SERT Self Tuning question
Big Kahuna HDF Member
Joined: Feb 2006
Posts: 23,881
Likes: 34
From: USA
RE: SERT Self Tuning question
ORIGINAL: Steve Cole
One needs to remember that the HDECM is produced down the same assemble line that the GM marine ECM is made. Let's just say they share a fair amount of hardware between each of them. The new 2008 DBW ECM event switches to the same connector GM has been using for years. The Knock method has always been remove first and add back. When tuned properly it works pretty well this way.
One needs to remember that the HDECM is produced down the same assemble line that the GM marine ECM is made. Let's just say they share a fair amount of hardware between each of them. The new 2008 DBW ECM event switches to the same connector GM has been using for years. The Knock method has always been remove first and add back. When tuned properly it works pretty well this way.
Thread Starter
|
Road Master
Joined: Mar 2007
Posts: 1,177
Likes: 23
RE: SERT Self Tuning question
I have a new question.
You set the entire AFR map to 13.2 and do the perfect tune thru the entire VE map to 13.2 AFR
You set the entire 70 KPA column of the AFR table to 14.5 AFR
Now you set the entire low load section of the AFR table to 14.6 to force closed loop
Youmax outthe o2 bias table to force the entire center of the map to 14.2 AFR
You go out and drive the bike for a few hundred miles to get the ADF and ADV homed in on some new answer that you cant see from the SERT software.
now you get the motor to 70 KPA back on on the dyno
The question is:
Will the code attempt to hit 14.5 AFR as you see in the AFR table and actually go leaner (remember it was at 14.2 due to the biasing)
OR
will the code try to hit 14.1= (14.5 / 14.6) * 14.2
OR
will the code try to hit 14.1= 14.2 - (14.6 - 14.5)
OR
some other answer
AW
You set the entire AFR map to 13.2 and do the perfect tune thru the entire VE map to 13.2 AFR
You set the entire 70 KPA column of the AFR table to 14.5 AFR
Now you set the entire low load section of the AFR table to 14.6 to force closed loop
Youmax outthe o2 bias table to force the entire center of the map to 14.2 AFR
You go out and drive the bike for a few hundred miles to get the ADF and ADV homed in on some new answer that you cant see from the SERT software.
now you get the motor to 70 KPA back on on the dyno
The question is:
Will the code attempt to hit 14.5 AFR as you see in the AFR table and actually go leaner (remember it was at 14.2 due to the biasing)
OR
will the code try to hit 14.1= (14.5 / 14.6) * 14.2
OR
will the code try to hit 14.1= 14.2 - (14.6 - 14.5)
OR
some other answer
AW
Outstanding HDF Member
Joined: Apr 2006
Posts: 2,537
Likes: 21
RE: SERT Self Tuning question
Any value OTHER than 14.6, the bike is out of Closed Loop...
That is the "Switch" to activete the CLB in that zone.....
Any value less than 14.6, that is what will be generated, providing that the VE's were dialed in to the DESIRED AFR....
That is the "Switch" to activete the CLB in that zone.....
Any value less than 14.6, that is what will be generated, providing that the VE's were dialed in to the DESIRED AFR....
Thread Starter
|
Road Master
Joined: Mar 2007
Posts: 1,177
Likes: 23
RE: SERT Self Tuning question
But.....
If the entire VE table was perfect I think First the 02 Integrater wouldjump to 1.035 everyware, theAFT would in migrate to 1.035 everyware in an attempt to get the 02 integrator to an average of 1. Last the code slowlyslides the AFT to a 0 average and the code moves the 1.035 to the AFV.
Now when the AFR table finds 14.5 AFR(open loop mode)it still uses the 1.035 AFV and winds up at 14.5/1.035 and ends up at 14.01 AFR
This would also happen also at WOT so if you really want 13.0 AFR you have to set the targets at 13*1.035 or 13.5
It may bepartlywraped upin that hidden, pesky, non resetableand non viewable AFV.
note: if you look at a SERT data log when the motor is in open loop, the VE New is still not the same as VE. Something ischanging it.
I hate not having full and total access to the programmers. This is the logic part of the Megasquirt that I worked with the programmers every day to get the code rock solid. I always took the side "If we can't explain it in the manual, we can't expect people to tune it."
AW
If the entire VE table was perfect I think First the 02 Integrater wouldjump to 1.035 everyware, theAFT would in migrate to 1.035 everyware in an attempt to get the 02 integrator to an average of 1. Last the code slowlyslides the AFT to a 0 average and the code moves the 1.035 to the AFV.
Now when the AFR table finds 14.5 AFR(open loop mode)it still uses the 1.035 AFV and winds up at 14.5/1.035 and ends up at 14.01 AFR
This would also happen also at WOT so if you really want 13.0 AFR you have to set the targets at 13*1.035 or 13.5
It may bepartlywraped upin that hidden, pesky, non resetableand non viewable AFV.
note: if you look at a SERT data log when the motor is in open loop, the VE New is still not the same as VE. Something ischanging it.
I hate not having full and total access to the programmers. This is the logic part of the Megasquirt that I worked with the programmers every day to get the code rock solid. I always took the side "If we can't explain it in the manual, we can't expect people to tune it."
AW
HD EFI Guru
Joined: Jan 2008
Posts: 4,462
Likes: 4,718
RE: SERT Self Tuning question
ORIGINAL: whittlebeast
I have a new question.
You set the entire AFR map to 13.2 and do the perfect tune thru the entire VE map to 13.2 AFR
You set the entire 70 KPA column of the AFR table to 14.5 AFR
Now you set the entire low load section of the AFR table to 14.6 to force closed loop
Youmax outthe o2 bias table to force the entire center of the map to 14.2 AFR
AW
I have a new question.
