ms2extra 3.2.0 alpha gslender v1
12-04-2011, 08:17 AM
#1
Junior Member
Thread Starter
iTrader: (1)
Join Date: Jun 2011
Location: Australia
Posts: 178
Total Cats: 3
ms2extra 3.2.0 alpha gslender v1
Yo... I've just posted up a new MS2 mod firmware with good bits 
Check it out >>>
http://www.msextra.com/forums/viewto...p?f=91&t=43176
Check it out >>>
http://www.msextra.com/forums/viewto...p?f=91&t=43176
Reply
1
1
12-04-2011, 11:43 PM
#4
Junior Member
Thread Starter
iTrader: (1)
Join Date: Jun 2011
Location: Australia
Posts: 178
Total Cats: 3
You'd be best to vist the 1st post on ms2e (the link I provided) and read the summary and then the included documentation. It outlines and explains each feature and you can then decided for yourself.
I don't think it is sensible to discuss them again invidually here - if you get stuck or can't make sense of the docs, then by all means, post away your questions....
G
Reply
0
0
12-05-2011, 06:09 PM
#6
Junior Member
Join Date: Jun 2007
Posts: 411
Total Cats: 0
To tune voltage correction, I disabled the AC (uncheck WLED), set AC duty adder to 1 (zero turns off the counter)- so only fan load is seen when the AC switch is turned. Set PID to 0/10/0. Consistent and repeatable.
Set floor to 13.0v (a hair under steady state voltage), and correction to 10. Still had noticeable fan droop. Correction factor of 16 did the trick.
Once I was satisfied with voltage correction, I turned on the AC and the corresponding AC idle up correction (15 for my car). It worked pretty well! Then I returned my normal PID values 25/25/120. Even smoother
RPM fluctuation is +/- 20 RPM from target at no load and with AC/fan load. Headlights still upset it, ~ +/- 40 RPM from target. Light years better compared to a few months ago! Remember I have some cam lope to deal with, stock cams should get better results.
Set floor to 13.0v (a hair under steady state voltage), and correction to 10. Still had noticeable fan droop. Correction factor of 16 did the trick.
Once I was satisfied with voltage correction, I turned on the AC and the corresponding AC idle up correction (15 for my car). It worked pretty well! Then I returned my normal PID values 25/25/120. Even smoother
RPM fluctuation is +/- 20 RPM from target at no load and with AC/fan load. Headlights still upset it, ~ +/- 40 RPM from target. Light years better compared to a few months ago! Remember I have some cam lope to deal with, stock cams should get better results.
Reply
0
0
12-05-2011, 06:47 PM
#8
Junior Member
Join Date: Jun 2007
Posts: 411
Total Cats: 0
Here's a screenshot of my enhancements window. Remember I have large throttled volume, so some settings may be off for most guys. I suspect the AC idle up duty is a lot higher in my case, to fill up that volume.
My MAT correction settings are untouched from the previous prerelease versions. It used to be constantly added to the PID duty output, but that caused some jitter in steady state. Ken pointed out that it didn't need to be constantly applied. So now it is applied only during the calculation of the entry duty, on throttle lift. Regarding the settings, I got the 1 duty% per 5 degrees F value off my previous datalogs, I suppose it would vary depending on the local climate conditions. The baseline temp, I set to my normal MAT temp (pretty hot here). Below that baseline temp, the correction will be a negative value. Above it, positive.
So I suggest everyone take the data for their individual car/engine/climate, in order to get the best out of these enhancements.
My MAT correction settings are untouched from the previous prerelease versions. It used to be constantly added to the PID duty output, but that caused some jitter in steady state. Ken pointed out that it didn't need to be constantly applied. So now it is applied only during the calculation of the entry duty, on throttle lift. Regarding the settings, I got the 1 duty% per 5 degrees F value off my previous datalogs, I suppose it would vary depending on the local climate conditions. The baseline temp, I set to my normal MAT temp (pretty hot here). Below that baseline temp, the correction will be a negative value. Above it, positive.
