Motronic Long Term Fuel Trim

[roger 04 rt]

As part of my Wideband O2 project I tried to find out if the Motronic MA 2.4 makes use of a Block Learn Term or as it is also called, a Long Term Fuel Trim. The idea of this trim is that it is computed in Closed Loop operation by comparing the Fueling Tables to the actual fuel needed to toggle lean/rich around 14.7:1. Many ECUs (perhaps most) use this concept to correct for fuel (ethanol), wear, battery voltage, injector wear, etc.

 

One thing I know it learns is TPS settings which we have to run closed to open a couple times after a battery or fuse replacement. My question is does anyone have certain knowledge of whether something like a Long Term Fuel Trim is calculated and saved?

 

Many Thanks,

RB

[dan cata]

I doubt it If it would have done so, that info would have been all over the web

 

Dan.

[TestPilot]

Roger,

 

You might send an Email to the folks at Hex Diagnostics ( Hex) , the people that build the GS911. They seem to be the experts on how the various ECUs perform.

[dirtrider]

Morning Roger

 

 

That's a good question.

 

Given it is dynamic system that steps into & out of 02 control many times per operational cycle it would almost have to have some type of fuel trim learning. (at least short term)

 

I don't have the tools at home to verify just how much block learn fuel trim (adaptives in BMW talk) the motorcycle version of the Ma 2.4 uses but when I was doing a lot of tuning on the 1150 Ma 2.4 system you could feel the difference in light throttle tip in & out fueling as the system learned fueling offsets.

 

What I can't say is if the block learn was all short term or a combination of volatile short term & some sort of longer term also.

I had a switch rigged up on my 1150 2.4 system to kill the Motronic's power to clear the block learn & give me a clean learn table but I was unable to tell if it was only a short term dynamic learn that went away with each fresh engine start or a combination of short & long term.

 

I worked with a different company's fueling computers during the period of non sequential injection & the software I used would not talk to the Ma.2.2 or Ma 2.4 BMW system (even on the automobile side).

 

I had a friend that worked in Germany for Opel & they at one time did a lot of work with the Motronioc 2.4 systems. In some of the 2.4 documentation he sent me it showed the automotive side did use a fueling adaptive but didn't explain if it was all volatile or a combination of volatile & long term.

 

What I found when I was working with the Ma 2.4 system is that very little info was/is available on the motorcycle side with just a bit more on the automotive side. Thing I could never figure out was how close the motorcycle 2.4 was in operation to the automotive 2.4.

 

You might try corresponding with the GS-911 people as they must have some operational knowledge of the motorcycle 2.4 system & it's use of adaptives.

 

[roger 04 rt]

I doubt it If it would have done so, that info would have been all over the web

 

Dan.

 

Hey Dan,

At first I doubted to too. But most every ECU that has a detailed following of modifiers, if you dig enough, comments in some way on the Short and Long Term Trim.

 

Here is the familiar language from the RT Service Manual:

 

Switch on the ignition.

• Without starting the engine, fully open the throttle

once or twice so that the Motronic control unit

can register the throttle-valve positions.

L Note:

Disconnecting the battery deletes all entries (e.g.

faults, settings) stored in the Motronic control unit’s

memory.

Loss of settings can temporarily impair the operating

characteristics when the engine is restarted.

 

Since the TPS settings have just been resorted, one wonders just what other settings they're talking about, fuel level?

[roger 04 rt]

Roger,

 

You might send an Email to the folks at Hex Diagnostics ( Hex) , the people that build the GS911. They seem to be the experts on how the various ECUs perform.

 

I think this is a good idea, e-mail sent, waiting to hear back from them. I've also written to the Bosch DME group on Yahoo.

 

All GM products, most other Motronic products and even Megasquirt produce Long and Short Term Trims. Given the ease of calculating the term, it's hard to believe that it doesn't exist in the bikes.

[Boffin]

On a very subjective level, I have noticed a difference in my bike between normal day-to-day riding - a mix of commuting and the occasional 250-mile day - and touring.

 

When I tour, I ride more sedately, with gentler acceleration and lower top-speeds. For the first day or two, the bike uses around the same amount of fuel and is a little rough at lower RPM. By the end of a week, the bike runs smoother and uses less fuel. Then, when I return to commuting etc, the bike seems less-responsive and eager for the first day or two.

 

As I said, very subjective and no hard data, but suggestive of some long-term learning.

 

Andy

[roger 04 rt]

Morning Roger

 

 

That's a good question.

 

Given it is dynamic system that steps into & out of 02 control many times per operational cycle it would almost have to have some type of fuel trim learning. (at least short term)

 

I don't have the tools at home to verify just how much block learn fuel trim (adaptives in BMW talk) the motorcycle version of the Ma 2.4 uses but when I was doing a lot of tuning on the 1150 Ma 2.4 system you could feel the difference in light throttle tip in & out fueling as the system learned fueling offsets.

 

What I can't say is if the block learn was all short term or a combination of volatile short term & some sort of longer term also.

I had a switch rigged up on my 1150 2.4 system to kill the Motronic's power to clear the block learn & give me a clean learn table but I was unable to tell if it was only a short term dynamic learn that went away with each fresh engine start or a combination of short & long term.

 

I worked with a different company's fueling computers during the period of non sequential injection & the software I used would not talk to the Ma.2.2 or Ma 2.4 BMW system (even on the automobile side).

