When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Ignition/Tuner/ECM/Fuel InjectionNeed advice on ignition issues? Questions about a tuner? Have questions about a EFI calibration or Fuel Injection? Tips on Engine Diagnostics, how to get codes, and what they mean. Find your answers here.
I know adaptive fuel factor /value is applied outside of closed loop— I can see it in my logs. What I don’t know is how does it get set outside of CL with narrow bands? I don’t use auto tune, so I know it’s not that.
Adaptive fuel correction is based on the closed loop integrator (short term fuel trim), however adaptive correction has different resolution than the rolling, live, closed loop correction. Adaptive fuel factor is learned correction and stored as a cluster/band of values similar to block learn. Adaptive fuel corrections can/will extend into the open loop regions of the tune.
Adaptive fuel correction is based on the closed loop integrator (short term fuel trim), however adaptive correction has different resolution than the rolling, live, closed loop correction. Adaptive fuel factor is learned correction and stored as a cluster/band of values similar to block learn. Adaptive fuel corrections can/will extend into the open loop regions of the tune.
thanks Jamie or Lucus 😀 But I am still not sure I understand. To make any correction via CLI outside of closed loop goes against everything I have learned. How can the Ecm make an accurate decision with narrow bands outside of nb capability? I guess the same goes for auto tune. Forcing the Ecm into closed loop throughout the afr map does not seem like it could be accurate. At least it’s not for me. I tried 😀
thanks Jamie or Lucus 😀 But I am still not sure I understand. To make any correction via CLI outside of closed loop goes against everything I have learned. How can the Ecm make an accurate decision with narrow bands outside of nb capability? I guess the same goes for auto tune. Forcing the Ecm into closed loop throughout the afr map does not seem like it could be accurate. At least it’s not for me. I tried 😀
CLI is a rolling, real time fuel correction; Adaptive fuel (AFF) learned. Think of them as short term/long term corrections, however as I noted earlier they have different resolution. CLI corrects quickly to maintain the AF/Lambda command, AFF is developed based on what what is being corrected but this is a "trim" is a much broader area, a moderate region of cells. For example if your Lambda table is closed loop to 75Kpa, what is learned at 75Kpa may be extended to 80/90/100 KPa due to the side of each region that is learned, this "region" can be well into the open loop area of the calibration. Also consider the VE, Lambda, and other CL related tables have different resolution and X/Y axis. This works under the intention the calibration is within reasonable accuracy in the first place, and can accommodate for differences such as fuel pressure etc..
CLI is a rolling, real time fuel correction; Adaptive fuel (AFF) learned. Think of them as short term/long term corrections, however as I noted earlier they have different resolution. CLI corrects quickly to maintain the AF/Lambda command, AFF is developed based on what what is being corrected but this is a "trim" is a much broader area, a moderate region of cells. For example if your Lambda table is closed loop to 75Kpa, what is learned at 75Kpa may be extended to 80/90/100 KPa due to the side of each region that is learned, this "region" can be well into the open loop area of the calibration. Also consider the VE, Lambda, and other CL related tables have different resolution and X/Y axis. This works under the intention the calibration is within reasonable accuracy in the first place, and can accommodate for differences such as fuel pressure etc..
Ok. Thanks! This explains what I am seeing. I didn’t know long term was applied across multiple cells. But it does beg the question of why it would apply correction to an area that it doesn’t evaluate.
I’m asking this because I am not sure if I should adjust fuel in an open loop area that the ECM is routinely pulling or adding.
Because the ECM is presuming if given cell(s) are known rich or lean, the adjacent/surrounding area also needs coinciding correction.
Right on Jamie. This is why it’s so important to get the VEs corrected so area changes provide expected results. Setting target AFR will never produce expected results if the VE calibration is incorrect.
Ok. Thanks! This explains what I am seeing. I didn’t know long term was applied across multiple cells. But it does beg the question of why it would apply correction to an area that it doesn’t evaluate.
Adjacent cells do not flow air significantly differently (there are exceptions on the idle 2/5/10% (that's why they added 7% in some cals). Other exceptions are cams that are not correct for the base cal and you get some big differences in adjacent cells but that is not what we are talking about. A really fast way I used to rough in a map I was adjusting manually was to do a 30% and adjust 20-40, and an 80% and adjust 60-100 with the same increases or decreases as the tested positions. With this dynamic, adaptive can be estimated off of adjacent cells. It allows learned correction of a the map based off narrow band input.
Last edited by Ed Ramberger; Oct 14, 2024 at 10:08 PM.
Adjacent cells do not flow air significantly differently (there are exceptions on the idle 2/5/10% (that's why they added 7% in some cals). Other exceptions are cams that are not correct for the base cal and you get some big differences in adjacent cells but that is not what we are talking about. A really fast way I used to rough in a map I was adjusting manually was to do a 30% and adjust 20-40, and an 80% and adjust 60-100 with the same increases or decreases as the tested positions. With this dynamic, adaptive can be estimated off of adjacent cells. It allows learned correction of a the map based off narrow band input.
One needs to be very careful with these things as the Adaptive is really the exact same thing as the industry standard name of "Long Term" The way the Delphi ECM handles it it that when the bike is running in Closed loop and goes into Open Loop it retains the Adaptive value until it returns to Closed Loop Operation. So when you leave Closed loop at 1500 RPM one time and another time you leave Closed Loop at 3000 RPM your going to get different Adaptive Values being used until you return to Closed Loop. Not having the VE values so that that all the adaptive are the same or close to it, raises hell with a tuneup. Think about it like a coarse tune **** on a old radio and a fine tune ****. The Adaptive is the coarse tune ****, and the integrator is the fine tune ****. The amount and size of the adaptive cells varies from base calibration to base calibration within the same model year so there is no one size fits all answer.
7 Surprising Harley-Davidson Products that Are Not Motorcycles
Slideshow: The bar-and-shield logo shows up on far more than motorcycles, some of the company's most unexpected products have nothing to do with riding.
Slideshow: From the troubled AMF years to modern misfires, these bikes earned reputations for reliability issues, questionable engineering, or disappointing performance.
Crazy Bunderbike Build Looks Amazing, But Is It Impossible to Ride?
Slideshow: The Swiss custom shop has taken a Harley Softail and stretched it into something so long and low that it looks closer to a rolling sculpture than a conventional motorcycle.
Engraved Rebellion: Inside Bundnerbike's Glam Rock II
Slideshow: A standard cruiser becomes an intricate metal canvas in the hands of a Swiss custom house known for pushing Harley-Davidson platforms far beyond their factory brief.
Slideshow: Harley-Davidson's challenges aren't abstract; they show up in dropping shipments, shrinking dealer traffic, and strategic decisions that aren't yet translating into growth.