So I suggest everyone take the data for their individual car/engine/climate, in order to get the best out of these enhancements.
Reply
0
0
12-05-2011, 11:20 PM
#9
Junior Member
Join Date: Jun 2007
Posts: 411
Total Cats: 0
Am chasing ghosts now, the differences are so small. But some more fiddling with the voltage compensation has taken care of the headlight oscillation! It just needed a smidge more correction factor (18 from 16).
I also showed how to use the timing of the fan to smoothen the transition by flattening the RPM surge from the AC adder duty.
I also showed how to use the timing of the fan to smoothen the transition by flattening the RPM surge from the AC adder duty.
Reply
0
0
12-10-2011, 05:21 AM
#13
Junior Member
Join Date: Jun 2007
Posts: 411
Total Cats: 0
Testing a prerelease firmware 3.2.0 gslender v1.1, with an new feature- modified idle advance, which is now based on RPM.
Options are to have it respond to MAP or RPM, and to have the output as an absolute value, or as an adder to the spark advance table.
These are the settings I tested with (RPM target = 900). Since we are after RPM control anyway, I believe the advance would be better if directly related to the RPM, or RPM error. Correction for MAP can be built into the Spark Advance table as a hill going upwards.
The results:
I was amazed at how stable the steady state idle was. I added some smoothing factor to the idle valve duty output, from 2 to 3. Idle valve is very quiet now at steady state. And I just spent 10 minutes on it!
The video:
Bottom right indicators show spark advance and PWM idle duty values.
My minimum PID duty is 32, which is why at the start (no load), the spark couldn't bring the RPM closer that 20 RPM over target. I believe that one dip below 750, I went too far with the advance and fell off the spark- torque plateau, making the dip worse. Should be able to adjust for that. Thanks to gslender for the great work!
Options are to have it respond to MAP or RPM, and to have the output as an absolute value, or as an adder to the spark advance table.
These are the settings I tested with (RPM target = 900). Since we are after RPM control anyway, I believe the advance would be better if directly related to the RPM, or RPM error. Correction for MAP can be built into the Spark Advance table as a hill going upwards.
The results:
I was amazed at how stable the steady state idle was. I added some smoothing factor to the idle valve duty output, from 2 to 3. Idle valve is very quiet now at steady state. And I just spent 10 minutes on it!
The video:
Bottom right indicators show spark advance and PWM idle duty values.
My minimum PID duty is 32, which is why at the start (no load), the spark couldn't bring the RPM closer that 20 RPM over target. I believe that one dip below 750, I went too far with the advance and fell off the spark- torque plateau, making the dip worse. Should be able to adjust for that. Thanks to gslender for the great work!
Last edited by Greg G; 12-10-2011 at 05:36 AM.
Reply
0
0
12-12-2011, 05:44 AM
#14
Junior Member
Join Date: Jun 2007
Posts: 411
Total Cats: 0
I thought about how to go about tuning this scientifically- so I did a spark vs MAP test. Set the idle valve to a fixed duty and ran the spark from -4 to 40. Got some interesting data out of it!
Data indicates that near the idle area, torque is directly/linearly related to spark. And since idle control is basically balancing all the different torque loads on the engine, we should be operating at this area. 10-12 is the sweet spot for the base idle- this leaves a margin of at least 10 degrees below (to reduce torque) and 10 degrees above (to increase torque). Above 20-21, the torque still increases, but at a decreased rate.
I figured that my making a slope of increasing advance as MAP goes up was negating the effect of the RPM based spark correction, so on gslender's advice, I flattened the entire idle area. I also steepened the slope above target RPM, to approximate the slope below target.
The results are even better than the previous settings! I was surprised to see the big RPM surge after AC turned off almost disappeared!
The steady state idle is really smooth. Steady state idle plays within 10 RPM off target.
Data indicates that near the idle area, torque is directly/linearly related to spark. And since idle control is basically balancing all the different torque loads on the engine, we should be operating at this area. 10-12 is the sweet spot for the base idle- this leaves a margin of at least 10 degrees below (to reduce torque) and 10 degrees above (to increase torque). Above 20-21, the torque still increases, but at a decreased rate.