 

I had a friend that worked in Germany for Opel & they at one time did a lot of work with the Motronioc 2.4 systems. In some of the 2.4 documentation he sent me it showed the automotive side did use a fueling adaptive but didn't explain if it was all volatile or a combination of volatile & long term.

 

What I found when I was working with the Ma 2.4 system is that very little info was/is available on the motorcycle side with just a bit more on the automotive side. Thing I could never figure out was how close the motorcycle 2.4 was in operation to the automotive 2.4.

 

You might try corresponding with the GS-911 people as they must have some operational knowledge of the motorcycle 2.4 system & it's use of adaptives.

 

Morning DR,

 

Interesting, it's hard to believe that such terms don't get created but I have searched once a day for a week and have found very little specific info on the MA 2.4 or 2.2. As I mentioned earlier, the manual strongly suggests that there are several terms that need to be relearned. I need to think of an experiment to prove it one way or the other.

 

Given the ease of calculating these terms once you have an O2 sensor:

 

As the Motronic goes into Closed Loop it would start with the pulse duration value for a given RPM and TPS in the stock tables, call that value X. Then in good closed loop you know that to trip the O2 lean takes Y and to trip the O2 rich takes Z, you can say that X should equal (Y+Z)/2. From that you can take the ratio (Y+Z)/2 : X and come up with a multiplying factor. With a few more RPMs and TPSs you could compute a long-term trim factor. Or a more appropriately complicated version of something like that.

 

Such factors would handle low battery voltage which makes the injectors less responsive, changes in fuel stoichiometric ratio, injector performance, air cleaner condition, small air leaks, etc.

 

Thinking further about this, since you can add an AIT (air inlet temp) modifier, run E10 fuel, or in my case shift Closed Loop, there is some kind a a correction factor being produced.

 

Once I can get a better idea of the whats and whens of these corrections, I think there are some implications that are worth discussing about how the R1100s and R1150s run, especially the "lean surge" issue.

 

RB

 

(Okay here is one of the thoughts, hypothetical as it is: If I set my O2 sensor 3% rich to Lambda=0.97, then lock closed loop, it has to run 3% richer than the fuel table to stay locked, which it does. I'm running E10 fuel. Because the fuel is 4% leaner, the O2 toggle must be 4% richer. So now, the fuel table has been richened by 7% from stock. My O2 Closed Loop is locking, so this is happening. More later after I know more.)

 

(Or ask oneself what would happen if the stock Lambda sensor aged to the point where it sent a rich signal (less O2) easily but a lean signal (more O2) with difficulty.)

 

Fun stuff.

[roger 04 rt]

On a very subjective level, I have noticed a difference in my bike between normal day-to-day riding - a mix of commuting and the occasional 250-mile day - and touring.

 

When I tour, I ride more sedately, with gentler acceleration and lower top-speeds. For the first day or two, the bike uses around the same amount of fuel and is a little rough at lower RPM. By the end of a week, the bike runs smoother and uses less fuel. Then, when I return to commuting etc, the bike seems less-responsive and eager for the first day or two.

 

As I said, very subjective and no hard data, but suggestive of some long-term learning.

 

Andy

 

Yes, and the manual says that it will run differently after reset.

 

I realized yesterday that I can program my wideband LC-1 to 450 mV which will make the Motronic think that the O2 is unplugged. So my plan is to cruise a test course with O2 lock at 14.2, confirming with the wideband gauge that is part of the LC-1.

 

Then set the O2 to 450 mV and cruise the same course and note the AFR. This will give me an Open Loop "adapted" reference since it has been a while since I reset the Motronic.

 

Then reset the Motronic, with the O2 still set at 450 mV so that I will remain Open Loop. Cruise the course, note the AFR.

 

Finally, just to confirm the effect of the BoosterPlug in Open Loop, I will remove it and note the AFR over the same course.

 

[Peter Parts]

Can somebody explain this new learning, please.

 

What are the referents? How or why or when does the ECU decide to.... I am not sure...

 

Thanks.

 

Ben

[roger 04 rt]

Can somebody explain this new learning, please.

 

What are the referents? How or why or when does the ECU decide to.... I am not sure...

 

Thanks.

 

Ben

 

Got out and made a few test runs on Saturday so here is my 2 cents worth of thinking at the moment, and with apologies if any of it seems controversial ...

 

Since Bosch is relatively secretive about its Motronic product, it has been like pulling teeth to get any information about the simple question: When do the Motronics MA 2.2 and 2.4 apply Closed Loop fueling adaptations to Open Loop fueling?

 

Although I'm still slowly looking into this, all the written (Bosch articles, handbooks and patents) and experimental evidence says to me that it is much more likely than not that Motronics with O2 sensors use what they learn in Closed Loop operation to adjust Open Loop fueling.

 

As the Motronic learns how much more or how much less fuel it takes at each RPM/TPS position to reach Lambda=1, it is making small corrections for fuel type (e.g. E10), airflow, fuel-flow, cylinder volume, battery voltage, etc. With my programmable-for-NarrowBand LC-1, I can reset the target air:fuel ratio and watch the Motronic adjust--I can see that happen for sure.

 

When I let the motorcyle adapt to a new AFR, and then force the motorcyle to Open Loop (through programming) the motorcyle runs much differently and the AFR displayed by the LC-1 wideband channel confirms this being either richer or leaner than before. Then if I reset the Motronic by pulling its fuse and waiting a while, the AFRs are different still. (There are more things I want to try.)

