Motronic Long Term Fuel Trim

[dirtrider]

Evening Roger

 

OK, found it in the 1150 manual under Motronic replacement.

 

Nothing written in the 1100 manual about clearing adaptives or ridability issues on first start.

 

Only says:

 

Remove Motronic control unit.

• Remove connector strip.

• Install in the reverse order of work.

 

[dirtrider]

Roger

 

Something I've been meaning to ask you.

 

If you ever get a scope on that bike can you move the trigger over on the injectors then do a cold start.

 

I found some old Motronic 2.4 tuning data & it says the fuel injectors get TWO pulses pre rev (TDC & BDC) for cold starting & for a very short time after cold start-- could you verify that?

 

Then see if you can look at dropped throttle fuel shut off. My tuning data also mentions the closed throttle fuel shut off isn't operational below 70°c-- be nice to know if the motorcycle Motronic 2.4 uses this available parameter.

 

[roger 04 rt]

Roger

 

Something I've been meaning to ask you. 

 

If you ever get a scope on that bike can you move the trigger over on the injectors then do a cold start.

 

I found some old Motronic 2.4 tuning data & it says the fuel injectors get TWO pulses pre rev (TDC & BDC)  for cold starting & for a  very short time after cold start-- could you verify that?

 

Then see if you can look at dropped throttle fuel shut off. My tuning data also mentions the closed throttle fuel shut off isn't operational below 70°c-- be nice to know if  the motorcycle Motronic 2.4 uses this available parameter.       

 

 

Hi DR,

 

Thanks for the input on the 1100s. It seems surprising that Bosch would implement Closed Loop without adding Open Loop adaptation values. However unless I could see it in writing or measure it myself, it will remain a mystery for a while longer.

 

I still hope to borrow a scope and look at the O2 and LC-1 waveforms. The test you suggest would be interesting but hard to set up. Every engine cycle and every injector firing would have to be captured. I have seen the following enrichments described, so it wouldn't surprise me if the Motronic had them:

 

--Ignition ON single injector firing as an initial short purge

--Cold cranking enrichment for a fixed number of cycles

--Sliding scale warm-up enrichment.

 

Regarding dropped throttle fuel shutoff, is this the same as fuel overrun cutoff? I did specifically gather data to see if the injectors were shut off when the engine is cold, when you abruptly close the throttle--they are! This behavior really makes the bikes hiccup at times. As an aside, Closed loop operation seems to kick in well below 70 C. I think it happens nearer 55C.

 

I will keep you in mind when/if I get the scope.

[Peter Parts]

Very interesting to learn about these brilliant adaptation systems - which seems to require having a smart little man (or woman) inside the ECU making subtle decisions.

 

Very easy to say, "When the ECU detects low power, it lengthens the spritz'". But how?

 

Soooo, what I'd really appreciate from Testpilot, Roger, or Dirtrider is a blow-by-blow description of just what the ECU looks for (must be a sensorable physical parameter, of course, not just "low power" or "sluggish") and how it decides to do what and when. And if you are patient in explaining this in simple bricks-and-mortar way, how it modifies these adaptation values when your gas or additives or riding style or weather changes again? Besides just pulling the fuse.

 

I see the light dawning, but I think this wishful-thinking discussion has to include the computer-logic that is happening in order to separate the pure wishful-thinking from the real Motronic capabilities.

 

If your description is based on firm knowledge of the Motronic, please say so. Otherwise, just say, "My guess is...."

 

 

If there are double sparks for a cold engine and a disabled fuel over-run shut-off under 70, then a fooler that fiddles with the oil temp (or maybe air temp) sensor will screw-up those functions (as well as the normal functions for those sensor systems.

 

Roger, accepted wisdom is that the fuel over-run comes into play when you snap the throttle/TPS below maybe .4 v. So if you are getting too many hiccoughs, that's the place to check.

 

Ben

[dirtrider]

 

 

Morning Ben

 

You always seem to doubt everything Roger & I say about the Motronic 2.4 system operation.

 

So even if we did the extensive research for you you would question it, or say you misunderstand it, or doubt it.

 

I think it's about time for you to do your own research then you can choose to believe what you want about how the 2.4 or other Alpha N systems works.

 

Maybe start by working on more than one (unmolested) 2.4 system sample as you need a sample size much larger than only one machine to gather useful data.

 

A good starting point for you would be as Roger has done & set up your own system then use GS-911 to gather data & FULLY research the systems involved & how they interact with each other.

 

Or do as I have done like working with electronically controlled fueling systems for over 40 years. Or maybe talking directly to BMW & Bosch engineers about Motronic Alpha N systems operation. Then start gathering years of Motronic tuning data.

 

A good place to start is by researching Alpha N fueling systems & how they utilize learn capability & fueling adjustments.

 

Or reading the below links. They aren't for the Motronic systems but fueling control is pretty basic across the board.

 

 

 

http://www.gnttype.org/techarea/faq/BLMINT.html

 

http://carprogrammer.com/Z28/PCM/FAQ/How%20GM%20Electronic%20Fuel%20Injection%20Works.htm

 

[roger 04 rt]

This is really interesting stuff ...

 

Ben,

 

A couple things:

 

There really are little men and women inside the Motronic, and in a sense they are doing what you suspect. In fact, there are hundreds, if not thousands of them. What I'm referring to is the thousands of man years of physical R&D and software development which led to the code inside the Motronic. Those lines of code are the embodiment of those individuals running the ECU, moment by moment.

