Jump to content

How long does it take for the Motronic to "learn"?


Willie

Recommended Posts

I am in the middle of trying out a few different types of plugs, with and without the CCP installed. I do the fuse pull, wait, twist 3x deal every time I do anything with the CCP, but I remember someone saying before that the Motronic takes awhile to learn how the run the bike even after doing the fuse and twist thing. How long does it take to learn and what exactly does it learn? I am wondering now if my testing will really mean anything if the bike is going to change how it runs in a few hundred miles. Any help? Terry

Link to comment

Terry I’m not sure what the learnability capabilities of the Motronic 2.4 system are.. I seems it at least has that basic learning ability of the TPS’s WOT & idle threshold (I would imagine that would be learned from that first key on throttle twist from non choked base idle to WOT).. As far a basic spark table learn, it might, but without the vehicle having a spark knock sensor or individual cylinder spark control it is doubtful there would be much to learn or even if it would make a noticeable difference if it did.. There could be some basic fuel cell learn (like the Delphi Block learn or integrator cell learn, but if did, that would probably be a fast learn for the basics (probably under 10 miles of consistent riding)..

 

Again, I’m not sure of it’s (2.4 system) learning capability but if it does in fact have other than TPS learning capability it wouldn’t take many miles to at least get close to getting it right.. You would just need to ride it for the first few miles like you plan on riding it as usual… If it does have other than basic TPS learning & you horse it hard right after the Motronic Re-set it might have some rough spots in the program until it settles out & re-learns the way you usually ride it..

 

It’s a pretty basic system so (my guess here) probably only uses fixed spark & fueling tables based on RPM, throttle position, & load, with the maps modulated a bit by engine temperature & intake air temperature..

 

Twisty

Link to comment
ShovelStrokeEd

Actually, Ken, I think there is more than that. Over the longer term, the bike seems to "learn" your riding habits. Nearly the same thing on my Honda. The most noticable thing is in the fuel consumption. This applies to both my 1100S and the Blackbird. If I have been doing a lot of around town riding, fuel consumption runs in the low to mid 30's for both bikes. Off on a trip and the fuel consumption does not instantly improve. Takes about 3 tanks worth to optimize. First tank will be about 35 to 36 mpg, second will be right at 39 and the third will get me 40-42 on the S and about 40 on the bird. Has been this way for over 50K miles on the S and 35K on the 'bird. IOW, over a course of 500 miles of freeway style riding, 84 mph moving average, the bike picks up nearly 8 mpg due to operation at near constant throttle for long periods.

 

I'm not sure how all this works as really the RPM range of operation is about the same, around town 3.5K is the norm, on the freeway 5K. If anything, I would think the extra aerodynamic drag would worsen fuel milage but it just does the same thing, time after time. I would have to think that the Motronic or Honda PGM looks at rate of change and frequency of change of throttle position and modifies the map richer or leaner to improve drivability and economy.

Link to comment

I do not have the answer, but if like many computerized vehicles with a learning prom it should be somewhere around 11 miles or so, what they actually do is relearn the driving habits constantly that way.

 

and it has been a while things may have changed

Link to comment

Actually, it was one of Ed's previous answers to an earlier post that reminded me of this. But he is not the only one who believes this though. One of the local dealers also told me that the bike will learn over a period of time how you ri de it and adjust accordingly. Now, I don't know if it is standard procedure to do the fuse pull throttle twist thing or not at the dealership level, but maybe if they knew about that technique, it would take care of him telling me that it does take quite a few miles for the bike to learn. I'm just grasping at straws right now out of frustration about this surge characteristic. I just did my valves again (they were good, but I still tweeked them more) and did my TBS with both a water manometer and a Twinmax, and I'm still having problems. I want to get my bike back to were it was between 12k and 18k where it was virtually surge free. The only thing I have to do yet to get it back to the same configuration is to pull the CCP and install Bosch 4418's. If that doesn't work... Techlusion here I come. The most frustrating part of this whole thing is that I had the surge tuned out, but now it's back and I can't shake it. In all honesty, it is probably mild compared to some bikes, but it bothers me. Also, it seems to be over a broader range in the rpm's. I checked for intake leaks when I had it apart by spraying carb cleaner, but will probably try again using a propane bottle. I do notice sometimes when revving the engine that on the downward fall of the rpm's that it hangs at about 1400 for just the blink of an eye. Sound like an intake leak or anything to you? I would thing from past experience that a leak you hang up the r's longer over a wider range. This is just a VERY short hesitation getting back to idle. Other than the surge, the bike idles and runs extremely smooth and since I have the CCP installed again, fires right up and even holds a steady idle with the fast idle lever off on a cold engine. I'm running out of ideas. The Techlusion sounds like the next step for me. So, anymore thoughts as to how long it takes for the computer to figure things out. I have been riding about 10-20 miles before doing changes to the plugs or CCP, so I hope that is enough to let the bike figure out exactly what it is doing. Believe it or not, now I am having the best results with the CCP installed using a set of stocker NGK plugs that where in the bike when I bought it. Tomorrow I have off work and might ride down to Bob's with the CCP out and see if the surge is still noticable. If not, back to 4418's and then the Techlusion I think. Thanks for all the help, Terry