 

So opposed to the many positive factors from Adaptive Fuel trimming, here are some possible negatives:

 

1. If your O2 sensor has aged and it is not efficient at registering lean mixtures, the Motronic will over-lean the mixture, learn that fact, and then lean out the Open Loop as well.

 

It is hard to test an O2 sensor for accuracy. I'm not sure how any of us would easily do that. You can't tell how it reacts to lean or rich and you can't tell if it is 14.7, 14.5 or 15.1:1.

 

2. If you add an aftermarket product to adjust fueling, it seems likely that at first you would see a good improvement and over time the improvement might be "adapted" out. However, when you disconnected it things would get worse for a while (until the Motronic readapted) so one might be convinced that the aftermarket product was indeed helping.

 

3. If you disconnect the O2 altogether, you might get a good result or a less than good result depending on how the stock fuel tables matched the specific tolerances of your motorcycle.

 

The results of disconnecting the O2 might get better or worse if the Motronic was reset--meaning the Adaptation factors that had been previously stored while in Closed Loop were erased. I don't know.

 

One thing I know for sure is that my particular R1150RT runs much stronger (not that it ran bad before) at an accurate Lambda=0.97 (AFR 14.2:1) than it did with the stock O2 sensor. What I don't know yet is just what that stock O2 sensor was doing. I will eventually go back and look at that.

 

[Peter Parts]

More good information. Thanks.

 

Here's my question again: when or why or what causes, triggers, or alerts the ECU to change the A/F map? Just how does it know it is time to change the map? What happens to trigger that change in the map stored value?

 

For sure, the ECU stores the anchor values of the TPS. Simple. It just says,"What is the highest voltage the TPS has produced? What is the lowest?" and that determines the anchor points albeit in volatile memory.

 

But what makes it say, "Gosh, my map is too lean at 45 degrees of butterfly and 6000 rpm"?

 

Gets weirder when you try to explain "learning" ethanol or changes during a long trip, or other testimony above.

 

... and that's why I think the 2.4 doesn't learn much. MotoOne thinks it does.

 

Ben

[roger 04 rt]

More good information. Thanks.

 

Here's my question again: when or why or what causes, triggers, or alerts the ECU to change the A/F map? Just how does it know it is time to change the map? What happens to trigger that change in the map stored value?

 

For sure, the ECU stores the anchor values of the TPS. Simple. It just says,"What is the highest voltage the TPS has produced? What is the lowest?" and that determines the anchor points albeit in volatile memory.

 

But what makes it say, "Gosh, my map is too lean at 45 degrees of butterfly and 6000 rpm"?

 

Gets weirder when you try to explain "learning" ethanol or changes during a long trip, or other testimony above.

 

... and that's why I think the 2.4 doesn't learn much. MotoOne thinks it does.

 

Ben

 

A very basic answer to your question would go like this:

 

The motorcycle is warmed up and the TPS says 20 degrees (roughly steady) and the RPM is 3000. The Motronic says to itself, this is a good time to try Closed Loop operation and drive the AFR to 14.7:1 plus or minus.

 

So the Motronic first goes into a Closed Loop routine and tries a value from the fueling table, expecting to read either 1 volt or 0 volts from the O2 sensor, indicating richer or leaner than 14.7. Depending on the answer it gets it tries a richer or leaner value by a few percent. Let's say the first answer it got was 1 volt--a rich mixture. It would then take small steps to "lean" the mixture until the answer from the O2 sensor was 0 volts. Then it would wait for a fraction of a second in the lean position. Next it would take small steps to make the mixture richer until it read 1 volt. It would then hold that value for a fraction of a second. Next going to the lean state, then to the rich state and so on.

 

After that process, the Motronic will have a lean value and a rich value that it can compare to its first try from the stock fuel table. If, for example, the mid point between the rich and lean value it just "learned" is 4% richer than the stock fuel table, it can store a number that it richens all future fuel table entries by that amount--whether open or closed loop.

 

That is a rough idea of how it "learns". The next time the motorcycle is started, it can have these correction factors in memory. (Just as with the TPS which it "learns" while you rotate the throttle.)

 

 

 

 

 

 

Here is a simplified formula for one type of ECU:

 

BPW = BPC*MAPP*T*F33C*BLM*DFCO*DE*CLT

 

BPW = How long to turn on the injector

 

BPC = Basic Pulse Width (value from TPS/RPM table)

MAPP= Measured Air Pressure (value from lookup table)

T = Measured Air Temperature(value from lookup table)

F33C= Measured Battery voltage compensation

 

BLM= Block Learn Term (what the ECU has learned from the O2 sensor)

DFCO= Decel Fuel Cutoff Term (when to turn fuel off on decel)

DE= Decel enleanment term (when to lean instead of cutoff)

CLT= Closed loop correction term (from O2 sensor learning)

 

So the BLM and CLT values are "learned" from the O2 sensor and can be applied proportionally.

[Peter Parts]

Roger - you are a worthy member of the group of people who turn to our bikes for intellectual stimulation as well as other stimulation.

 

But your explanation, if you will excuse my joshing, seems to require a little man inside the ECU to make decisions. I suppose it could be a woman as well, but she'd have to be real small. Or both.

 

I ask my question in all sincerity and I suspect there is a good answer, as least in theory and at least for far more sophisticated ECUs, but maybe for the 2.4. But I think the issue is "the referent" or criterion the ECU uses to say "gotta be longer spritzs".

