Open loop running: throttle and RPM?

[AZBaldur]

The other day I was checking my TPS and noticed that the voltage reading stopped increasing at a certain amount of throttle. Don't recall whether it was a quarter throttle, half, or what. Got me thinking though, and perhaps someone here can educate me. Under what conditions does the Motronic switch from closed loop to open loop running? At changing throttle, over half throttle, over 4500 RPM...? When is the bike running "on the map"?

[RFW]

The other day I was checking my TPS and noticed that the voltage reading stopped increasing at a certain amount of throttle. Don't recall whether it was a quarter throttle, half, or what. Got me thinking though, and perhaps someone here can educate me. Under what conditions does the Motronic switch from closed loop to open loop running? At changing throttle, over half throttle, over 4500 RPM...? When is the bike running "on the map"?

It runs closed loop a)When the O2 sensor has heated up enough to operate (a few minutes after starting), and b)at all throttle openings short of "wide open throttle", and c)Under closed throttle deceleration when all fuel is cut off. Wide open throttle means slightly different things depending on the engine, but is generally at least 3/4 throttle.

 

As for when it is on the map, it is always on the map. The map is calibrated to richen the mixture at WOT. More correctly, the mixture set by the map is constantly fine-tuned using feedback from the O2 sensor, in order to keep the mixture stoichiometric, except under conditions a) and b)above.

[T__]

AZBaldur, it sounds like you were just monitoring the low speed (low throttle opening) part of the TPS.. There is a second part of that TPS that measures the second half of the throttle opening curve (basically ½ throttle up to WOT)..

 

As far as closed loop vs open loop? There are a lot of variables on that one but basically from what I have seen on my 1100 & 1150 is once it has been started & run for a few minutes it goes closed loop at warm idle & stays closed loop in the low to moderate throttle opening & steady cruise range.. It seems to go open loop on almost any throttle opening or closing change, as well a high RPM’s, high power demand, or high vehicle speed.. I’m sure there are many other operational parameters than the simplistic stated above..

 

Twisty

[smiller]

What Twisty said. There are different contacts for different parts of the TPS range, but the entire range of throttle position (from closed to wide open) is measured and fed to the ECU as a resistance value.

[ShovelStrokeEd]

It has been my understanding that closed loop operation as it applies to the O2 sensor is only in effect at or below 13% throttle opening. Above that, the system returns to open loop mode or n-alpha where throttle position and engine RPM are the primary source of injector pulse width calculations. As stated earlier, it runs on the map all the time. My information comes from a couple of guys I know from the R1100S board who have actually gone to school for the Motronic system, not BMW school, Bosch school.

[smiller]

It has been my understanding that closed loop operation as it applies to the O2 sensor is only in effect at or below 13% throttle opening.

But that doesn't seem to make sense in that (since at freeway cruise you are running way more than 13% throttle position) it would mean that the bike is running open-loop the vast majority of the time and closed-loop a small amount of time... which is the opposite of how it should work. Also, the Power Commander maps show that open-loop occurs only at very low and high throttle openings and closed-loop operation occurs everywhere in-between, including most cruise conditions. This is exactly what one would expect.

 

I don't think that closed-loop occurs only at 13% throttle or below, someone must have something confused there... I think...

[ShovelStrokeEd]

Put a marked disk on your throttle and see just how little of the throttle you need to maintain say, 80 mph in high gear. I'd be greatly surprised if it's more than 10%. I could be wrong about the 13%, just a number I pulled from memory, maybe it's more like 20% but I'm pretty sure it is the case.

 

Think of it this way, Seth. The O2 sensor really can't cope with anything other than steady throttle conditions. In fact, it does poorly under either acceleration or deceleration. The necessary enrichment required for even moderate accleration with good throttle response will just rail it to full rich indication. The last thing you want when you add some throttle is for the motor to bog down due to a lean condition.

 

That disk can be useful if you decide to do some long economy runs. Learning to just squeeze the throttle enough to get to speed is one of the tricks.

[smiller]

Put a marked disk on your throttle and see just how little of the throttle you need to maintain say, 80 mph in high gear.

