Alternate R1200 synchronization procedure

[MarcS]

Here's a procedure that should achieve the same results as a full synchronization by a dealership using the BMW computer and tools. It's inspired by the procedure in the RepROM, easy and it seems to work well. Anyone wanna give it a shot and give me some feedback?

 

The BMW procedure does not adjust the idle mixture/synchronization. This is done automatically, constantly, in real time by your bike, through the use of a servo on each throttle body that controls residual airflow when the throttle valve is closed.

 

The BMW procedure only ensures that the throttle cables are synchronized. With the throttle cables synchronized, the valves open at exactly the same time. The idle control system makes up the difference in airflow between cylinders automatically.

 

The BMW procedure has 3 stages. First, the idle synchronization is verified. If idle synchronization is not achievable with the servos active, something is wrong with the engine or intake. The servos are disabled for the next two stages. The throttle cables are adjusted so that the valves open at the same time. The position of the idle servos does not matter as long as they do not change and there probably will be a vacuum differential between the two cylinders at idle and at load for this stage. The differential goes away when the servos are activated. Unlike a TB/carb sync on almost anything else, we're not trying to match the vacuums. The computer does that. We just just adjusting the throttle cables to account for stretch. The third stage is a verification that the synchronization has been achieved.

 

I don't think you can use a TwinMax with this procedure since it only measures relative differences. You need the absolute values for each cylinder, so that means using two traditional manometers. I use a Morgan CarbTune.

 

This is a simple one-beer operation...

 

You will need:

 

2 * 10mm open-ended wrenches. The smaller/thinner, the better.

2 * manometers (mercury sticks, a Morgan CarbTune, &c). You need to be able to obtain two separate absolute vacuum readings.

 

1. Start the motorcycle and warm it up to operating temperature. A 20 mile ride should suffice nicely. I think the folks with plenty of plastic have to take some off.

 

2. Clean the throttle bodies, especially around the nipples. Don't wanna get grit in the engine.

 

3. Attach the manometers to the throttle body nipples. You will need to remove the nipple cover on the right side and the hose from the emissions canister on the left first.

 

Nipple:

 

 

It's looking right at you...

 

Stage 1 - make sure idle sync is OK

 

4. Start the bike and let it idle. Blip the throttle a few times. The manometer average reading for both cylinders should be identical with the bike warmed up. On my bike, until the bike is warmed up, the readings will be a bit different (2cmHg or so).

 

 

Idle vacuum with idle control taking care of bidness

 

4a. If the idle sync isn't established, find out what is wrong. Check for intake obstructions or a messed up valve clearance. Maybe the idle control system failed.

 

Stage 2 - Adjust throttle cable synchronization

 

5. Turn the bike off. Unplug the connectors from both servos. You can do this by gently lifting on the tab at the top and pulling back on the connector. It's got a seal around the sides so it takes a tiny bit of effort. Restart the bike. It should idle normally. I wouldn't be surprised if it's a little bit rough but my bike with 6k miles on the clock has a perfect idle w/ the servos disconnected. There WILL be a difference in vacuum between the two cylinders. Probably. Pay attention to whether the left cylinder vacuum is higher or lower than the right cylinder vacuum.

 

Servo connector

Servo disconnected

 

 

Vacuum at idle with no idle control.

 

6. Slowly open the throttle. Both manometers should begin changing at the same time. The relative difference between the vacuums does not matter -- only that both columns begin moving at exactly the same time. If they do, skip to step 8.

 

7. Adjust the right-hand side cable.

Adjuster:

The locknut is the 10mm nut on the bottom. The adjuster nut is the 10mm on the top.

 

7a. Undo the locknut.

 

7b. If the column for the right cylinder moves BEFORE the column for the left cylinder, the cable needs more free play, so turn the adjuster nut clockwise. If the column for the right cylinder moves AFTER the column for the left cylinder, there is too much free play, so turn the adjuster counterclockwise. Small movements (1/8 a turn, or less) will suffice. Trial and error.

 

7c. When you get the two columns to change at the same time, tighten the locknut, and then re-check the setting to make sure you didn't lose your synchronization.