 

After all, if it could be done automatically by the ECU itself you could just ship bikes with no map at all and you wouldn't need megabuck map development time at the Nurburgring.

 

And you wouldn't need different color code plugs for ethanol in Fiji or Iowa because the little couple inside the ECU could figure it out.

 

Ben

[TestPilot]

Roger - you are a worthy member of the group of people who turn to our bikes for intellectual stimulation as well as other stimulation.

 

But your explanation, if you will excuse my joshing, seems to require a little man inside the ECU to make decisions. I suppose it could be a woman as well, but she'd have to be real small. Or both.

 

I ask my question in all sincerity and I suspect there is a good answer, as least in theory and at least for far more sophisticated ECUs, but maybe for the 2.4. But I think the issue is "the referent" or criterion the ECU uses to say "gotta be longer spritzs".

 

After all, if it could be done automatically by the ECU itself you could just ship bikes with no map at all and you wouldn't need megabuck map development time at the Nurburgring.

 

And you wouldn't need different color code plugs for ethanol in Fiji or Iowa because the little couple inside the ECU could figure it out.

 

Ben

 

Actually the ECU can figure it out, but it has to have a basic map to at least get the engine running without destroying itself. The Autotune option that Power Commander offers does just what you propose.

 

I doubt if you will ever discover exactly what criteria Bosch or BMW use to determine long-term fuel trim values. They're not about to release the basic code for their ECU software.

 

Long-term fuel trim first came about as a result of the automotive OBD-II regulations, to keep emissions within reasonable levels as sensors wore out and systems slowly degraded. Since motorcycle emission rules are rapidly catching up with automobile rules, especially in Europe, it's not surprising that the same techniques are used.

[NJNeal]

At the end of the day, in my practical experience, the Motronic doesn't seem to work that well. I've owned my 1150RT for about five months now and spent most of that time trying to solve for lean surging. I can feel the ECU making the changes as it searches for the right AFR. It's annoying as hell.

 

As some posters on this list know, I've tried ALL the suggestions, and installed a Techlusion. None of it really worked that well.

 

I finally, and very, very reluctantly just pulled the CCP, and that's given me, so far, the best balance of ride quality and MPG. I have no idea what damage I may be doing to the CAT, and really don't care...at least the bike is ridable again.

 

I was actually starting to shop around for other bikes as a result of my experience on this RT, and I learned that this lean surging is common among a lot of bikes between about 2000 and 2006. Ducati, Harley, Guzzi...many makers had this issue.

Neal

[dirtrider]

Evening Karl

 

Long term fuel trim as well as short term fuel trim learning has been around a lot longer than OBD-II.

 

Most automotive electronic fuel injection systems have had it since the 02 sensor was incorporated.

 

On the Delco/Delphi system "block learn" Multiplier as the long term fuel trim & "integrator" as short term has been used since the 1980's.

 

[TestPilot]

D.R.,

 

Thanks for the correction.

[dirtrider]

---

I finally, and very, very reluctantly just pulled the CCP, and that's given me, so far, the best balance of ride quality and MPG. I have no idea what damage I may be doing to the CAT, and really don't care...at least the bike is ridable again.

 

-

 

Evening Neal

 

Removing the CCP on the 1150 2.4 systems does some strange things. Unlike the 1100 2.2 system it doesn't force open loop but just forces a very crude underdeveloped basic closed loop map with a poor upper mid range spark mapping.

 

You might try reinstalling the OEM CCP as that will bring the good spark curve & snappier mid upper range back then unplugging the 02 sensor as that should kill off most of your light throttle lean surging. Be sure to remove the Motronoic fuse for about 30 seconds after 02 disconnect as that will re-set the Motronic learned adaptives & give you a clean slate to start out with. Then after Motronic fuse reinstall do a new TPS re-learn.

 

[NJNeal]

Thanks. Will try that.

Do you need to plug the hole where the 02 sensor was? or just unplug it and wrap up the lead?

[NJNeal]

And D.R. Another CCP question. I have read that people use different colors of CCP in an attempt to drive different mapping. My bike was sold to me with a pink/rose colored CCP. From the parts schema it LOOKS like a yellow one was spec'd. Shall I go find a yellow one?

[dirtrider]

Evening Neal

 

Just unplug the 02 pigtail from the main harness (covering the connectors would be a good idea), leave the 02 sensor in the exhaust.

 

 

I see you have an 03 1150? Is it a single or twin spark?

 

[dirtrider]

And D.R. Another CCP question. I have read that people use different colors of CCP in an attempt to drive different mapping. My bike was sold to me with a pink/rose colored CCP. From the parts schema it LOOKS like a yellow one was spec'd. Shall I go find a yellow one?

 

Evening Neal

 

You can try a yellow CCP if you want (or just jumper the terminals in the fuse box to do the same thing-- "pins 30 to 87") but there is just a slight difference in spark mapping as the 02 sensor still forces the same lean light throttle fueling with either one.

 

[NJNeal]

My 2003 is a single spark. Interested in what color CCP other 2003 RT riders have luck with.

[NJNeal]

Thank you D.R.

[dirtrider]

My 2003 is a single spark. Interested in what color CCP other 2003 RT riders have luck with.

 

Evening Neal

 

Just make up a 4 pigtail jumper wire & try them all yourself. Just clear the Motronic adaptives at each change then do a new TPS re-learn.