As a matter of coincidence I recently did just that, or something like it in an attempt to find out where I was in the map at cruise speeds. At 70-80 mph the throttle position is way more than 18%, more like 40%.

 

The O2 sensor really can't cope with anything other than steady throttle conditions.

Agreed, and agreed that the ECU is in open loop at any other time than steady throttle conditions. But there is definitely not a cutoff of closed-loop operation above 18% throttle as that would mean that the bike couldn't run in closed loop under even under a steady throttle position at cruise speeds, and I think you'll agreee that is precisely the mode of operation that closed-loop is intended for.

 

Certainly there is a point in the map when the bike goes strictly open loop, but that is more like 70% throttle, not 18%.

[ShovelStrokeEd]

40%? Really? I'm leaving in a couple of hours to ride up to Oak Ridge. I'm taking the Blackbird but, it is geared exactly the same as my 1100S. I'll fashion a disk and let you know what I find.

[T__]

It appears you guys are using hand grip rotational position to equate to throttle plate opening degrees.. They are not a one to one ratio..

 

13°-15° (throttle plate opening degrees) or around 5000 RPM's seems like it would be a likely place that the Motronic 2.4 would give up lambda control & the injection pulse time become strictly determined by the RPM and throttle position (alpha/N) modulated by intake air & engine operating temperature .. The EPA regs on motorcycles are a lot less stringent than automobiles (at least until 2010) so they are probably only worried about converter operation/protection not low emissions at high operating speeds..

 

In monitoring the lambda cross counts on both my 1100R & 1150RT that throttle plate opening & engine RPM would be close to where I observed the lack of O2 toggling..

 

One of these days I will find access to the Motronic tuning hand book & become a little more knowledgeable on these Motronic systems..

 

Twisty

[Don_Eilenberger]

Twisty wrote:

One of these days I will find access to the Motronic tuning hand book & become a little more knowledgeable on these Motronic systems..

 

Twisty

I'd recommend rather highly - "Bosch Fuel Injection" by Robert Probst. Great book explaining how all the Bosch systems work (up to about 5 years ago..) with good explainations of theory for each system, and then implimentation for each system. Someplace in that book it mentions 4,000 RPM as being the switchover from O2-closed loop to open-loop-map operation IIRC. Since the design of the oilhead Motronic was probably completed around 1991-2 I think the theory is probably well covered in Probst's book.

 

I'll have to dig out my copy and see what he says about an un-metered-air system with O2 sensor and see what he says about open-close-loop crossover.

[smiller]

It appears you guys are using hand grip rotational position to equate to throttle plate opening degrees.

I marked the position on the throttle grip by referencing the actual throttle position (in percent) as displayed on the PC interface of the Power Commander. I don't remember how linear it all was or wasn't, but I do know that the tick marks did accurately represent the actual throttle position.

 

Also, the gray area in the Power Commander maps that represent closed-loop operation (and which were derived from testing the bikes and observing ECU operation on the dyno) show that open loop operation occurs at low and high throttle settings and the meat of the powerband is closed-loop (just as one would expect.) The closed-loop area varies for different R1100/R1150 models, but always terminates at rather high throttle positions near WOT. I'm not trying to represent Power Commander data as some kind of oracle of absoulte truth but it does at least represent actual measurements.

 

A screenshot of the R1150RT map is attached and it is kind of what I would expect to see in that closed loop operation occurs during cruise conditions (and these throttle position/rpm ranges do seem to correlate with R11xx cruise operation as far as my limited testing indicates.) In any event it doesn't seem logical (to me) that closed-loop operation could have a hard termination at only 18% throttle.

 

Well, maybe on a Blackbird...

[T__]

It appears you guys are using hand grip rotational position to equate to throttle plate opening degrees.

I marked the position on the throttle grip by referencing the actual throttle position (in percent) as displayed on the PC interface of the Power Commander. I don't remember how linear it all was or wasn't, but I do know that the tick marks did accurately represent the actual throttle position.

 

Also, the gray area in the Power Commander maps that represent closed-loop operation (and which were derived from testing the bikes and observing ECU operation on the dyno) show that open loop operation occurs at low and high throttle settings and the meat of the powerband is closed-loop (just as one would expect.) The closed-loop area varies for different R1100/R1150 models, but always terminates at rather high throttle positions near WOT. I'm not trying to represent Power Commander data as some kind of oracle of absoulte truth but it does at least represent actual measurements.