 

Stage 3 - Verification of settings

 

8. Rev the engine to 3000 rpms. Whichever cylinder had lower vacuum in step 6 should still have lower vacuum. If they don't then you botched step 7. The vacuums are not expected to match.

 

 

The vacuum readings are very close, but you can see the vacuum for the right cylinder is slightly lower than the vacuum for the left cylinder.

 

9. Turn off the bike, reconnect the servos, and restart the bike. Let it idle for a moment, blip the throttle a few times.

 

10. The vacuum readout at idle should be pretty much identical. The RepROM states within 15 mBar, or about 1 cmHg/10mmHg. If they are not, go back to step 4. Most likely you over-tightened the throttle cable and the right throttle valve is being held slightly open with the throttle closed.

 

11. The vacuum readout at 2k-3k-4k RPMs should be pretty much identical. Again within 15 mBar, or about 1 cmHg. If they are not, go back to step 7.

 

12. Disconnect manometers and reassemble motorcycle.

 

That's it! This may be a good time to apply dielectric grease to the various electrical connections. It's not necessary -- they have seals, and those have NEVER failed...

 

Now, why is this better than revving the bike to 4k in your garage and setting the sync by matching the vacuum levels? Because your engine will heat up rapidly at 4k RPMs and the idle control system will start dancing around. You will see identical vacuums but your throttle valves will not be opening at the same time. You could probably get the exact same result by actually riding your bike while you are adjusting it (so that the engine is homeostatic -- constant load, constant temperature, &c). But this is easier, safer and faster...

[bakerzdosen]

Uh... BOOKMARK.

 

Thanks for putting the effort into this writeup, I'll have to try that at my 18k.

 

Also, that Carbtune is looking better and better...

[Jim VonBaden]

Very interesting.

 

It seems to me that you could achieve the same thing by noting the offset on a Twinmax and using that with the high rpm sync.

 

Though, to be honest, I am not sure why it matters? If you are adjusting the butterflys at idle fot timing instead of vacuum sync, why not use the Twinmax to adjust the higher RPM settings, and adjust the idle static? So far it appears to work well for me.

 

Jim

[Ken H.]

Very interesting.

 

It seems to me that you could achieve the same thing by noting the offset on a Twinmax and using that with the high rpm sync.

 

Though, to be honest, I am not sure why it matters? If you are adjusting the butterflys at idle fot timing instead of vacuum sync, why not use the Twinmax to adjust the higher RPM settings, and adjust the idle static? So far it appears to work well for me.

 

Jim

Jim, that's pretty much the same thoughts I had about the procedure. Seems to me there are two key differences between this and the "standard" way. First he is advocating disconnecting the servos before starting the bike. Presuming this puts them in the same position as the GT-1 parks them (see the other thread going on about this), doing so could be part of it all.

 

Secondly is making the right-hand cable adjustment just off idle. Then later allow the (reconnected) servos to make the fine adjustments needed to sync on the fly. This would seem to agree with my experimentation that supported the idea that the servos affect sync at any RPM, not just idle. I mean after all, they're just automated air bypass ports, they have to affect it.

 

I have observed the "hunting" referred to while trying to do high RPM sync with the stepper servos connected, and wondered about how much heat changes, and by extension O2 sensors' output was causing the dog to chase the tail so to speak.

 

I agree though, an offset indication with a TwinMax or differential manometer will tell you the same thing as absolute readings with Carb Stixs. I mean, matched is matched no matter how you read it.

 

Interesting... The collective knowledge slowly grows...

[MarcS]

I happen to like where the servos are parked when the bike is off. The air bypass valve is kept somewhat open. I imagine this is a failsafe so that the bike still runs if a servo fails. It does not matter where the servos are stopped as long as the bike runs OK. It is simply required that they are both in the same position and they do not move.

 

The first stage of the official procedure checks the idle vacuum difference with the servos parked and specifies a tolerance of 25 mBar. This is slightly under 2 cmHg, which is a good guideline tolerance for many motorcycles. I imagine that this is also the range the servos can most effectively balance.