I figured that my making a slope of increasing advance as MAP goes up was negating the effect of the RPM based spark correction, so on gslender's advice, I flattened the entire idle area. I also steepened the slope above target RPM, to approximate the slope below target.
The results are even better than the previous settings! I was surprised to see the big RPM surge after AC turned off almost disappeared!
The steady state idle is really smooth. Steady state idle plays within 10 RPM off target.
Reply
0
0
12-13-2011, 08:26 PM
#16
Elite Member
Join Date: Jul 2005
Posts: 6,420
Total Cats: 84
I thought about how to go about tuning this scientifically- so I did a spark vs MAP test. Set the idle valve to a fixed duty and ran the spark from -4 to 40. Got some interesting data out of it!
Data indicates that near the idle area, torque is directly/linearly related to spark. And since idle control is basically balancing all the different torque loads on the engine, we should be operating at this area. 10-12 is the sweet spot for the base idle- this leaves a margin of at least 10 degrees below (to reduce torque) and 10 degrees above (to increase torque). Above 20-21, the torque still increases, but at a decreased rate.
Data indicates that near the idle area, torque is directly/linearly related to spark. And since idle control is basically balancing all the different torque loads on the engine, we should be operating at this area. 10-12 is the sweet spot for the base idle- this leaves a margin of at least 10 degrees below (to reduce torque) and 10 degrees above (to increase torque). Above 20-21, the torque still increases, but at a decreased rate.
Note that the characteristics you describe will be a bit different when the a/c is loading the engine.
Reply
0
0
12-13-2011, 09:28 PM
#17
I got this installed and it made my car run so much smoother, I was running 3.0.3 before though, so i was unaware that my misfires were because of a firmware related sequential injection problem. I still need to fiddle with the settings to make my idle stay constant when i turn on my fan and lights though. This feature should work even if i dont have ac correct? I have messed with it a little, I put in greg g's settings and didn't notice any effect on the idle droop.
Reply
0
0
12-15-2011, 09:51 PM
#19
Junior Member
Join Date: Jun 2007
Posts: 411
Total Cats: 0
I repeated the spark test at 36.2 idle valve duty (idle range), with more complete results:
Which means is since I have base advance at 12, I can retard up to 22 degrees and still get a linear decrease in RPM, and advance up to 18 degrees with a linear increase in RPM. Notice the rates are linear above and below sweet spot, but the slope is slightly different. You can adjust the idle advance values to reflect this.
Which means is since I have base advance at 12, I can retard up to 22 degrees and still get a linear decrease in RPM, and advance up to 18 degrees with a linear increase in RPM. Notice the rates are linear above and below sweet spot, but the slope is slightly different. You can adjust the idle advance values to reflect this.
Reply
0
0
12-18-2011, 04:19 PM
#20
Junior Member
Thread Starter
iTrader: (1)
Join Date: Jun 2011
Location: Australia
Posts: 178
Total Cats: 3
Now up to v2 (beta5) of my mods.
Check it at >> http://www.msextra.com/forums/viewto...p?f=91&t=43370
This v2 update is a major release to the mods I've previously added and includes significant changes from previous release.
The changes and new features include:
Check it at >> http://www.msextra.com/forums/viewto...p?f=91&t=43370
This v2 update is a major release to the mods I've previously added and includes significant changes from previous release.
The changes and new features include:
- New Memory Page - now ms2 has lots more space for more future mods.
- New MS3 Initial Value Tables (ivt) - in addition to last-known-good (lkg) for idle duty (replaces MAT correction).
- New MS3 PWM voltage correction tables (replaces Volt correction).
- Extra bin and same page (burb fix) for load/rpm idle advanced settings.
- Latest MS3 Bug fixes in AC idle handing
- Change to RPM input smoothing from IAC output smoothing (smoother still)
Reply
0
0
Thread
Thread Starter
Forum
Replies
Last Post