 

A screenshot of the R1150RT map is attached and it is kind of what I would expect to see in that closed loop operation occurs during cruise conditions (and these throttle position/rpm ranges do seem to correlate with R11xx cruise operation as far as my limited testing indicates.) In any event it doesn't seem logical (to me) that closed-loop operation could have a hard termination at only 18% throttle.

 

Well, maybe on a Blackbird...

 

I guess I'm still not sure we are all on the same page here.. You are talking percent, I’m referring to degrees of throttle opining,, it appears Ed is talking degrees of throttle opening also..

 

While I’m not real sure what the actual throttle PLATE opening is from closed, I would imagine it is somewhere around 80° or so.. So we are talking 13°-15° of 80° (guess on my part here).. You are probably talking about 18% of a possible 100% throttle opening..

Due to the TBI shaft cam shapes I don’t think the hand grip is a linier control of the actual throttle plates anyhow..

 

Not trying to start an argument or criticized anybody here just trying to get us all on the same page so we can understand how this system operates..

 

Twisty

[smiller]

I guess I'm still not sure we are all on the same page here.. You are talking percent, I’m referring to degrees of throttle opining,, it appears Ed is talking degrees of throttle opening also..

Ah, OK, good point. But it seems to be more practical to discuss in terms of throttle position (percent of WOT) than degrees of throttle opening since the former is both more easily measurable and is the figure that the ECU actually sees. Also since Ed mentioned physically marking the throttle grip I assumed that he must also have been talking about throttle position vs. actual throttle opening because as you say the latter may not be very linear and thus not really subject to accurate meausure just by looking at the throttle grip.

 

But yes, in any event it does help to be talking about the same thing, thanks for clearing that up.

[nrp]

My theoretical understanding of these systems is that the injector pulse length map is only slowly updated based on RPM, air temp, TPS, and O2 history info obtained from the engine after it has warmed up.

 

But how slowly? Is it like seconds, minutes, hours, or even days?

 

Keep the discussions going. This is fascinating info! THX

[T__]

My theoretical understanding of these systems is that the injector pulse length map is only slowly updated based on RPM, air temp, TPS, and O2 history info obtained from the engine after it has warmed up.

 

But how slowly? Is it like seconds, minutes, hours, or even days?

 

Keep the discussions going. This is fascinating info! THX

 

Nrp, can't comment on how fast the Motronic updates it’s sensor inputs but when I was monitoring the 02 cross counts on my 1100 & 1150 I was seeing an 02 cross count swing at around 2 second intervals so it looked like the computer was ramping the fuel pulse width up & down to achieve an 02 crossover response at around 2 second intervals.. I never actually measured the cross count interval but 2 seconds would be a close guess.. Obviously TPS input would be looked at anytime there was any measurable throttle position movement.. Engine RPM would have to be kept track of pretty closely & fairly often but air intake temp & engine oil temp could be updated slowly as those change fairly slowly in operation..

 

Twisty

[AZBaldur]

Fascinating info indeed, and a special thanks to smiller for attaching that Power Commander fuel map image. If I understand it correctly, it looks like the PC is only adding a little extra fuel (compared to the stock Motronic map) at 80% throttle between 3000-4000rpm, and again at 100% throttle between 7000-8500rpm... is that right? What do the "8-9" numbers represent?

 

Also, and this is a bit of a tangent, but does anyone know which fuel maps are stored in my Motronic 2.4, and which terminals to connect to activate them?

 

Much obliged for the replies so far.

[smiller]

If I understand it correctly, it looks like the PC is only adding a little extra fuel (compared to the stock Motronic map) at 80% throttle between 3000-4000rpm, and again at 100% throttle between 7000-8500rpm... is that right? What do the "8-9" numbers represent?

You understand it correctly. The numbers represent additional fuel relative to the open-loop stock map, 8 = 8% richer. A negative number would indicate a leaner setting. The other thing that the PC would be doing in this case is maintaining the closed-loop mixture at 13.8:1, vs. the stock setting which would probably be much leaner. (The closed-loop mixture is set on one of the pull-down menus and is not visible on the screenshot.)

