BMW M3 Forum
BMW M3 Forum BMW M3 Gallery BMW M3 Reviews BMW M3 Social Groups BMW M3 Chat M3Forum Sponsors >>
Loading


Mobile M3forum
Go Back   BMW M3 Forum.com (E30 M3 | E36 M3 | E46 M3 | E92 M3 | F80/X) > BMW M3 Discussions > E46 M3 (2001-2006) > Coding and Tuning
Tire Rack Buy Winter Tires Now!
Not a member? Register Now!
Register Gallery All Albums Garage Search Today's Posts Mark Forums Read Calendar FAQ

Coding and Tuning Discuss all avenues of coding and tuning here!


Reply
 
Thread Tools Display Modes
Old Wed, Nov-13-2019, 06:48:09 PM   #1
Drewster
Registered User
 
Drewster's Avatar
 
Join Date: Aug 2014
Posts: 4,033
In the garage:
Reputation: 0 Drewster is on a distinguished road

United States




Default PSA - AFR isn't as important for power as you think


In the video, he uses a dyno to show the torque of the engine as he varies lambda from scary lean (1.15) to pig rich (0.75) - some of you might be surprised by the results based on the posts I've seen
Jump to top Drewster is online now   Reply With Quote
Sponsored Links
Register now and remove these ads
Old Wed, Nov-13-2019, 11:16:12 PM   #2
Blaikenstein
Registered User
 
Blaikenstein's Avatar
 
Join Date: May 2016
Posts: 414
In the garage:
Reputation: 0 Blaikenstein is on a distinguished road
Location: Denver

United States




Default Re: PSA - AFR isn't as important for power as you think

I though AFR was one of the variables people adjust when your getting a custom tune?

Maybe itís more of a engine safety thing when messing with other variables.
Jump to top Blaikenstein is online now   Reply With Quote
Old Thu, Nov-14-2019, 12:25:11 AM   #3
Drewster
Registered User
 
Drewster's Avatar
 
Join Date: Aug 2014
Posts: 4,033
In the garage:
Reputation: 0 Drewster is on a distinguished road

United States




Default Re: PSA - AFR isn't as important for power as you think

Quote:
Originally Posted by Blaikenstein View Post
I though AFR was one of the variables people adjust when your getting a custom tune?

Maybe itís more of a engine safety thing when messing with other variables.
It definitely is a variable adjusted in a custom tune - tuners should be making sure the car is operating at a safe AFR, and they might even make things a tad richer to avoid detonation.

However, enthusiasts and tuners alike often focus on AFR for power.. when it really doesn't have an impact as much as it does for safety/ longevity/ emissions.
Jump to top Drewster is online now   Reply With Quote
Old Thu, Nov-14-2019, 12:54:36 AM   #4
Blaikenstein
Registered User
 
Blaikenstein's Avatar
 
Join Date: May 2016
Posts: 414
In the garage:
Reputation: 0 Blaikenstein is on a distinguished road
Location: Denver

United States




Default Re: PSA - AFR isn't as important for power as you think

Really interesting video

^thanks for the further clarifications, Iím in the s54 for the long haul more than power.
Jump to top Blaikenstein is online now   Reply With Quote
Old Thu, Nov-14-2019, 02:43:39 AM   #5
ctres
Registered User
 
ctres's Avatar
 
Join Date: Jun 2015
Age: 25
Posts: 517
Reputation: 0 ctres is on a distinguished road
Location: Detroit Metro

United States




Default Re: PSA - AFR isn't as important for power as you think

This is a great illustration of the relationship between lambda and torque, thanks for posting this. As you alluded to, this seems to be something that many people still don't understand. This relationship is covered in one of the first few chapters of Heywood's ICE Fundamentals and is very well understood in the industry, yet it seems that many keyboard warriors still struggle with it, so it is very nice to see it demonstrated live.

This type of lambda sweep is done on production engines during the mapping phase and in modern control strategies the data for this curve is actually often stored in the calibration and used for torque estimation. In my experience from referencing numerous production engine calibrations I've never seen more than a 5-6% difference in torque between the peak and the lowest point over this same lambda range, but I suppose there are always outliers. From what I've seen the peak almost always occurs in the range of .88 to .92 lambda, and is relatively immune to different speeds/loads.

It's worth noting that this the relationship between lambda and torque demonstrated here is only characterizing the chemical/reaction/mixing impact of air fuel ratio alone, while all other factors are held constant. The change in torque you demonstrated as lambda is swept is primarily a function of burn rate. What this sweep doesn't demonstrate is that running richer than the lambda sweep 'peak' provides additional charge cooling (particularly on DI engines) which, on engines that are heavily knock limited, can allow you to safely advance the spark from where you would be able to if you ran right at the peak. Advancing the ignition towards MBT phases the combustion earlier allowing the piston to extract more work from the expanding gases, and as a result create more torque while also reducing EGT.

