Drivetrain Software Reset - fact or fiction?
Software Reset - fact or fiction?
I've read that the "software reset" function that can be performed through the odometer followed by several mile of "spirited" driving with lots of WOT improves performance. Likewise, I've read that the same reset process followed by very modest throttle inputs can help boost fuel mileage.
To the contrary, I've also read that the adaptive learning portions of memory can only be reset by the dealer using factory diagnostic software. This would seem to suggest that neither performance nor fuel mileage would be influenced by this software reset.
I'd appreciate it if someone knowledgeable could shed some light on this.
To the contrary, I've also read that the adaptive learning portions of memory can only be reset by the dealer using factory diagnostic software. This would seem to suggest that neither performance nor fuel mileage would be influenced by this software reset.
I'd appreciate it if someone knowledgeable could shed some light on this.
Here's my understanding of this:
The ecu continues to adapt slowly to driving style. If you change your
driving style, it will take a while for the ecu to fully adapt to it. If you
reset the ecu, it starts learning from scratch, and adapts much more
quickly (guessing 30-50 miles as opposed to a couple hundred miles).
If you reset it, drive like you stole it, then revert to granny style, it will
slowly adapt back to granny style. I think this is more like fine-tuning
of the fuel management curves, etc., rather than big differences.
The ecu continues to adapt slowly to driving style. If you change your
driving style, it will take a while for the ecu to fully adapt to it. If you
reset the ecu, it starts learning from scratch, and adapts much more
quickly (guessing 30-50 miles as opposed to a couple hundred miles).
If you reset it, drive like you stole it, then revert to granny style, it will
slowly adapt back to granny style. I think this is more like fine-tuning
of the fuel management curves, etc., rather than big differences.
With only anecdotal evidence to support them, I've always been a bit skeptical myself of the ECM adaptation claims. Reset the ECM and drive the car hard for a while and what - it makes more HP than if you drive it easy for a while? Conversely, if you drive it easy for a while after reset than I suppose it makes more mileage than the hard driving case. This sort of behavior implies that the two regimes, high HP and good mileage, are somehow long-term exclusive. Perhaps I'm missing the entire point of ECM adaptation, (willing to learn however
).
The only somewhat authoritative thing I've seen so far on adaptation is in the Bentley manual.
Page 130-9:
I haven't scanned the entire Bentley manual, but that's the stuff on ECM adaptation. Notice that it is claiming adaptation to environmental changes, not driving style or any long-term effects.
The only somewhat authoritative thing I've seen so far on adaptation is in the Bentley manual.
Page 130-9:
Adaptation. ECM adapts to driving conditions in order to maintain a stoichiometric air/fuel ratio (14.7:1). It is capable of adapting to various environmental conditions encountered while the vehicle is in operation (changes in altitude, humidity, ambient temperature, fuel quality, etc.). These ECM adaptations can only make slight corrections and can not compensate for large changes which may be encountered as a result of incorrect airflow or incorrect fuel supply to engine.
Within allowed adaptation range, ECM modifies injection rate during these states of engine operation:
Within allowed adaptation range, ECM modifies injection rate during these states of engine operation:
- During idle and low load mid range speeds (additive adaptation)
- During operation under normal load to higher load at higher ingine speeds (multiplicative adaptation)
From what I understand, below is the status of the adaptive ecu, adapted from my post in another thread:
1) On a dyno, full throttle, your car should be operating in open loop mode. This means that it should be operating solely based on the open loop parameters, which are coded in the file. It will not be reading the AF ratio from the oxygen sensors and adapting. Hence, adaptive ecus show no difference in _dyno_ power output with adaptation, as some have seen perplexingly.
2) On the road, you're only in this open loop mode when full throttle, so this usually happens just for a short while, if ever. The adaptability comes in here, in the closed loop operation [when your car is constantly reading the AFR's and altering fuel delivery appropriately, see http://www.autoshop101.com/forms/h58.pdf]. As the Bentley manual suggests, the car must learn its operating capabilities for part throttle situations, so as to improve economy and performance across geographies and seasons. It then runs more efficiently and with more perceived power, day to day, in closed loop mode.
But, adaptability is not perfect. ECU Adaptability, also known as "fuel trim," can only allow about plus or minus 20% changes, from what I've read. Therefore, if you put in huge 1000 cc injectors, your car can't throttle back enough, but if you've got slightly bigger injectors, it can, but only in closed loop [part throttle]. Open loop [full throttle] reads right off of the original ecu program.
Extra reading: http://www.hondata.com/techclosed.html
Hope this helps!
1) On a dyno, full throttle, your car should be operating in open loop mode. This means that it should be operating solely based on the open loop parameters, which are coded in the file. It will not be reading the AF ratio from the oxygen sensors and adapting. Hence, adaptive ecus show no difference in _dyno_ power output with adaptation, as some have seen perplexingly.