 

On the 1150 single spark the best all around CCP I have found is the Pink (30 - 87-87A). It forces a very good progressive spark mapping.

 

The thing to do is jumper to whatever you want to try then ride it a week (at ALL throttle positions & engine loads). Then switch back to the Pink as a controller, then ride that a day or two then switch to the next one you want to try.

 

If you don't clear the Motronioc adaptives & do a new TPS re-learn then about any change you make will probably feel a bit better for a short while until the adaptives get re-established so you really can't do quick switch between CCPs & tell very much long term.

 

[NJNeal]

Cool. Thanks. That's going to keep me busy for a while.

[dirtrider]

Evening Neal

 

If your 1150 single spark surges as bad as you say & even using a Techlusion didn't help then you might have some other underlying issues like a lazy or poisioned 02 sensor, or fuel injector issue, or basic tuning issue.

 

With a good 02 sensor I have had very good success in getting the single spark 1150 to run good at light throttle light load using the Techlusion 1032. The thing is a bad 02 sensor will drive a 1032 daffy.

 

Before doing much CCP trials do the 02 sensor disconnect first, if the 02 sensor is acting up all the CCP ride-offs will tell you nothing & maybe point you in the wrong direction.

 

[NJNeal]

Okay. Will do the 02 disconnect first.

If it was a failing sensor, I thought the dealer would pick that up with their diagnostic equipment. This bike has been regularly serviced by the dealer since I've owned it, and they tell me all the diagnostics say that the system is working well. As for basic tuning, valves are adjusted, throttle bodies carefully synced, tps adjusted.

Didn't mean to derail this valuable thread with my bike's issues. Sorry to OP.

I will disconnect the 02 sensor and PM you with my findings.

[roger 04 rt]

Okay. Will do the 02 disconnect first.

If it was a failing sensor, I thought the dealer would pick that up with their diagnostic equipment. This bike has been regularly serviced by the dealer since I've owned it, and they tell me all the diagnostics say that the system is working well. As for basic tuning, valves are adjusted, throttle bodies carefully synced, tps adjusted.

Didn't mean to derail this valuable thread with my bike's issues. Sorry to OP.

I will disconnect the 02 sensor and PM you with my findings.

 

When you disconnect the O2 you need to decide whether to reset the Motronic on not. Resetting it will clear anything it has learned with the current O2. Same for the CCP. I would think you want to reset it. (i now see that DR suggested the reset.)

 

It is very hard to test the O2 and fuel Injector performance is not so visible either.

 

My 04RT didn't surge but it runs a LOT stronger now that my O2 is set at 14.2.

 

 

[roger 04 rt]

Roger - you are a worthy member of the group of people who turn to our bikes for intellectual stimulation as well as other stimulation.

 

But your explanation, if you will excuse my joshing, seems to require a little man inside the ECU to make decisions. I suppose it could be a woman as well, but she'd have to be real small. Or both.

 

I ask my question in all sincerity and I suspect there is a good answer, as least in theory and at least for far more sophisticated ECUs, but maybe for the 2.4. But I think the issue is "the referent" or criterion the ECU uses to say "gotta be longer spritzs".

 

After all, if it could be done automatically by the ECU itself you could just ship bikes with no map at all and you wouldn't need megabuck map development time at the Nurburgring.

 

And you wouldn't need different color code plugs for ethanol in Fiji or Iowa because the little couple inside the ECU could figure it out.

 

Ben

 

The little man or woman inside is the team of sotware programmers who wrote the code. As Testpilot said you need the basic maps to get the ball rolling and for "limp home" modes.

 

It is not at all clear to me whether the CCP changes fuelIng or timing or both. Of the three different pins that can be grounded, one is low octane fuel, one indicates a CAT and another seems to be model variant.

 

 

 

[roger 04 rt]

At the end of the day, in my practical experience, the Motronic doesn't seem to work that well. I've owned my 1150RT for about five months now and spent most of that time trying to solve for lean surging. I can feel the ECU making the changes as it searches for the right AFR. It's annoying as hell.

 

As some posters on this list know, I've tried ALL the suggestions, and installed a Techlusion. None of it really worked that well.

 

I finally, and very, very reluctantly just pulled the CCP, and that's given me, so far, the best balance of ride quality and MPG. I have no idea what damage I may be doing to the CAT, and really don't care...at least the bike is ridable again.

 

I was actually starting to shop around for other bikes as a result of my experience on this RT, and I learned that this lean surging is common among a lot of bikes between about 2000 and 2006. Ducati, Harley, Guzzi...many makers had this issue.

Neal

 

I don't think the Motronic is the weak link. It seems pretty solid to me. After basic tuning issues it seems to me that:

 

a) The Boxer engine as it existed then didn't like to run lean.

b) Can be thrown off track by a bad O2 sensor.

c) Can't easily be fixed by simple aftermarket fixes unless the O2 is pulled. But if you pull the O2 you lose all other adaption.

d) Can be permanently fixed by a wideband O2 such as LC-1 or Powercommander and setting closed loop to a richer target than 14.7. The PC solution takes the Motronic out of closed loop adaption but may do some of that itself. The LC-1 leaves the motronic in the closed loop but is not an easy plug and play.

 

 

 

 

 

[NJNeal]

I would be interested in learning more about the LC-1 and will do some research today.

 

My experience with Techlusion was that when it was working well, really eliminating surge, it also killed mpg, down to about 30mpg, which is just unacceptable.