 

The servos do seem to affect mixture at any load condition but the amount of air relative to the throttle valve they let through becomes insignificant as the throttle valve opens. You yourself showed this in another thread when you said you saw different vacuum levels with the servos disconnected before and after the ignition was enabled. The servos are just stepper motors actuating a pin that opens a spring-loaded air bypass valve. The fact that BMW used stepper motors allows them to get around not having a MAF sensor -- they know how much air the bypass valve is letting through and can adjust the mixture accordingly.

 

The servo position may be determined by temperature and O2 sensors. Temperature is a given -- they need to be open a different amount to quickly start a bike, and I think the idle hunting/variance in vacuum between cylinders while a bike warms up is due to the servos not having O2 sensor data available yet (they are not hot enough).

 

What do you mean "matched is matched"? The reason I think a TwinMax would be suboptimal is that you're not actually adjusting anything to match the readings...you're just using the columns as an indicator that the throttle valves have begun to open. If you had 4 hands, you could probably do the sync with the bike off, a hand on the adjuster nut, a hand on each throttle cam, and a hand on the throttle.

 

You could possibly do it with two people, one touching the cams and directing, one cracking open the throttle and adjusting...if you ever get stuck in a parking lot needing to sync the TBs...

 

Random couple of points. First, some people advocate stretching the throttle cables by locking the throttle open overnight. Secondly, the BMSK should probably be reset after you perform this procedure. I have a hunch that the BMWS learns optimal servo positions as the bike runs, and running with the servos disconnected is probably going to confuse the heck out of the learned parameters. Also, if you've done a valve adjustment or changed anything related to the intake/exhaust, the servos will need to relearn idle synchronization positions before the bike will run well (in this case, maybe you should even reset the BMSK before you sync the TBs?)

 

How exactly DO you properly reset the BMSK? Sure, disconnect the battery overnight, but are there any things like fully opening and closing the throttle 30 times within six seconds or whatever the procedure is now that should be done as well? The RepROM implies there is a procedure that the GT1 walks the mechanic through.

[Ken H.]

Humm... If I understand you correctly then, you're saying that we are more concerned about when each side opens (movement of each side of the carb sticks at the same time) rather than how much each side opens (not concerned that the vacuum level is equal). Or to put it another way, we are only concerned with timing the physical opening of each butterfly together. Whether or not this results in even vacuum reading on one side vs. the other is irrelevant, the steppers take care of any remaining imbalance.

 

Correct to what you are saying?

 

If so, that certainly flys in the the face of traditional theory on carb or throttle body synchronization where we are more concerned with the effect (both sides have equal vacuum) than the physical aspects of the carbs or TBs that gets us there.

[MarcS]

The procedure is different but even with a normal TB/carb sync you are still trying to get the throttle valves to be open the same amount at any given throttle position.

[Jim VonBaden]

The procedure is different but even with a normal TB/carb sync you are still trying to get the throttle valves to be open the same amount at any given throttle position.

 

Not necessarily. Variations in flow are not going to allow a simple butterfly sync. Opening them at the same time, and same amounts will not necessarily, or even likely, cause an equal flow from side to side.

 

Variations in manufacture, burs in the intake, carbon build-up, and minute valve adjustment variations will cause the flow to vary side to side.

 

Adjusting in any way other than matching side to side flow, will un-sync the TB's and cause suboptimal performance.

 

Jim

[MarcS]

Airflow through butterfly valves is not linear with valve angle. Sure you can sync the vacuum by adjusting throttle valves for a single throttle position but you will lose that sync at other throttle positions, maybe badly so (Kawasaki Concours). I would fix it with pilot screws/idle mixture screws/idle control servos, and not butterfly valve position.

 

I couldn't do this procedure on a bike with mated carbs, for example, because all the valves are going to open at the same time, no matter what their relative position. So I use the vacuum reading as an indicator.

 

Entertain another thought. What state of tune would you consider a bike whose throttle valves did NOT open at the same time? Even if it didn't buzz while cruising?

[Emoto]

Ok, maybe I am missing something, but if the steppers take care of synch at idle, and one can synch the TBs at "above idle" (4k in your example) then what is it about your procedure that would make me want to adopt it? I mean, do I really care if they open at precisely the same instant?

 

Thanks,

 

Bob

[Ken H.]