 

In the case of my '04 RT, the fuel overrun cutoff seems to work as designed. However, now that I'm aware of it, I can feel when the Motronic re-enables fueling. Interestingly, while my cars surely do the same thing, it is undetectable to me--perhaps the difference between newer and older or 2 cylinder and 8.

 

Thanks for the critique and feedback.

 

DR,

 

In some ways I wish I had read those two great articles you provided the links for, before I did all my data gathering. If I had though, I wouldn't have worked nearly as hard to understand what the Motronic is doing--so better that I read it now.

 

This is another good read for anyone with further interest:

http://www.megasquirt.info/

 

Having read the two pieces you pointed us to, I now am thinking that I want my LC-1 programmed to the nearest Lambda to 1 that will produce the drivability that I'm looking for. That way, I won't over-correct the Open Loop design objectives of the Motronic. If I went too low on Lambda, the Open Loop enrichments might become too rich.

 

Also, it is more clear than ever that the best way to improve the basic performance of the R1150 is by slightly reducing Lambda (increasing AFR) and keeping the O2 connected to let the Motronic do all the good things that Closed Loop lets it do.

 

Many thanks for the articles and feedback.

[dirtrider]

-----

 

In the case of my '04 RT, the fuel overrun cutoff seems to work as designed. However, now that I'm aware of it, I can feel when the Motronic re-enables fueling. Interestingly, while my cars surely do the same thing, it is undetectable to me--perhaps the difference between newer and older or 2 cylinder and 8.-----

 

 

 

 

Morning Roger

 

One thing that has kind of kept me pondering is that on the Motronic 2.2 & 2.4 BMW motorcycle system the BMW/Bosch fueling engineers seem to have chosen to use total fuel shut off on dropped throttle or (closed throttle) above 1400-1500 RPMs or so then turn it abruptly back on at or below that same RPM range.

 

Most automobile systems I am familiar with use a combination of complete fuel shut-off at higher RPM dropped throttle then taper it back in using some from of controlled enleanment so you don't get that that drastic difference or step in the fueling program as you totally close the throttle on decel.

 

[Peter Parts]

We don't seem to be making any progress on learning how the Motronic addresses adaptation.

 

Wishful thinking or reality?

 

I sure don't have any picture of how the Motronic handles adaptation. Does anybody?

 

Dirtrider - can't say as I find write-ups on car ECUs too illuminating. Have you read Adam Wade's book on bike EFI (and only bike EFI)? Truly great read - nice pictures too. A couple of years old, like the Motronics, eh. You might learn a lot.

 

Ben

[JamesW]

Morning DR,

I must say this just gets more and more interesting. It is all I can do to resist tearing into my new R1100RSL. This bike was built in 06/93 and has a cat converter but no receptacle for a ccp and no connector for a CO POT. Roger mentioned in a previous thread that he suspects the CCP is hard wired and this I would like to know. I am thinking about disconnecting this wired in CCP, if there is one, installing a CO POT (1K linear POT) and disconnecting the narrow band O2 sensor. The end result would be a European version 1100RS always in open loop, I think. I also think, maybe, I would end up with a better running m/c with a smoother idle and a bit richer low to mid range in the surge prone area. What do you think? Maybe I should wait until I get a few miles on it first? I might have talked about this before but I can't recall, must be age related.

[roger 04 rt]

Morning DR,

I must say this just gets more and more interesting. It is all I can do to resist tearing into my new R1100RSL. This bike was built in 06/93 and has a cat converter but no receptacle for a ccp and no connector for a CO POT. Roger mentioned in a previous thread that he suspects the CCP is hard wired and this I would like to know. I am thinking about disconnecting this wired in CCP, if there is one, installing a CO POT (1K linear POT) and disconnecting the narrow band O2 sensor. The end result would be a European version 1100RS always in open loop, I think. I also think, maybe, I would end up with a better running m/c with a smoother idle and a bit richer low to mid range in the surge prone area. What do you think? Maybe I should wait until I get a few miles on it first? I might have talked about this before but I can't recall, must be age related.

 

Given the many virtues of adaptation that the O2 sensor can bring, I would be slow to disable it, even though there seems to be less known of adaptive values on the 1100.

[JamesW]

Hello Roger,

Yes, that may be true. Making these mods would be easy and inexpensive and could be easily reversed. You are correct when you point out that info on these early oilheads is somewhat sparse. Maybe because not many were sold and at the time ('93,'94,'95) the venerable and much loved airhead was still in production. Guys like D.R. with his years of experience are a valuable source for info no doubt about it. The motronic ma2.2 in my 1100 is more basic than the ma2.4 in the '04 but easy to work with in some ways like being able to clock out logged fault codes by counting flashes from an LED after grounding a pin on the 3 pin diagnostic connector.

[dirtrider]

Morning DR,

I must say this just gets more and more interesting. It is all I can do to resist tearing into my new R1100RSL. This bike was built in 06/93 and has a cat converter but no receptacle for a ccp and no connector for a CO POT. Roger mentioned in a previous thread that he suspects the CCP is hard wired and this I would like to know. I am thinking about disconnecting this wired in CCP, if there is one, installing a CO POT (1K linear POT) and disconnecting the narrow band O2 sensor. The end result would be a European version 1100RS always in open loop, I think. I also think, maybe, I would end up with a better running m/c with a smoother idle and a bit richer low to mid range in the surge prone area. What do you think? Maybe I should wait until I get a few miles on it first? I might have talked about this before but I can't recall, must be age related.