To illustrate, say for example that you have a knock limited engine with a lambda torque peak at .9. Lets say you lose 5% torque by running lambda = .8, given the same spark timing. However, if running at .8 provides enough charge cooling that you can now advance spark timing by 2 degrees, and gain 10% torque, that is an excellent trade off if peak power is your target, as opposed to fuel economy, emissions, etc.

Anyway, I don't have any specific experience on S54 regarding AFR, but now that all the tuning tools are widely available I think it's great that there is interest in topics like this and I would be very interested in hearing others' experience on this.
Jump to top ctres is offline   Reply With Quote
Old Thu, Nov-14-2019, 03:11:44 PM   #6
Drewster
Registered User
 
Drewster's Avatar
 
Join Date: Aug 2014
Posts: 4,033
In the garage:
Reputation: 0 Drewster is on a distinguished road

United States




Default Re: PSA - AFR isn't as important for power as you think

Quote:
Originally Posted by ctres View Post
This is a great illustration of the relationship between lambda and torque, thanks for posting this. As you alluded to, this seems to be something that many people still don't understand. This relationship is covered in one of the first few chapters of Heywood's ICE Fundamentals and is very well understood in the industry, yet it seems that many keyboard warriors still struggle with it, so it is very nice to see it demonstrated live.

This type of lambda sweep is done on production engines during the mapping phase and in modern control strategies the data for this curve is actually often stored in the calibration and used for torque estimation. In my experience from referencing numerous production engine calibrations I've never seen more than a 5-6% difference in torque between the peak and the lowest point over this same lambda range, but I suppose there are always outliers. From what I've seen the peak almost always occurs in the range of .88 to .92 lambda, and is relatively immune to different speeds/loads.

It's worth noting that this the relationship between lambda and torque demonstrated here is only characterizing the chemical/reaction/mixing impact of air fuel ratio alone, while all other factors are held constant. The change in torque you demonstrated as lambda is swept is primarily a function of burn rate. What this sweep doesn't demonstrate is that running richer than the lambda sweep 'peak' provides additional charge cooling (particularly on DI engines) which, on engines that are heavily knock limited, can allow you to safely advance the spark from where you would be able to if you ran right at the peak. Advancing the ignition towards MBT phases the combustion earlier allowing the piston to extract more work from the expanding gases, and as a result create more torque while also reducing EGT.

To illustrate, say for example that you have a knock limited engine with a lambda torque peak at .9. Lets say you lose 5% torque by running lambda = .8, given the same spark timing. However, if running at .8 provides enough charge cooling that you can now advance spark timing by 2 degrees, and gain 10% torque, that is an excellent trade off if peak power is your target, as opposed to fuel economy, emissions, etc.

Anyway, I don't have any specific experience on S54 regarding AFR, but now that all the tuning tools are widely available I think it's great that there is interest in topics like this and I would be very interested in hearing others' experience on this.
Jump to top Drewster is online now   Reply With Quote
Old Fri, Nov-15-2019, 06:39:43 PM   #7
Turb02
Registered User
 
Turb02's Avatar
 
Join Date: Aug 2016
Posts: 417
In the garage:
Reputation: 0 Turb02 is on a distinguished road

United States




Default Re: PSA - AFR isn't as important for power as you think

This is good stuff. Thanks for sharing.

So, when tuning alphaN, is it better to adjust AFR target or the alpha N map?

for example: if youre alpha map has a Z axis parameter of 120 in a cell, but when you log, you see 130 for that cell, is it better to adjust the number to 130 and then lean out the fuel table, or leave the fuel table at target AFR and adjust the alphaN value down to say 113?
__________________

ModeGrau Respray
Active Autowerke Gen3 Stage 2.5 C38-91-AA Headers-Supersprint-Meisterschaft-Vorsteiner-Brembo-AC Schnitzer-BBS-CSL
6MT swapped-RTD shifter-Amber Gang-Dual port Aquamist-Advanced Fuel Dynamics E85-CMP Auto Engineering topside subframe reinforcement-Redish v2 plates
Jump to top Turb02 is offline   Reply With Quote
Old Sat, Nov-16-2019, 01:27:26 AM   #8
ctres
Registered User
 
ctres's Avatar
 
Join Date: Jun 2015
Age: 25
Posts: 517
Reputation: 0 ctres is on a distinguished road
Location: Detroit Metro

United States




Default Re: PSA - AFR isn't as important for power as you think

The correct way to do it is to adjust the Alph-N map. The whole point of the Alpha-N map, or the VE map on a speed density system is to estimate the mass of air that is being trapped in the cylinder. With this information, the control system then references the lambda target table and calculates the mass of fuel that should be injected to meet the target lambda at the current conditions. If the injector flow rate and latency are calibrated correctly, the control system can then accurately convert this fuel mass into an injector pulsewidth. If everything was calibrated correctly the measured exhaust lambda should be close to the target lambda.