 

My personal reluctance to do a PC, after looking at it closely, is that there's no shop close to me with (1)a dyno, (2) experience with both PC and RT bikes. So, if I did it, I'd have to travel a good distance to find someone to set it up, tweak it over time, etc. I am in Northern New Jersey. I do have access to a dyno equipped shop about an hour away and the guys there seem open to experimentation, but that's not really filling me with confidence.

 

I am willing to buy that the O2 sensor may in fact be not functioning 100%. I'd also be willing to buy a better o2 sensor as a test.

[dirtrider]

Morning Neal

 

No real suggestions in this post just a basic statement on your problem.

 

The basic thing you have to keep in mind is that you have a SINGLE SPARK 1150.

That involves rather large pistons, only 2 cylinders, & a fueling computer that uses a narrow band 02 (lambda) sensor (basically a switch) to adjust fueling to a marginal fuel/air mixture on gasoline. To make it even worse the fueling is non sequential & both cylinders are fueled to only one 02 sensor.

 

Then you have to add in the single spark thing--only one spark plug in the center of the combustion chamber to (try) to get the combustion going evenly over that large piston area.

 

Then, if you run it on anything other than pure gasoline (that being any of the modern reformulated fuels or fuel with alcohol in it) the problem gets worse as the 02 sensor will drive the combustion even leaner than if operating on pure gasoline.

 

Your light throttle runability & surging answer is going to mean that you either need to richen your fueling up enough to cover the basic single spark shortcomings. Or add the lower second set of spark plugs, or somehow fool the stock fueling system into running richer than is now. Just playing with CCPs alone won't do this as none of the CCPs can change closed loop fueling as the 02 sensor leads the dance in closed loop operation not the CCP.

 

Problem is: at light throttle low engine loadings the stock 02 sensor operating as intended won't allow the system to operate & run rich enough to get even combustion at light throttle low engine loadings with only one spark plug in the center of the chamber to light it off.

 

This means to get it operating surge free "on only one spark plug per cylinder" you will more than likely need to deactivate the 02 sensor in the surging operational range & maybe even make it run richer yet in that range , or somehow trick the 02 sensor into doing that for you (see Rogers great attempt above).

 

We are pretty well stepping on Rogers original thread here so maybe start a new thread on this so we don't get too far off topic here & turn Rogers thread into a mess.

 

[roger 04 rt]

I would be interested in learning more about the LC-1 and will do some research today.

 

My experience with Techlusion was that when it was working well, really eliminating surge, it also killed mpg, down to about 30mpg, which is just unacceptable.

 

My personal reluctance to do a PC, after looking at it closely, is that there's no shop close to me with (1)a dyno, (2) experience with both PC and RT bikes. So, if I did it, I'd have to travel a good distance to find someone to set it up, tweak it over time, etc. I am in Northern New Jersey. I do have access to a dyno equipped shop about an hour away and the guys there seem open to experimentation, but that's not really filling me with confidence.

 

I am willing to buy that the O2 sensor may in fact be not functioning 100%. I'd also be willing to buy a better o2 sensor as a test.

 

Near the end of this Wideband Thread are the settings I ended up with for the Innovate Motorsports LC-1. You should also look at the Zeitronix ZT-3 and the Innovate MXT-L.

 

I feel that the benefits of Wideband over PowerCommander or pulling the O2 are that the programmable-Wideband approach leaves the Motronic "fully in the loop". The Powercommander removes the O2 connection to the Motronic, as does the Techlusion. The Powercommander has a Closed Loop setting so it may take on some of those functions but its documentation is silent on that.

 

The biggest problem with Wideband is that there is no Plug and Play solution. And even if you're good, it takes several hours to install. However, a new stock O2 Sensor and an LC-1 and Wideband Sensor are about the same price.

 

It's really hard to believe that the R1100/1150s can't all run great with a somewhat richer mixture provided by the Motronic. As I mentioned earlier it seems to me that the two hardest problems to diagnose are Injector and O2 Sensor performance--just no easy way to measure or monitor them.

 

RB

[TestPilot]

Then, if you run it on anything other than pure gasoline (that being any of the modern reformulated fuels or fuel with alcohol in it) the problem gets worse as the 02 sensor will drive the combustion even leaner than if operating on pure gasoline.

 

This might be the case with a carbureted engine, but in closed loop operation the O2 sensor will drive lambda = 1 with or without 10% ethanol.

[dirtrider]

---This might be the case with a carbureted engine, but in closed loop operation the O2 sensor will drive lambda = 1 with or without 10% ethanol.

 

Morning Karl

 

Well sort of, but lambda is different with different alcohol ratios.

 

I think Roger has done a bit of research on this for his LC-1 project so hopefully Roger can add to this.

 

[JamesW]

Time to chime right in. The way I see it 10% ethanol is not really a problem. Where I see a problem is when often times you can't be at all sure you get 10% and not maybe as high as 30% which is not a good thing especially on motors that run to the lean side anyway. I still think the easiest and most cost effective measure with these machines is to simply disconnect the O2 sensor and whenever possible fill up with pure gas. Oh, and just think what will happen when E10 becomes E15.

 

I consider myself fortunate that within easy walking distance of me are 3 gas stations that sell pure gas premium which is legal in Oregon. Check out puregas.org to see what may be available in your area.