 

Afternoon James

 

You are assuming there is more than one fueling/spark mapping calibration in your computer. My guess is that is assuming a lot.

 

The pre CCP Motronics were different per country/emission cal so why would BMW put an extra cal in there they couldn't access with an external a CCP jumper. I would be willing to bet you a dozen donuts those pre CCP Motronics were flashed at time of build with the one & only calibration for that unit.

 

The basic idea behind the CCP system was to allow the same Motronic base to be loaded with all the fuel/spark mapping for different areas then use different calibrators (CCP) to access that mapping (a big cost saving).

 

[dirtrider]

Afternoon Roger

 

I'm beginning to think you are on the money with that "use the 02 if possible assumption".

 

I did a little more searching in my Motronic archives that turned up a document that a fueling engineer friend sent me from Germany long ago.

 

This is in regards to the automotive Motronic 2.4 system but the basic box works about the same.

 

Loosely translated it says: There are no sensors for air flow or intake manifold vacuum used by the Ma/2.4 and due to the simple engine load sensing, it is heavily dependent on the lambda sensor for correct functioning.

 

There is also mention later in the doccument that the Ma/2.4 has always had adaptive closed loop lambda control. But it didn't say how extensive (cell coverage) or how far up the RPM/load ladder it goes.

 

[roger 04 rt]

Afternoon Roger 

 

I'm beginning to think you are on the money with that  "use the 02 if possible  assumption".  

 

I did a little more searching in my Motronic archives that turned up a document that  a fueling  engineer  friend sent me  from Germany long ago. 

 

This is in regards to the automotive Motronic 2.4 system but the basic box works about the same. 

 

Loosely translated it says:  There are no sensors for air flow or intake manifold vacuum used by the Ma/2.4 and due to the  simple engine load sensing, it is heavily dependent on the lambda sensor for correct functioning. 

 

There is also mention  later in the doccument that the Ma/2.4 has always had adaptive closed loop lambda control. But it didn't say how extensive (cell coverage) or how far up the RPM/load ladder it goes.

 

 

These are very big data points for me, thank you. The connection between Alpha-N fueling management and the lambda sensor makes good sense. I've read that Wideband EGOs are often used to fine tune VE (volumetric efficiency) tables to match target AFR tables using auto-tune software.

 

The Megasquirt designers say that for best fueling computations is is helpful to use MAP and Alpha-N, each where it is best suited. You can probably tell that I'm tempted to get one and interface it to the R1150. It just seems like too big a project.

 

From the data I've taken, the MA 2.4 goes Open Loop about 3000 RPM at light load but will hold Closed Loop to 5000 or 6000 RPM at higher loads. I don't think it's too important to know how many "longterm trim cells" there are. Just knowing they're there puts a premium on keeping the O2 sensor alive.

 

For now, my strategy is as follows:

 

1) Keep the stock CCP and Intake Manifold, believing that Bosch and BMW knew how to measure and set the VE tables and spark-timing far better than I could hope to. And that their big constraints were the catalytic converter needs and emissions.

 

2) Keep an O2 sensor installed and operating so that I get the maximum benefit from the Motronic's Adaptation Values functions.

 

3) Use a Wideband Sensor/Controller combo to reduce Lambda just enough to get the best combination of Closed Loop AFR and Open Loop Adaptive Values. At the moment a lambda between 0.94 and 0.96 works very well. I will try richer and leaner over the next few months.

[JamesW]

Oops, hadn't thought about that. Probably is no jumper for a cat code because there is only one installed fueling plan hence not only no CCP but no hard wired jumper either which means it is what it is. Maybe an LC-1 would do really good things in this '94 R1100RS situation.

 

I can see I am going to have to spend a few dollars either that or go nuts!

[dirtrider]

Evening Roger

 

It would be nice to use some form of MAF or MAP sensor to supplement the Alpha N system but that is difficult on the Boxer twin. Either would be a great engine load supplemental input.

 

A MAP sensor barely works on a 4 cylinder engine & about useless on a 360° 2 cylinder engine due to the harsh pressure pulses.

 

A MAF would work but with a throttle body on each side that would mean either 2 separate MAF's or somehow get the intake air flow as a combined.

 

The big gain with using a MAP or MAF would be a more redundant load input if the TPS failed as they will stay drivable on the MAP or MAF just not precise.

 

Even the way more advanced 1200 BMS-K with twin 02 sensors is still a basic Alpha N system. The BMS-K does use cylinder head temp input (early 1200 used two, one on each head) & uses twin knock sensors as well as camshaft position sensor for sequential injection with the idle RPM being stepper controlled & the later ones having a system-voltage sensitive idle RPM increase.

 

[dirtrider]

Oops, hadn't thought about that. Probably is no jumper for a cat code because there is only one installed fueling plan hence not only no CCP but no hard wired jumper either which means it is what it is. Maybe an LC-1 would do really good things in this '94 R1100RS situation.

 

I can see I am going to have to spend a few dollars either that or go nuts!

 

Evening James

 

I really haven't played much with the early (non CCP) BMW FI systems.

 

Personally if it were me the first thing I would do is get it on a chassis dyno with exhaust plotting capability then run it through the range in a few upper gears. First with then without the 02 operational just to get a baseline & see what you have both ways.