You set the entire AFR map to 13.2 and do the perfect tune thru the entire VE map to 13.2 AFR
You set the entire 70 KPA column of the AFR table to 14.5 AFR
Now you set the entire low load section of the AFR table to 14.6 to force closed loop
Youmax outthe o2 bias table to force the entire center of the map to 14.2 AFR
AW
Thread Starter
|
Road Master
Joined: Mar 2007
Posts: 1,177
Likes: 23
RE: SERT Self Tuning question
ORIGINAL: Steve Cole
When setting the bike up and when tuning the O2 bias table after the fact your going to get different results. The ECM is going to learn when in closed loop then apply what it's learned event in open loop. You need to start doing the basic stuff and see the results and quit jumping to conclusions about whats is going to happen. As I told you on the phone and again here it's not a mega squirt and it works differently.
When setting the bike up and when tuning the O2 bias table after the fact your going to get different results. The ECM is going to learn when in closed loop then apply what it's learned event in open loop. You need to start doing the basic stuff and see the results and quit jumping to conclusions about whats is going to happen. As I told you on the phone and again here it's not a mega squirt and it works differently.
I am just trying to get my head around how this system works. As I always tell the programmers over at MS, "If you can't explain it in the documentation, they can't tune it. So how are we going to explain that one?"
AW
Thread Starter
|
Road Master
Joined: Mar 2007
Posts: 1,177
Likes: 23
RE: SERT Self Tuning question
We got the sert log viewer working today
It is available for free download from http://www.ncs-stl.com/sert/MegaLogV...tup-Harley.zip
also download http://www.ncs-stl.com/sert/test12.DM2This is the sert data file
and http://www.ncs-stl.com/sert/SertOutput12.csvThis is the sert output file that was created that can be loaded into the viewer
Here is a screen shot from the viewer off my bike
Have fun
AW
It is available for free download from http://www.ncs-stl.com/sert/MegaLogV...tup-Harley.zip
also download http://www.ncs-stl.com/sert/test12.DM2This is the sert data file
and http://www.ncs-stl.com/sert/SertOutput12.csvThis is the sert output file that was created that can be loaded into the viewer
Here is a screen shot from the viewer off my bike
Have fun
AW
HD Forum Stories
The Best of Harley-Davidson for Lifelong Riders
8 Best Harley-Davidson Motorcycles Ever
Pouria Savadkouei
10 Worst Harley-Davidson Motorcycles Ever
Pouria Savadkouei
Killer Custom's Jail Break Is The Breakout That Refused to Blend In
Verdad Gallardo
Crazy Bunderbike Build Looks Amazing, But Is It Impossible to Ride?
Verdad Gallardo
Harley-Davidson Reveals Super Cool Cafe Racer Concept
Verdad Gallardo
Engraved Rebellion: Inside Bundnerbike's Glam Rock II
Verdad Gallardo
10 Motorcycles You Should Never Buy
Joe Kucinski
10 Things Harley-Davidson Needs to Fix in 2026
Verdad Gallardo
Southpaw Super Glide: A Left-Hand-Drive 1979 Harley FXE Built to Fit the Rider
Verdad Gallardo
HD EFI Guru
Joined: Jan 2008
Posts: 4,462
Likes: 4,718
RE: SERT Self Tuning question
The only way to "get your head around how it works" is to go out and start useing it. Then spend the time to review the data. Your not going to do it by sitting down and looking at the same old data the same old way. It's different and needs to be treated different.
Thread Starter
|
Road Master
Joined: Mar 2007
Posts: 1,177
Likes: 23
RE: SERT Self Tuning question
I choose to develope the tools to review the data first and develope the stock fuel and timing maps. Harley elected to
not include the factory original maps making it very difficult to return a bike back to stock without the help and mercy
ofthe dealer network. They are in the business of selling shop time at $100/hr.
When you want to change the wheels and tires on your car the first two thing you go find is a lug wrench and a jack.
Without those two tools nothing else in the project matters. When I build aracecar, I first design and prove the entire
concept in FEA.The number of options and pitfalls of chassis design is impossible to comprehend without access to the
tools to locate the pits. EFI tuning at home is similar. Purchasing a dyno is out of the question to tune one bike. I
would never be able to afford the dyno time at a dealer,required to get my bike to the level of drivability I demand. I
have been there, done that. I can do it again with agood set of tools.
The guys from MSconverted the MS data logger in about 30 lines of code. Everything was designed to be adjustable
from day one. The next release available for free from http://www.efianalytics.com/MegaLogViewer/will include the option for
reading Harley Delfi data log files.
AW
not include the factory original maps making it very difficult to return a bike back to stock without the help and mercy
ofthe dealer network. They are in the business of selling shop time at $100/hr.
When you want to change the wheels and tires on your car the first two thing you go find is a lug wrench and a jack.
Without those two tools nothing else in the project matters. When I build aracecar, I first design and prove the entire
concept in FEA.The number of options and pitfalls of chassis design is impossible to comprehend without access to the
tools to locate the pits. EFI tuning at home is similar. Purchasing a dyno is out of the question to tune one bike. I
would never be able to afford the dyno time at a dealer,required to get my bike to the level of drivability I demand. I
have been there, done that. I can do it again with agood set of tools.
The guys from MSconverted the MS data logger in about 30 lines of code. Everything was designed to be adjustable
from day one. The next release available for free from http://www.efianalytics.com/MegaLogViewer/will include the option for
reading Harley Delfi data log files.
AW