[Joe Frickin' Friday]

Put a marked disk on your throttle and see just how little of the throttle you need to maintain say, 80 mph in high gear.

As a matter of coincidence I recently did just that, or something like it in an attempt to find out where I was in the map at cruise speeds. At 70-80 mph the throttle position is way more than 18%, more like 40%.

 

This agrees surprisingly well with a (really) rough estimate that I recently did.

 

An RT's top speed is ~130 MPH; this is at wide-open throttle, or max load. Aero drag varies with the square of speed, so if you drop to 80 MPH, your drag (and therefore engine load, and therefore throttle opening) should be 80^2/130^2 = 38% of max. IOW, you would expect a throttle opening of around 38% when cruising @ 80 MPH.

 

Monitoring the O2 sensor to see when the engine is/isn't in closed-loop operation has been on my to-do/wish list for quite a while now. It would be easy enough to see with a DVM taped to the fuel tank: a steady signal of about .9 volts indicates a constant rich-of-stoich condition (open-loop), while a signal that oscillates between .9 and .2 volts (a couple of times per second) indicates closed-loop operation.

[AZBaldur]

The other thing that the PC would be doing in this case is maintaining the closed-loop mixture at 13.8:1, vs. the stock setting which would probably be much leaner. (The closed-loop mixture is set on one of the pull-down menus and is not visible on the screenshot.)

Ah, there had to be more to it, thanks for clearing that up. I read somewhere that the stock closed-loop mixture ratio is 14.7:1, i.e. exactly stoichiometric, which is optimal for the cat converter but does driveability no favors. A slightly richer across-the-board closed-loop mixture of 13.8:1 would help driveability, at some cost to cat converter efficiency.

 

I found a partial answer to my own "tangent" question above. The following are CCP plugs that select two different maps stored in the Motronic 2.4:

 

• Gold CCP - for R1150GS w/cat, stock - connects pins 30 and 87 - BMW part # 61368366282
• Grey CCP - for R1150GS w/cat, low octane fuel (91 ROZ) - connects pins 30, 86 and 87 - BMW part # 61368366623

The R1150R has the GS motor, so it'd be interesting to try the grey CCP option and see how the bike responds.

[ShovelStrokeEd]

Lessee,

80^2/180^2=19.7% OK, that is about right for my 1100XX.

I'll check my TPS voltage when I get home to be sure. There might be some difference related to the actual power the bike is capable of producing at part throttle. A real 80 mph on my bike in high gear (gearing is almost identical to that on my 1100S) my motor is only running at a little over 4800 RPM so it is well down from its power peak which occurs at about 9700 RPM. It also makes enough power to go past its peak in speed with stock gearing which will also skew things a little. Red line starts at 10800 RPM, which it will pull and the rev limiter kicks in at 11500, which my bike will not pull so actual top speed is ca 182 mph.

[AZBaldur]

smiller, how much did driveability improve after you installed the Power Commander? I use my bike primarily for commuting and encounter fairly heavy traffic regularly. I'm particularly interested in the off-on throttle transition and steady low-RPM running in 1st and 2nd gear.

[smiller]

smiller, how much did driveability improve after you installed the Power Commander?

A lot. Prior to installation I had learned all the tricks to keeping the bike tuned properly tuned properly so the surging wasn't really too bad, but like almost every R1100/1150 it was still present to some degree. After installing the PC, setting the mixture to 13.8:1, and richening the mixture a bit at low engine speeds and throttle openings there is no trace of surge left. Also much smoother accelerating from a stop through first gear at small throttle openings (some roughness there that I just couldn't get rid of had always bothered me), and better throttle response at cruise. Feels like a different bike now, in those respects at least. But none of that is unexpected as it's no secret that the engine runs on the lean edge at stock settings, so no surprise that a little extra fuel would make a positive difference. This is similar to what the Techlusion or removing the CCP does (although I didn't find removing the CCP to be nearly as effective), but I wasn't too happy about changing the mixture in the blind and liked the complete control that the PC provides.