 

Also rig up an adjustable pot on the air intake sensor to play with while the 02 is disconnected. That thing might respond very favorably to some intake sensor spoofing as long as the 02 is disabled.

 

 

[JamesW]

Good thoughts DR. Closest DYNO is Boise about 145 miles from here. Yeah, disconnect O2 and go open loop and play with intake air temp sensor. I looked at OEM Parts on-line and the motronic part number is the same for both EURO and US bikes. Another odd thing is I got the original '94 service manual along with electrical drawings with the m/c and everything shows the presence of a CCP socket along with the different CCP configurations. The only model with no CCP also has no cat converter and does have a CO POT. My model really isn't listed. I guess that is why I am so curious about the fueling computer in these beasts. As long as my only oilhead was the '04RT I was all right but now.....? Got to wait for better weather.

[dirtrider]

Evening James

 

You got me thinking on that 94 Motronic. Somewhere I have an old service bulletin on that era 93/94 1100RS Motronic usage.

 

OK, I found it!

 

The 93RS is showing a socket for the CCP with a note that a cat could be retrofitted. Only seems to be two maps--one for no-cat & one for cat.

Some of the other 93's RT/GS etc had the cat coding jumper wire in the wire harness itself.

 

The bulletin says 94 1100 RS from 94 on has 6 engine versions available. As far as I can tell the CCP forced ones are all for cat usage (or cat retrofitting) in different countries.

 

Looks like no CCP = No cat, therefore no CCP = no closed loop operation.

 

[roger 04 rt]

...

 

The bulletin says  94 1100 RS from 94 on has 6 engine versions available. As far as I can tell the CCP forced ones are all for cat usage (or cat retrofitting)  in different countries.

 

Looks like no CCP = No cat, therefore no CCP =  no closed loop operation.  

 

 

DR,

I went over the 6 configs once and 4 of the appear to be for cat and 2 for non cat. Any CCP with 86 or 87 grounded is a cat config.

 

Also, has anyone pulled the CCP and then affirmatively verified that the 1100 doesn't go closed loop? (on the 1150 the motronic still runs the closed loop program with no cat. I confirmed it)

[dirtrider]

Evening Roger

 

Those 93's/94's seem to be an odd duck when it comes to Motronic programming. I know what you are saying on the 95 up Ma 2.2.

 

All I have to go on the 94RS is what my service bulletin shows & it only shows the CCP's were used for a cat. or cat. retrofit with different CCPs used for different countries.

 

Yes, on the 95 up 1100RT/GS Ma2.2 they definitely go open loop with no CCP. That also allows them to respond to an idle air trim potentiometer.

 

 

The 1150 Ma 2.4 (single spark) has all closed loop mapping inside (as far as I could ever tell) & (as far as I can tell) the CCP's are mainly used for fuel differences & evap purge control in different regions.

 

On the 1150 twin spark Ma 2.4 I never ran all the maps for being closed loop (I don't see why they wouldn't be but I can't say they are with total confidence).

 

[Peter Parts]

Gettin' warmer.

 

Thanks to DR's search, it looks like the bike Motronic has some adaptation in close-loop mode. Actually, it is barely conceivable that you could operate closed-loop without some adaptation to bracket the range of spritz lengths and adapting further on the fly as you ride and things change.

 

Now, does the ECU save any of that information (which it can quickly develop each and every time closed-loop begins) let alone store it in cells to trim open-loop functioning in short- or long-term use?

 

I tend to Occam's Razor when it comes to the Motronics but maybe it does store adaptation trims.

 

I've long been puzzled that bikes don't use MAP control - and I think Adam Wade hints at the same puzzle. In days of yore - K-bikes - MAP sensors were mechanical and cumbersome and fragile in a bike setting. Not so with modern non-mechanical MAP sensors and likewise, with some electronic time-constant fiddling, could work OK on a single cylinder.

 

If you take a hammer and smash your TwinMax (which reflects how I feel about the TwinMax) you'll find a dandy air pressure sensor.

 

Ben

[dirtrider]

Evening Ben

 

MAP sensors are pretty useless on the singles or twin cylinder 360° firing engines. If they are dampened or hindered in response time they become pretty lame for immediate engine load input.

 

RPM/throttle position (alpha-N) works pretty good for engine load input at all except altitude compensation & that is where the MAP system shines.

 

The Motronic Ma 2.2 & Ma 2.4 have built in barro sensors so they really don't need a MAP for altitude compensation.

 

A MAP would be a nice addition but Bosch/BMW probably didn't think the gain warranted the added engineering needed to get one to work properly on the 2 cylinder boxer.

 

On the saved adaptives ?--

 

The Ma 2.4 definitely saves some of the adaptives. If you clear the Motronic 2.4 you can definitely feel it go into closed loop the first few times it does after the clearing. After a bit of riding the closed loop becomes seamless again.

 

The 1150 service manual also cautions of impaired engine operation right after starting after adaptives clearing. How much of the long term retained I can't say. Definitely the short term would disappear at key off. The 1150 2.4 also retains the TPS learned parameters at key off but not battery disconnect.

 

[TestPilot]

I've long been puzzled that bikes don't use MAP control - and I think Adam Wade hints at the same puzzle. In days of yore - K-bikes - MAP sensors were mechanical and cumbersome and fragile in a bike setting. Not so with modern non-mechanical MAP sensors and likewise, with some electronic time-constant fiddling, could work OK on a single cylinder.