 

All this said I still find this thread to be most interesting and wouldn't it be nice if one could obtain design info for the Motronic? That will be when pigs fly.

 

FWIW, I like the green code plug best for my '04RT twin spark.

 

Just bought a '94R1100RSL so might yet find out what surging is really all about. Might end up saying thank God for EBay. lol....

[roger 04 rt]

Then, if you run it on anything other than pure gasoline (that being any of the modern reformulated fuels or fuel with alcohol in it) the problem gets worse as the 02 sensor will drive the combustion even leaner than if operating on pure gasoline.

 

This might be the case with a carbureted engine, but in closed loop operation the O2 sensor will drive lambda = 1 with or without 10% ethanol.

 

 

---This might be the case with a carbureted engine, but in closed loop operation the O2 sensor will drive lambda = 1 with or without 10% ethanol.

 

Morning Karl

 

Well sort of, but lambda is different with different alcohol ratios.

 

I think Roger has done a bit of research on this for his LC-1 project so hopefully Roger can add to this.

 

Here is what is going to happen when you run E10 which is 4% leaner than gasoline:

 

--Assuming the Motronic is reset, and that the fuel tables have been constructed for (pure) gasoline ...

 

--The engine warms up, running leaner than designed

 

--The Lambda sensor goes into operation, starting with the factors for gasoline.

 

--In a few seconds (I've watched this happen on the GS-911) the Lambda sensor, strives to drive Lambda=1 and drives the mixture richer by the few percent needed.

 

--An interesting side note, at Lambda=1 with E10 fuel, the Motronic gets the Air Fuel Ratio to 14.13:1 if there is truly 10% Ethanol. Newer vehicles may start closer to this figure than older ones. This is an important reason why I prefer to keep my O2 in the Closed Loop Motronic world and not disconnect it.

 

--Next time the Motronic goes Closed Loop, it takes the "adaptation" factor it learned and starts there.

 

--At some point, this becomes a long term correction factor for Open Loop. In Motronic terms, long term may only mean it takes a few minutes. (This I don't know, it may be somewhat longer or shorter.)

 

On my 04 R1150RT, E10 fuel wants a 4% richer mixture, by using the LC-1 my target of Lambda=0.97 means it wants a 3% richer mixture. So if I did nothing, just for these reasons, the Motronic would have to develop a long term correction of about 7%.

 

I am using a BoosterPlug to tell the Motronic the air is 20C cooler than actual so that the Motronic starts adapting with a 6% richer mixture than it normally would. This means my Motronic is only correcting by about 1% (for the factors I've mentioned).

 

[NJNeal]

With the BoosterPlug fooling your engine on temp, what is your average mpg? Just curious.

[TestPilot]

[quote=dirtriderMorning Karl

 

Well sort of, but lambda is different with different alcohol ratios.

 

I think Roger has done a bit of research on this for his LC-1 project so hopefully Roger can add to this.

 

I believe Roger got it right. A/F ratio changes with different fuels, but Lambda = 1 means stoichiometric regardless of the fuel.

 

O2 sensors do not measure A/F ratio; they measure excess oxygen. The switching point is around 2% O2. With less than 2% O2 in the exhaust gas, the sensor indicated rich (Lambda < 1); with more than 2% O2, lean (Lambda >1). So the O2 sensor in the Motronic should drive the system toward Lambda = 1 regardless of the fuel (within the limits of the system).

[Glenn Reed]

I consider myself fortunate that within easy walking distance of me are 3 gas stations that sell pure gas premium which is legal in Oregon. Check out puregas.org to see what may be available in your area.

 

Just FYI- the site is actually pure-gas.org

 

Somebody is squatting on the other one.

[dirtrider]

Afternoon

 

The problem with web sites like pure-gas.org is that there is no testing or verification if the gasoline sold is actually Alcohol free.

 

A lot of people post to sites like that without knowing alcohol free gasoline from coffee.

 

Some think that just because it isn't posted on the gas pump it must be alcohol free. Others simply believe what the pimple faced kid behind the counter tells them & still others hear it from another misinformed person.

 

I know if I owned a gas station in a high alcohol content area I would be posting all over those pure-gas.org type web sites saying I sold alcohol free gasoline. Would have my friends & neighbors posting the same there also. With no BS control & no penalty why not.

 

Just show me the (independent testing data) then I will believe that sites like that have some merit.

 

[roger 04 rt]

With the BoosterPlug fooling your engine on temp, what is your average mpg? Just curious.

 

I don't check it that often, but with the BoosterPlug, but not the Lambda reduction from 1 to 0.97, I was getting mid-40s. I guess that after the Lambda reduction it's a percent or two less.

[JamesW]

I dunno about all entries on pure-gas but I do know that in my travels in these parts (eastern OR/WA) I have not had a problem finding pure gas that I know is pure gas without doubt. For example there is a Conoco in Dayton, WA that even sells pure gas regular, never miss that one. Around here the thing that prohibits a lot of stations from selling the pure stuff is they (stations) don't have 3 storage tanks. Plus grade is 50/50 regular/premium E10 so only need two storage tanks. The older stations are the ones more likely to have extra storage for pure-gas. Also, many entries on pure-gas turn out to be card lock stations so unless you have the card you're SOL. All I'm saying is that websites like pure-gas are a source for info not to say that info is alway 100%. Kind of like info on internet m/c forums, present thread excluded of course.