You can generally achieve the same results my editing the target lambda table, but that is sort of a fudged solution, because it really just means that something else is wrong in your cal. On a modern control strategy that relies on accurate torque estimation, this approach would result in some really wacky behavior. Also, the other big advantage to doing it the correct way is that if you change something in your injection system i.e. bigger injectors, different fuel pressure, etc. All you need to update in your cal is the injector parameters, and the system will not require any re-tuning.
Jump to top ctres is offline   Reply With Quote
Old Tue, Nov-19-2019, 03:40:10 PM   #9
Turb02
Registered User
 
Turb02's Avatar
 
Join Date: Aug 2016
Posts: 417
In the garage:
Reputation: 0 Turb02 is on a distinguished road

United States




Default Re: PSA - AFR isn't as important for power as you think

Quote:
Originally Posted by ctres View Post
The correct way to do it is to adjust the Alph-N map.
So if the target fuel map had super rich or super lean spots, thats fine, as long as the Alpha-N map was providing enough of a calculation to hit a safer AFR?

In my example above, if the cell that has 120 in it, has a target AFR of 12.2 in the fuel map, are you saying I should adjust the alpha to 130, but lean out my fuel table from 12.2 to something that is "off target" (leaner)?

Forgive me, Im still learning.

Ive been trying to clean up my tune and have set my fuel tables to values that I feel are pretty solid, then log, and determine where Im rich and lean, and have been adjusting the alpha-N map up or down to ensure Im hitting the targets that I set in the fuel map. Is this correct?

There were some areas in my original tune, that were targeting 15.xx+ but the alpha-N values were really high (I would still log a safe AFR, but...), so Im reversing it and normalizing AFR and lowering the alpha-N values, until I see a consistent +-2.5 trims.

Is it common practice for a tune to be like this? Higher alpha-N values AND higher target AFR values? I always ASSumed it should have a "spot on" AFR target, then adjust around that (alpha-N values). Big question for me.
__________________

ModeGrau Respray
Active Autowerke Gen3 Stage 2.5 C38-91-AA Headers-Supersprint-Meisterschaft-Vorsteiner-Brembo-AC Schnitzer-BBS-CSL
6MT swapped-RTD shifter-Amber Gang-Dual port Aquamist-Advanced Fuel Dynamics E85-CMP Auto Engineering topside subframe reinforcement-Redish v2 plates
Jump to top Turb02 is offline   Reply With Quote
Old Tue, Nov-19-2019, 05:41:36 PM   #10
ctres
Registered User
 
ctres's Avatar
 
Join Date: Jun 2015
Age: 25
Posts: 517
Reputation: 0 ctres is on a distinguished road
Location: Detroit Metro

United States




Default Re: PSA - AFR isn't as important for power as you think

The alpha N map is an AIR map, NOT a fuel map. If you calibrate the alpha N table correctly then your measured exhaust lambda will closely match what you put in the desired lambda table ASSUMING that your injector parameters are correct. Does that make sense?

The way I would suggest to tune it is to temporarily set all your lambda targets to a single nice rich value, probably .75 to .8 or so. This will prevent lean issues during tuning that could potentially cause damage to the engine. Then tweak the alpha n table until your measured exhaust lambda is equal (or as close as possible) to whatever you put in the desired lambda table. Once you've acheived this you can go back to your desired lambda table and set it to whatever lambdas you actually want to run, and if you've done everything correctly then you should see that your measured exhaust lambda is the same as or very close to your new targets. Let me know if this isn't making any sense.
Jump to top ctres is offline   Reply With Quote
Reply

Bookmarks

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump



All times are GMT. The time now is 09:33:05 PM.


Powered by vBulletin® Version 3.8.6
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
M3Forum.com and M3forum.net is in no way sponsored, endorsed or affiliated by or with BMW NA / BMW AG or any of it's subsidiaries or vendors.
BMW and M3 (E90 M3 | E92 M3 | E93 M3 | E46 M3 | E36 M3 | E30 M3) are registered trademarks of BMW AG.
M3Forum Terms of Service
Copyright ©1999-2017 M3Forum.com
Discussing PSA - AFR isn't as important for power as you think in the Coding and Tuning Forum - Discuss all avenues of coding and tuning here! at BMW M3 Forum.com (E30 M3 | E36 M3 | E46 M3 | E92 M3 | F80/X)