 

Ben,

 

If you're referring to the K75s and K100s, they had mechanical MAF sensors, not MAP sensors. The L-Jetronic system on my K100RS was almost identical to the system Bosch used in automobiles.

[roger 04 rt]

As far as selecting among the most likely scenario for Adaptation values, here (again) is what it says in the BMW Repair Manual:

 

 

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.

 

That makes MULTIPLE adaptation values most likely and the theory to be disproven for me, each to her/his own though. Frankly, I just don't see the case against it given what's written and what's been measured by myself on the 1150.

[roger 04 rt]

If you look carefully at this AFR plot below you will see that the whole plot centers around 14.2:1 even though this trip was only in Closed Loop half the time.

 

 

If adaptation had not taken place, much of this plot would be at 14.7:1. Occam's razor is sharp. (And of course Occam's razor is itself a theory to be proven. )

[Peter Parts]

Thanks for keenly-argued informative posts.

 

Not sure there's really much difference between MAF and MAP - job is the same. But on K-bikes, the less said the better.

 

Without in any way discounting the DR conclusions that air sensing has serious drawbacks (nor would I want to casually discount his opinions on other bike matters also... as I have said MANY times before), it should be recognized that air flow or air mass are basic parameters of combustion while butterfly position is only a crude proxy for air mass (and is faulty in ways besides altitude pressure too). That is one reason I'd always favor air sensing over butterfly position sensing.

 

So it seems there might be something saved in volatile memory between runs*. What? How much?

 

Roger - Einstein also had trouble with Occam's Razor as a "theory" as you put it. He said, "A theory should be simple. But not too simple."

 

Pity Bosch is so secretive about their Motronics.

 

Ben

*Can't honestly say I've ever noticed a problem in running poorly after a fuse removal. Just a few times and mostly had other things to watch for at such times. Shouldn't the open-loop map(s) work just fine whether or not there's any adaptation fudge factor? They are supposed to.

[roger 04 rt]

....

 

So it seems there might be something saved in volatile memory between runs*. What? How much?

 

 

Since you favor the razor, the ball is now in your court to make the case against adaptation, with appropriate data (e.g. show some data that exhibits adaptation not happening).

[Peter Parts]

....

 

So it seems there might be something saved in volatile memory between runs*. What? How much?

 

 

Since you favor the razor, the ball is now in your court to make the case against adaptation, with appropriate data (e.g. show some data that exhibits adaptation not happening).

 

Sure, Roger: as each closed-loop period ends, a single trim factor for all rpms and TPSs is retained by the ECU. I think that is what has been called here the short-term adaptation and I think any long-term adaptation (albeit volatile memory) is more conjectural (to use a polite term).

 

This is pure speculation.

 

My loose understanding is that the three other trim sensors have some kind of polynomial or look-up table that takes their reading, say "oil temp is 25C" and converts to a single trim multiplier that then acts on all cells of the fueling map... but not the spark map. Unless you've installed a PowerBoosterGizmo. That's why I offer this simple model for adaptation storage, if there is any.

 

Now, I have never observed this myself, as I indicated in my footnote. But it does suggest some experimental tests to verify the notion.... provided the adaptation or re-adaptation or de-adaptation takes place slowly enough to be butt-tested. Since some folks have said they can feel it happening, then maybe they can test it.

 

 

Ben

[JamesW]

Morning to you Roger,

Just one little question. With your GS911, can you tell me how long it takes for your '04R1150 to go into closed loop after a cold start up with the m/c just at idle and no throttle blipping?

 

Man, I am beginning to think these motronics are maybe worth their high list price.

 

Going to talk to better half about ordering the LC-1 today. I got big trouble, she is really interested in a 3D LED TV.

[roger 04 rt]

Just one little question.  With your GS911, can you tell me how long it takes for your '04R1150 to go into closed loop after a cold start up with the m/c just at idle and no throttle blipping?

 

If you waded through the original Wideband O2 post I started you can find some data on that, in case there are other things you'd like to know. I have tried to keep it up to date. But for the sake of time, after a cold start my 1150 takes about 10 minutes in my 40F garage to warm up to about 130F oil temperature. At that point, the Motronic enables the Closed Loop program. When on the center stand, it allows Closed Loop to 3000 RPM. On the highway, I have seen Closed Loop at 5000 RPM and higher (don't have the data in front of me but might be in the O2 thread).

 

I have not taken any data on an 1100 and have not seen any. I would really like to get a PDF copy of an R1100RS/RT BMW Repair Manual. I have one for the 1150. Can you or anyone e-mail me a copy?

 

What I'm looking for is information related to the 1100 adaptation strategy.

 

As far as putting an LC-1 on an 1100, I don't see any reason why it wouldn't work. However, I would like to know what the O2 sensor lo and high pins measure with the O2 disconnected and power on, using a sensitive DVM. My 1150 measures 140mV and 590mV (from memory). Knowing this let's us determine how much the Motronic offsets the ground, for its own diagnostic purposes.

 

[JamesW]

My shop manual for the 1100RS/GS probably wouldn't give you what I think you are looking for. Not sure I understand "adaption strategy", however. I am going to remove plastic and gas tank tomorrow and have a look at the motronic and see what jumper (if any) is present to represent a code plug. Will also install wiring to allow a CO POT. Will do some measurements of O2 sensor output while I'm at it.