 

 

[roger 04 rt]

While we're all pondering the operation of the Motronic in the O2 Sensor thread (and wanting to spark some controversy), here's a note that is in the R1150 BMW Repair Manual in three places:

 

NOTE: Disconnecting the battery means that the entries in the fault memory of the Motronic MA 2.4 control unit are deleted and the adaptation values are reset. This can temporarily impair the operating characteristics when the engine is restarted.

 

What does everyone/anyone think the term "adaptation values" means?

[TestPilot]

Adaptation values are sometimes called long term fuel trim. When operating closed loop, the ECU uses the injections times stored in the fuel map as a starting point and looks at the O2 sensor to see if it's running rich or lean. It corrects from there until it sees the O2 sensor input cycle from rich to lean and back. If, over time, the ECU determines that it's always starting too lean or too rich at certain points in the fuel map, it stores correction factors in a separate table. It uses these correction factors to make open loop operation closer to closed loop. When the battery is disconnected, these correction factors are lost.

[roger 04 rt]

Adaptation values are sometimes called long term fuel trim.  When operating closed loop, the ECU uses the injections times stored in the fuel map as a starting point and looks at the O2 sensor to see if it's running rich or lean.  It corrects from there until it sees the O2 sensor input cycle from rich to lean and back.  If, over time, the ECU determines that it's always starting too lean or too rich at certain points in the fuel map, it stores correction factors in a separate table.  It uses these correction factors to make open loop operation closer to closed loop.  When the battery is disconnected, these correction factors are lost.

 

That's what I understood. BMW and Bosch don't say much about this but other ECU manufacturers do. It means that the Motronic has a powerful ability to adapt to things like:

 

Fuel Pressure which can vary +/- 5%

Injector performance ?%

Volumetric efficiency  (accuracy of the fuel maps) ?%

Gasohol 4-5%

 

If the Motronic couldn't adapt (or if you disconnect the O2) actual open loop AFR could vary significantly. For example if an open loop TPS/RPM setting was expected to yield an AFR of 14.5:1, a 5% error could mean the actual AFR was anywhere between 13.8 to 15.2. Or if you ran gasohol and your fuel pressure & injector performance was 5% low you would end up at a very lean 16:1 instead of the 14.5:1 the Motronic was shooting for!

 

It's no wonder that different bikes show different results from disconnecting O2 sensors and changing coding plugs.

 

The other thing is that most efforts to richen the mixture will be thwarted by these adaptation values. You install a device to richen the mixture and the Motronic leans it back out for Open and Closed Loop.

 

On the other hand, this adaptation values function implies that if you use a Wideband sensor to richen the mixture by, say, reducing lambda by 6%, the Motronic will richen Open Loop automatically. And it will still continue to adjust for fuel and pressure/injector performance.

 

Food for thought.

[dirtrider]

----

 

On the other hand, this adaptation values function implies that if you use a Wideband sensor to richen the mixture by, say, reducing lambda by 6%, the Motronic will richen Open Loop automatically. And it will still continue to adjust for fuel and pressure/injector performance.

 

Food for thought.

 

Evening Roger

 

That would be my assumption also. The adaptives are based on the closed loop 02 info so if you offset the closed loop fueling you would surly think that should force the adaptives to match that at open loop entery.

 

The adoption values are mainly used to allow a seamless entry into closed loop & a seamless entry back into open loop. You really wouldn't want to jump from 13.2:1 into 14.7:1 closed loop then back into 13.4:1 again while riding along as that would make a very raspy fueling control.

 

Adaptives can also be used to keep the open loop fueling somewhat correct as engine & fueling system parts wear during long term use.

 

Based on other fueling computers that I am familiar with I would be willing to bet there are both short term & long term adaptives. The short term is used for immediate adaptive trim & when the short term runs out of trim ability it bumps the long term to the next cell then centers the short term in it's cell & starts again.

 

On the old 1150 (single spark) after you rode it a while then cleared the Mortronic (that also cleared the adaptives), then did a new TPS learn you could definitely feel it enter closed loop on the first ride after clearing the adaptives.

 

[roger 04 rt]

----

 

On the other hand, this adaptation values function implies that if you use a Wideband sensor to richen the mixture by, say, reducing lambda by 6%, the Motronic will richen Open Loop automatically. And it will still continue to adjust for fuel and pressure/injector performance.

 

Food for thought.

 

Evening Roger

 

That would be my assumption also. The adaptives are based on the closed loop 02 info so if you offset the closed loop fueling you would surly think that should force the adaptives to match that at open loop entery.

 

The adoption values are mainly used to allow a seamless entry into closed loop & a seamless entry back into open loop. You really wouldn't want to jump from 13.2:1 into 14.7:1 closed loop then back into 13.4:1 again while riding along as that would make a very raspy fueling control.

 

Adaptives can also be used to keep the open loop fueling somewhat correct as engine & fueling system parts wear during long term use.

 

Based on other fueling computers that I am familiar with I would be willing to bet there are both short term & long term adaptives. The short term is used for immediate adaptive trim & when the short term runs out of trim ability it bumps the long term to the next cell then centers the short term in it's cell & starts again.

 

On the old 1150 (single spark) after you rode it a while then cleared the Mortronic (that also cleared the adaptives), then did a new TPS learn you could definitely feel it enter closed loop on the first ride after clearing the adaptives.

 

Evening DR,

Makes sense to me. Does the 1100 manual, which I can't download at the moment, have a similar statement to the one I quoted for the 1150?