Link to comment

Terry, first off you need to keep in mind,, what’s there to learn.. On the simple Motronic 2.4 without wide band 02 & without a spark knock sensor, not very much.. Secondly anything it would need to learn (assuming it could) would have to be learned pretty darn fast as your riding habits, weather, engine temps, fuel type, rider, carried weight, terrain, etc, are CONTINUALLY changing from day to day, hour to hour, minute to minute, even second to second…

 

Now what could it learn & when does it use what it learned.. Keep in mind here that I am not an expert (or even close) on the Motronic 2.4 system but I have worked with many ECM. PCM. & a multitude of power-train controllers over the years.. Most of the later units have a basic learning ability of the parameters of the external sensors like TPS learn.. Those are usually learned on initial key-up & don’t change rapidly or even at all.. Those are usually used to eliminate a precise setting by the initial installing assembly worker or allow adaptation to different usages of the same parts, or to allow a higher tolerance to mis-adjustment or wear related calabration change..

Next would be the longer term learned things like spark table cells or fueling cells.. As a rule those depend on something like moving throttle, changing load, spark knock sensor input, wide band 02 return input.. Without a spark knock sensor or wide band 02 the basic Motronic 2.4 wouldn’t get enough return input to even be able to learn much.. In most all the power-train computers I have worked with the learning ability only pertains to changing throttle or changing load in relation to the spark tables or initial fueling cell selection during throttle position change.. At steady state throttle with a narrow band 02 & in closed loop there is nothing the Computer can do anyhow but react to the 02 signal even the spark tables would be fixed as there is no knock sensor to allow it to react to that area..

 

You seem to be only chasing a surging problem at the moment.. That is usually ONLY in closed loop (don’t believe that just disconnect the 02 sensor & ride the bike).. There is no amount of learning by anything in that computer that can change the closed loop (under 15% throttle) fuel air ratio.. Basically it uses fixed spark tables (no way to be creative there) & relies on the 02 to tell it to go rich or lean from where it is running at the moment (pretty well runs along at 14.7:1 fuel air ratio regardless).. About the only thing you can change in this (closed loop) area are the spark tables & that would be effected by trying a different CCP that forces a different set of spark tables.. About the only other option here is to pull the CCP completely or disconnect the 02 as that would cause it to use open loop spark & fueling tables modulated by air temp input & engine temp input..

 

Again, in closed loop all fueling duties are controlled by the queen (that’s that little lambda sensor).. If your surge is in closed loop (probably is as most are).. You just can’t learn around or effect closed loop with any (Computer learning), (TPS adjustment), (sensor cheating),(ETC).. No matter what you do or change it will go right back to 14.7:1 --IF IT STAYS IN CLOSED LOOP.. Now you can cheat a sensor, run the TPS out of a usable range, otherwise change something & that can effect the fuel air ratio but if it isn’t around 14.7:1 then it isn’t using the 02 sensor so WON’T be in closed loop anymore- PERIOD.. Like I said above about all you personally have control of in close loop are the spark tables..

If you have the ability to make some internal computer changes you could speed up or slow down the rate at which the computer processes the 02 input so the reaction time to the 02 could be changed & that could effect the surging.. If the computer takes too long to react to an 02 crossover change it could allow the fuel/air ratio to get way out of kilter before correcting & that in itself could cause a surge as the system over reacts to try & compensate.. (in fact that could be part of the surging problem on the 2.4 system for all I know)..

 

Now to your surging problem.. If it surges while in closed loop it means that the engine just doesn’t like the lean 14.7:1 fuel air ratio in closed loop (mechanical problem, or unbalanced side to side, or not enough spark to properly propagate ignition of a lean mixture).. OR, the 02 is getting lazy & won’t properly react to the change in oxygen content in the exhaust gas (that’s possible if your bike used to not surge).. Or the fuel you are using now has a different ignition property so makes the engine more susceptible to a lean/run condition..

 

There seems to be some speculation & even some rather compelling believable data that using this spark plug or that spark plug can effect or eliminate surging.. That’s possible but I haven’t shown any change on my personal 1150RT.. Even two people using the same wazzo?? spark plug get different reactions.. Possibly they also did some other change or adjustment at the same time & that was the improvement,, or the spark plugs got indexed differently than the originals when they were installed.. On my Harley I can pick up well over 1 horsepower just by pointing the open end of the spark plug electrode gap towards the intake valve..

 

One other thing to consider when chasing a surge.. If you aren’t actually looking for it, or become sensitized to the surge area of operation it is easy to drive right through it & not really notice it.. I have a friend that thought his bike was pretty decent when it came to the Boxer surge.. I rode his bike & it was worse than mine.. He just didn’t notice it much the way he rode the bike..

 

Twisty

Link to comment

Well... I's sure it adjust to current conditions, MAP, air temp, O2 sensor input, etc., and that's going to take a bit of time from a baseline reset. But I've never seen any documentation that it is adaptive to user input. E.g. - Changes fuel delivery map to suit the style of the rider, etc.

Link to comment

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...