 

I assume your reference to adaption strategy refers to installation of the LC-1 on an 1100? I looked at the LC-1 installation, calibration, and programming manual on-line and it should be no problem for the 1100. Pretty straight forward for both models. Being able to know exactly what the AF ratio actually is makes the LC-1 the best option, imo, compared to any other method. I'm totally sold and thanx for your very good work. Going to reread these threads again too easy to have missed something.

[Galactic Greyhound]

Roger,

 

I have a hardcopy BMW Repair Manual for my 1996 R1100RT.

 

I am unable to find any reference in this manual to the Motronic MA 2.2 unit storing any adaptive values.

 

The closest thing I can find related to this is: 'Group 12 - Engine Electrical System - Ignition System - Motronic (mapped characteristic control)'. There is nothing related in 'Group 13 - Fuel preparation and Control'.

 

My bike is to UK spec and does not have a catalytic converter. The space for the Encoding Plug is empty. I have a CO pot fitted for the exhaust gas emission control.

 

The only Motronic inputs seem to be from the HES, the TPS, Air Temp sensor in the airbox and the engine oil temp sensor.

 

Bit of a dinosaur I suppose - but it works and it's reliable!

[JamesW]

Hello Ced,

On your EURO spec bike you no doubt, I think, have just a CO POT and no O2 sensor. Do you experience any tendency to surge and do you ever adjust the CO POT yourself much as one might tweak the mixture adjust on a carburetor?

[roger 04 rt]

... Will do some measurements of O2 sensor output while I'm at it.

 

I assume your reference to adaption strategy refers to installation of the LC-1 on an 1100? 

 

 

 

Thanks for the measurement offer. By adaptation strategy I mean the way it applies (or doesn't) what it learns during closed loop operation to open loop adjustment. It took several times through the 1150 manual before I spotted it.

 

...I am unable to find any reference in this manual to the Motronic MA 2.2 unit storing any adaptive values.

...

 

The only Motronic inputs seem to be from the HES, the TPS, Air Temp sensor in the airbox and the engine oil temp sensor.

 

Bit of a dinosaur I suppose - but it works and it's reliable!  

 

The 1150 has the same inputs, plus the O2 sensor. That's all it takes to make a good Alpha-N ECU. If you spot anything about what happens after you disconnect the battery or remove the Motronic fuse, please let me know.

 

In the meantime if anyone has the R1100 manual in PDF format I would appreciate a copy.

[Galactic Greyhound]

Hello Ced,

On your EURO spec bike you no doubt, I think, have just a CO POT and no O2 sensor. Do you experience any tendency to surge and do you ever adjust the CO POT yourself much as one might tweak the mixture adjust on a carburetor?

 

Hi James,

Confirmed - I have a CO pot and no O2 sensor (no cat). I have had this bike from 1997 and never experienced any surging problems and never touched the CO pot. I check the throttle synchronisation about once a year at the winter overhaul using a twin tube mercury manometer and always use the standard Bosch FR6DDC plugs. The HES was replaced last year for the first time after a failure on the road at around 59,000 miles.

 

If you spot anything about what happens after you disconnect the battery or remove the Motronic fuse, please let me know.

 

Hi Roger,

I have never noticed any change to the performance after disconnecting the battery or Motronic fuse.

 

[roger 04 rt]

 

... I have had this bike from 1997 and never experienced any surging problems and never touched the CO pot. I check the throttle synchronisation about once a year at the winter overhaul using a twin tube mercury manometer and always use the standard Bosch FR6DDC plugs. ...

 

If you spot anything about what happens after you disconnect the battery or remove the Motronic fuse, please let me know.

 

I have never noticed any change to the performance after disconnecting the battery or Motronic fuse.

 

 

Two things stand out for me:

 

--No surging ever with straightforward tuning.

--No effect from resetting the Motronic.

 

Since you don't have an O2 sensor, there is no feedback mechanism to allow for the creation of adaptation values. So the performance before and after any resetting would be the same.

 

The only Motronic tune would therefore be CO setting. I could only guess where the factory would set CO and whether there was any ethanol in the fuel during tuning. But my guess would be 1.5% CO and no ethanol--1.5% is the center of the 1% to 2% spec for CO. (Does your current fuel have any ethanol in it?)

 

Tables for CO vs AFR show the following:

 

1%    = 14.11:1

1.5% = 13.94:1

2%    = 13.77:1

 

So I'll go out on a limb and guess that your idle AFR is running around 14:1 and that much of the cruising range is in that neighborhood. Higher power levels would be richer. (I'm running an AFR of 13.8:1 now and find that the 1150 runs smooth as silk there, with lots of low RPM grunt.) If you started running E10 without readjusting CO, the idle AFR would go to about 14.5:1 I believe.

 

If someone put an LC-1 on an 1100 with CO adjusted to 1.5%, ran a no-cat plug and didn't connect the O2 sensor, they could quickly get a good idea of the fuel map by reading the gauge or by connecting to a computer and logging AFR as they rode.

 

Now I'll go way out on a limb and bet that 1100s, reasonably tuned, running AFRs between 13.8 and 14.0:1 would not exhibit surging. But that bet will wait till someone tries it out.

