most people use is designed to run at 39Hz, so use 39Hz.
I've never used the sensitivity slider so I can't talk to it.
Last edited by Braineack; 08-21-2014 at 07:56 AM.
#27
Elite Member
iTrader: (1)
Join Date: Jun 2006
Location: Warrington/Birmingham
Posts: 2,642
Total Cats: 42
The developer explained it once to me. He said he basically reworked the code that makes tuning PID as simple as moving the slider.
If you want to use the old method, you can just input your PID values.
Otherwise the slider sensitivity thing is it's own algorithm.
so, one or the other i suppose.
If you want to use the old method, you can just input your PID values.
Otherwise the slider sensitivity thing is it's own algorithm.
so, one or the other i suppose.
The slider just replaces the valve open / closed at numbers. Adjusting those just adjusted the sensitivity of the algorithm, the slider simplifies this.
If you use the 'simple' tuning method , the PID values are whatever are in the grey'd out boxes, default is 100,100,0 IIRC
Go advanced and mess with the PID values, to fine tune everything, but the sensitivity slider is still in use. Go back to simple and the PID values used are still what ever is in the PID boxes
This entire thread is worth a read, but Ken's post here confirms the slider is always active
http://www.msextra.com/forums/viewto...b9bd1&start=80
#29
Elite Member
iTrader: (1)
Join Date: Jun 2006
Location: Warrington/Birmingham
Posts: 2,642
Total Cats: 42
This thread explains it pretty well.
http://www.msextra.com/forums/viewto...=51203&start=0
The sensitivity slider is a 'gain' for the PID routine. It doesn't replace it.
#30
Boost Czar
iTrader: (62)
Join Date: May 2005
Location: Chantilly, VA
Posts: 79,499
Total Cats: 4,080
most people can just use the default PID settings and tune [idle pid] using the slider. And if they can't get it to where they're happy they can turn on "advanced" mode and tune the PID gains.
#32
So after upgrading to the latest firmware I still get oscillation at full throttle. However at 50-75% it builds boost perfectly to the target with no oscillation.
At full throttle, 3rd gear, starting at 2500RPM with the slider low (<50) I get a spike past my target to boost cut (10kpa over target). With the slider high (125) I get oscillation of 2.5-3 psi and it increases until the boost cut is hit on the 4th or 5th oscillation. Graphed it's clear the pressure oscillations are increasing as are the duty cycle corrections from the CL Boost control.
Any thoughts on why it works at partial throttle? I'm thinking it's possible I need to add some PID to make it work, but most people seem to get it working with just the slider.
At full throttle, 3rd gear, starting at 2500RPM with the slider low (<50) I get a spike past my target to boost cut (10kpa over target). With the slider high (125) I get oscillation of 2.5-3 psi and it increases until the boost cut is hit on the 4th or 5th oscillation. Graphed it's clear the pressure oscillations are increasing as are the duty cycle corrections from the CL Boost control.
Any thoughts on why it works at partial throttle? I'm thinking it's possible I need to add some PID to make it work, but most people seem to get it working with just the slider.
#40
Frustrating....tried a series of D, I and P values. Varying results but still no stable boost control.
If it weren't for the fact the OL gives good results I'd think I had a mechanical issue. It's almost as if the system doesn't respond fast enough to quick changes in boost pressure. Perhaps my cold side location for the EBC solenoid is too far from the turbo? I really doubt it.
For now the plan is to continue to experiment the slider and PID settings and their impact on performance.
Question: I have pretty good results with the slider in "Basic" mode, but still 1-2 PSI oscillation. When I go to advanced I get much bigger oscillations or when I lower P I shoot right past my target. Suggestions on how to set P so that I get the same results as just the slider in Basic mode?
If it weren't for the fact the OL gives good results I'd think I had a mechanical issue. It's almost as if the system doesn't respond fast enough to quick changes in boost pressure. Perhaps my cold side location for the EBC solenoid is too far from the turbo? I really doubt it.
For now the plan is to continue to experiment the slider and PID settings and their impact on performance.
Question: I have pretty good results with the slider in "Basic" mode, but still 1-2 PSI oscillation. When I go to advanced I get much bigger oscillations or when I lower P I shoot right past my target. Suggestions on how to set P so that I get the same results as just the slider in Basic mode?