2) On the road, you're only in this open loop mode when full throttle, so this usually happens just for a short while, if ever. The adaptability comes in here, in the closed loop operation [when your car is constantly reading the AFR's and altering fuel delivery appropriately, see http://www.autoshop101.com/forms/h58.pdf]. As the Bentley manual suggests, the car must learn its operating capabilities for part throttle situations, so as to improve economy and performance across geographies and seasons. It then runs more efficiently and with more perceived power, day to day, in closed loop mode.
But, adaptability is not perfect. ECU Adaptability, also known as "fuel trim," can only allow about plus or minus 20% changes, from what I've read. Therefore, if you put in huge 1000 cc injectors, your car can't throttle back enough, but if you've got slightly bigger injectors, it can, but only in closed loop [part throttle]. Open loop [full throttle] reads right off of the original ecu program.
Extra reading: http://www.hondata.com/techclosed.html
Hope this helps!
Good thread - thanks for asking this!
I've wondered about this myself. Andy recently said in another thread that any mods you make will show up on the dyno right away - no need to reset the ECU. Now, with Ingsoc and Bradley99's explanations, some of this stuff is starting to make sense. When you run at WOT (like on a dyno), you bypass the adaptive stuff. It's starting to make sense!
But I'm still wondering about "driving style." When I first got Tillie broken in, her throttle felt really touchy. Even a little movement of the gas pedal and BAM! - lots of power online. It seemed very difficult to transition from cruising to moderately hard acceleration. The gas pedal seemed almost like it was either on or off. Three-quarters throttle didn't seem to exist.
A year later, the car seems much more manageable. Is it just me getting used to driving such a fast car? (The MCS is definitely the fastest car I've ever owned.) Or did the car somehow adapt to me?
I've wondered about this myself. Andy recently said in another thread that any mods you make will show up on the dyno right away - no need to reset the ECU. Now, with Ingsoc and Bradley99's explanations, some of this stuff is starting to make sense. When you run at WOT (like on a dyno), you bypass the adaptive stuff. It's starting to make sense!
But I'm still wondering about "driving style." When I first got Tillie broken in, her throttle felt really touchy. Even a little movement of the gas pedal and BAM! - lots of power online. It seemed very difficult to transition from cruising to moderately hard acceleration. The gas pedal seemed almost like it was either on or off. Three-quarters throttle didn't seem to exist.
A year later, the car seems much more manageable. Is it just me getting used to driving such a fast car? (The MCS is definitely the fastest car I've ever owned.) Or did the car somehow adapt to me?
Fact and Fiction
There is no 'software reset', but you can reset the adaptations and smooth running values through a BMW GT1 or Displus.
When you do the cluster reset all you are doing is a CLUSTER RESET. In fact, if you don't turn the car off for ten seconds before restarting it, you might notice that the tachometer is incorrect.
When you do the cluster reset all you are doing is a CLUSTER RESET. In fact, if you don't turn the car off for ten seconds before restarting it, you might notice that the tachometer is incorrect.
Trending Topics
Thanks for all the info.
Taken together, what you all contributed makes sense to me.
If anybody has solid information on the rate of adaptability, that would be greatly appreciated. For example, if it is consistently driven in a manner that would cause it to optimize for performance, and the driving style changes from spirited to conservative, how long would it take to adapt to a mapping that would optimize for fuel mileage?
If anybody has solid information on the rate of adaptability, that would be greatly appreciated. For example, if it is consistently driven in a manner that would cause it to optimize for performance, and the driving style changes from spirited to conservative, how long would it take to adapt to a mapping that would optimize for fuel mileage?
Okay, so I understand that the computer is continiously adjusting itself according to the values coming from the sensors in every case except WOT. to figure out everything it needs from the environment. Why do people believe that it takes 50 miles or even hundreds of miles for the computer to "adapt?" Isn't this a myth? Won't the computer figure out the proper air/fuel ratio in a matter of seconds once the engine is and running and fine-tune any slight changes from then on? And also, isn't there flawed logic to think the computer adapts to performance if you're accelerating more rapidly or for ultimate fuel efficiency if your just cruising? Isn't the computer ALWAYS trying to figure out the most efficient air/fuel ratio regardless of how aggressive or tame you are driving? It's a computer.. what does it know what it is to be aggressive or tame, it just needs to keep the engine running the best it can?
Antranik, You are correct. As soon as you start the car, the computer sends and receives voltages from various sensors, telling it what and where to compensate. And yes, it is constantly trying to maintain the ideal air/fuel ratio of 14.7 to 1.
002, is also 100% correct. A BMW GT1 is the only way to completely reset the the ECU. The cluster reset just resets the cluster.
002, is also 100% correct. A BMW GT1 is the only way to completely reset the the ECU. The cluster reset just resets the cluster.
Thread
Thread Starter
Forum
Replies
Last Post
R50/53 Help please.. Odd issue.
PsychoRallye
R50/R53 :: Hatch Talk (2002-2006)
21
Feb 17, 2019 06:08 AM
fkrowland
R50/R53 :: Hatch Talk (2002-2006)
5
Sep 30, 2015 10:30 AM
simo711
R52 :: Cabrio Talk (2005-2008)
3
Sep 29, 2015 11:53 AM






. (let the modifications ensue...)