 

[dirtrider]

Morning Roger

 

The Co. trim pot on the 1100 2.2 systems is ONLY for idle Co adjustment. Non cat spec is 1.5 (+/-.5). With the CCP in place for closed loop operation (pin 87) the MA 2.2 doesn't look for or respond to a trim pot even if one is in place.

 

The trim pot acts similar to the idle mixture screws on a carburetor in that it is most effective right at curb idle. By default it's control does travel up the light load RPM range just a bit (about like carb idle screws) , but at moderate throttle under road load has no appreciable effect on fuel tuning.

 

The trim pot was used for non cat countries that had a basic idle Co. requirement or even annual idle Co. testing. Not much alcohol fuel in use during the 1100 Ma 2.2 era.

 

Without a trim pot in place & assuming no CCP or at least a CCP that doesn't force closed loop (pin 87) then the thing defaults to programmed rich Co. idle. I measured that years ago when I had a 1100 Ma 2.2 no CCP on my friends Dyno but I can't find my info at the moment (that was before I kept computer records).

 

If I remember correctly it was rich but not outrageous. (well above the emission setting of 1.5 though.

 

For most of us that had 1100 Ma 2.2 systems back in the day we would just remove the CCP & run them with no trim pot. You could smell a bit rich at idle but I never saw one overheat the cat. On mine I always gutted the stock cat's just as a precaution.

 

Even the few that I know that ran a Co. pot (or homemade trim pot) never set the idle Co. with a meter we just set for best & highest idle. (I would imagine well above the 1.5 threshold)

 

[roger 04 rt]

Morning Roger

 

The Co. trim pot on the 1100 2.2 systems is ONLY for idle Co adjustment. Non cat spec is 1.5 (+/-.5). With the CCP in place for closed loop operation (pin 87) the MA 2.2 doesn't look for or respond to a trim pot even if one is in place.

 

The trim pot acts similar to the idle mixture screws on a carburetor in that it is most effective right at curb idle. By default it's control does travel up the light load RPM range just a bit (about like carb idle screws) , but at moderate throttle under road load has no appreciable effect on fuel tuning.

 

The trim pot was used for non cat countries that had a basic idle Co. requirement or even annual idle Co. testing. Not much alcohol fuel in use during the 1100 Ma 2.2 era.

 

Without a trim pot in place & assuming no CCP or at least a CCP that doesn't force closed loop (pin 87) then the thing defaults to programmed rich Co. idle. I measured that years ago when I had a 1100 Ma 2.2 no CCP on my friends Dyno but I can't find my info at the moment (that was before I kept computer records).

 

If I remember correctly it was rich but not outrageous. (well above the emission setting of 1.5 though.

 

For most of us that had 1100 Ma 2.2 systems back in the day we would just remove the CCP & run them with no trim pot. You could smell a bit rich at idle but I never saw one overheat the cat. On mine I always gutted the stock cat's just as a precaution.

 

Even the few that I know that ran a Co. pot (or homemade trim pot) never set the idle Co. with a meter we just set for best & highest idle. (I would imagine well above the 1.5 threshold)

 

Afternoon DR.

On the left coast at the moment. Just watched the Santa Monica PD put a half dozen 1200 RTPs through training manuevers in a parking lot. Pretty good riders.

 

Since there still seems to be more than a few 1100s that surge, fifteen years after they were made, I'm going to pursue a few points.

 

Given that the Motronic is a digital, alpha-n ECU, and setting aside enrichments and enleanments, the non-cat fueling algorithm would be to read a fueling value based on TPS and RPM, modify for oil temp to a limit, modify for air pressure, modify for air temp, modify for bat voltage and modify for the CO pot. Excepting the CO pot for a moment, the modifiers are multiplicative scale factors.

 

So for the CO pot. if as you suggest it only acts at low tps/rpm, it could be an additive/subtractive value (like the injector on constant) which would mean its effect diminished as pulse width increased. Or it could be a multiplicative factor like the others but limited to certain rpm/tps conditions, either are possibilites.

 

The problem I see with both those scenarios is that the designers would have given up a chance to have an adjustment for aging, fuel pressure and injector performance by allowing the CO pot to be set at idle but scale the whole fueling equation like the other modifiers. If they had done that, wouldn't that also have been felt more at low tps/rpm? Given a richer fuel table hypothesis for non-cat I see that as a real possibility, even as most likely.

 

if someone puts an AFR gauge on we might find out.

[dirtrider]

Evening Roger

 

I'm pretty sure the BMW fueling engineers only had what was available in the 2.2 box to work with.

 

The big concern was to allow the idle Co. to be adjusted to meet emission compliance testing in certain countries.

 

Seems they wouldn't want the idle Co. adjustments to effect the part throttle & mid throttle fueling by driving it too rich or too lean off the basic sensor trimmed fueling tables. (at least that's my though on it)

 

Maybe an idle Co. pot along with an upper fueling trim adjustment but that probably isn't in the basic Motronic 2.2.

 

[roger 04 rt]

Yes, but the trim pot modifies a resistance/voltage that the Motronic must digitize just like the other sensors. Then it must add, multiply or lookup and modify by digital arithmetic ... hence the three scenarios I posed in my last post.

 

All of those options would have been available.

 

Since BMW and Bosch won't tell us we have to measure to find out.

[roger 04 rt]

I suppose it is somehow limited to low rpm/TPS.

[JamesW]

Evening guys,

Well, I now have a EURO version R1100RSL. Have no O2 sensor and no ccp but do have a cat converter and a 9 turn 1K CO POT. It did have a jumper instead of a ccp which I cut. Warmed it up and adjusted CO POT for best idle, not hard to do and you know by the sound of it where that best idle is. Also, checked TB sync and had to slow the idle down just a touch. Went for very short, brrrrrr, ride and I gotta tell ya it rivals if not equals my old R100 for just plain smooth from idle right up through the revs with no trace of surge. Gotta say I'm very satisfied. The jumper simulated a yellow code plug with pin 10 of motronic connector grounded which is what the shop manual calls for on an RS with a cat converter. Had to solder a wire directly to pin 28 on the motronic because the cable connector has no female pins for any unused male pin, worked good and I did use a battery powered soldering iron so as not to take chances with the motronic. Need to find an exhaust gas analyzer because would like to know what AF ratio is. Maybe better I just enjoy what I have.

[roger 04 rt]

Evening guys,

Well, I now have a EURO version R1100RSL. Have no O2 sensor and no ccp but do have a cat converter and a 9 turn 1K CO POT. It did have a jumper instead of a ccp which I cut. Warmed it up and adjusted CO POT for best idle, not hard to do and you know by the sound of it where that best idle is. Also, checked TB sync and had to slow the idle down just a touch. Went for very short, brrrrrr, ride and I gotta tell ya it rivals if not equals my old R100 for just plain smooth from idle right up through the revs with no trace of surge. Gotta say I'm very satisfied. The jumper simulated a yellow code plug with pin 10 of motronic connector grounded which is what the shop manual calls for on an RS with a cat converter. Had to solder a wire directly to pin 28 on the motronic because the cable connector has no female pins for any unused male pin, worked good and I did use a battery powered soldering iron so as not to take chances with the motronic. Need to find an exhaust gas analyzer because would like to know what AF ratio is. Maybe better I just enjoy what I have.

 

Way to go. Very happy for you. It will be very interesting to hear what the AFRs are if you ever get the chance to look.

 

I'm not surprised about the faster idle with the richer mixture. Your bike is happy.

[JamesW]

Yes bike seems happy. I had to solder a #24 wire directly to male pin #28 on motronic and while I did a pro job of this I really would like to obtain a female pin that goes in the cable harness motronic connector. Any idea where to begin looking? It would be tedious should one have to unplug the motronic. Can be done but would be a pain. If I could find a pin I could install it when I do 1K service. I just like things right.

 

Now I got to think about the '04 R1150. I think an LC-1 would be the best thing. I would think about selling it but I am rather fond of the beast and with the Russell it is hard to beat for comfort.

[JamesW]

If I could find a dealer that has a wiring harness from a wreck then I could maybe talk them into cutting the motronic connector off and mailing it to me then I would have some female pins. About the only way I'll ever find one I think. Of course how many times do you ever unplug your motronic? Not going to loose sleep over it.

[dirtrider]

Afternoon James

 

If you have a cat with Co. pot with no 02 that sound like a retrofit bike for import.

 

 

On the Motronic terminal. That Motronic is the same as the car units so try a VW dealer or auto parts store that carries import parts or even a BMW car dealer. About anybody that carries Bosch terminals should have one.

 

[JamesW]

Well I couldn't be more pleased with the way this '94 RSL runs after I converted it to a EURO. Nice weather today so I did some cruising around. Some around town and some freeway riding. The power delivery is absolutely smooth. Only way you know you're close to red line is to look at the tach. Putting along at 25 in first or second is also smooth as you could ask and the idle reminds me of a very good running airhead as it never varies just sits at 1K RPM. The really good part is the plugs are now a light tan color instead of chalk white. This thing is a keeper. I thought my '04RT was a pretty good and smooth runner but it just isn't as good as this RS. Not sure it ever will be. Why BMW ever went to a twin plug design with stick coils is beyond me when the had already produced bikes like the '94RS. And why that joke of an ABS system compared to that found on the '94?? No more new Beemers for me.

 

Will check with some car dealers. Thanks for tip.

[roger 04 rt]

Well I couldn't be more pleased with the way this '94 RSL runs after I converted it to a EURO. Nice weather today so I did some cruising around. Some around town and some freeway riding. The power delivery is absolutely smooth. Only way you know you're close to red line is to look at the tach. Putting along at 25 in first or second is also smooth as you could ask and the idle reminds me of a very good running airhead as it never varies just sits at 1K RPM. The really good part is the plugs are now a light tan color instead of chalk white. This thing is a keeper. I thought my '04RT was a pretty good and smooth runner but it just isn't as good as this RS. Not sure it ever will be. Why BMW ever went to a twin plug design with stick coils is beyond me when the had already produced bikes like the '94RS. And why that joke of an ABS system compared to that found on the '94?? No more new Beemers for me.

 

Will check with some car dealers. Thanks for tip.

 

I think you've gotten your AFR to the high 13s. I think you'll be just as pleased as I am with the 04 if you get it to the same AFR.

[JamesW]

We shall see as I am going for the LC-1 for the 04. No gain in selling it. Too bad I can't just do to it what I did to the '94. One thing I would like to do is get rid of those stick coils. I wonder if anyone has ever tried to drive standard coils with the output signal from the motronic. Doubt it would hurt the motronic if I were to give it a go. I see no gain with those stk coils at all just expensive to replace and not nice to troubleshoot.