I happen to like where the servos are parked when the bike is off. The air bypass valve is kept somewhat open. I imagine this is a failsafe so that the bike still runs if a servo fails. It does not matter where the servos are stopped as long as the bike runs OK. It is simply required that they are both in the same position and they do not move.

Marc,

 

This thing about the steeper motors/servos parking in a specific spot when the bike is turned off intrigued me. I knew they went through a specific initialization procedure when the bike is first powered on, but not about the same when the bike is turned off.

 

So today I did some measuring. I rode the bike all last week and haven't turned it on since. So today, still without turning it on I pulled the steppers and measured the position of the piston / air bypass. Sure enough, both of them were in the exact same position - .840" extended.

 

Next I put them back in, reconnected them and powered the bike on. But did not start it. I then disconnected them with the power on. Then removed them again and again measured the position of the pistons. This time they both measured - .658" extended. More open, more air bypass than in the bike off position.

 

So I don't know if either of these positions, power on, or power off, correlate to the position that the dealer's GT1 places them in prior to doing a cable sync. but it does seem to add some credence to the argument that they should both be in the same position prior to doing a cable sync., and the best way we have to get them there is to disconnect them prior to starting the bike.

 

My just above idle cable adjustment sync. was way off compared to where I had set it with the servos connected and the bike running. And I had been suspicious of it anyway as I had had to turn the right cable adjuster almost all the way in. Now, with the steppers disconnected while adjusting the right cable, its adjuster position is more in its mid-range. When I was done fiddling with the cable, I shut the bike back down and reconnected the steppers of course.

 

It will be interesting to see how the bike runs this week.

[MarcS]

Ken, cool! Did you try and match vacuums with this sync or did you follow the alternate procedure and just make them open at the same time?

[Ken H.]

Ken, cool! Did you try and match vacuums with this sync or did you follow the alternate procedure and just make them open at the same time?

Well to be honest with you, I'm not sure I understand the distinction. If the vacuums match, that is my manometer stays balanced as each TB's butterfly opens, as the throttle is raised off of idle, to me that means they are opening at the same time. And just as importantly opening an equal amount.

 

If that wasn't so, my manometer would swing dramatically to one side as one TB opened sooner than the other, and disrupted the balance.

 

Or what am I missing in your distinction between equal opening and equal vacuum?

[Jim VonBaden]

Ken, cool! Did you try and match vacuums with this sync or did you follow the alternate procedure and just make them open at the same time?

Well to be honest with you, I'm not sure I understand the distinction. If the vacuums match, that is my manometer stays balanced as each TB's butterfly opens, as the throttle is raised off of idle, to me that means they are opening at the same time. And just as importantly opening an equal amount.

 

If that wasn't so, my manometer would swing dramatically to one side as one TB opened sooner than the other, and disrupted the balance.

 

Or what am I missing in your distinction between equal opening and equal vacuum?

 

I'm about to try this tomorrow. I'll let you know how it goes.

 

Jim

[Emoto]

Guys, I still don't see the need for this, in terms of real world results. Can someone clue me in please?

 

Thanks,

 

Bob

[Marty Hill]

Does this mean my bikes have been running badly all these years?

[Ken H.]

Guys, I still don't see the need for this, in terms of real world results. Can someone clue me in please?

Well here's the thing / theory...

 

First you have to accept that the stepper controlled air bypass ports affect sync at any RPM, not just idle. And history with the LBSs on the oil heads support that they do. A change in the position of a LBS on an oil head will affect sync. at any RPM.

 

With the steppers we know that they are adjusting sync on the fly based on inputs to the BMS-K module, primarily from , we think, the O2 sensors.

 

The challenge becomes when we try to adjust above idle cable sync. especially if we are doing it at higher RPMs (like 3-4K) and the steppers are trying to make adjustments at the same time we are. It is certainly possible that the steppers are trying to move sync. one way, while we are trying to move it the opposite way with the cable adjustment. Especially as we rev. the bike, return it to idle, repeat, airflow amount change, the engine goes in and out of closed loop, etc., all changing the values the BMS-K is trying to use to set sync. Which I think accounts for a lot of the “hunt” and inconsistency that we see when trying to sync. a hex head ourselves.

 

Another potential issue is that the stepper only has so much adjustment range to work with. What happens if the cable position for best sync. we happen to end up with also results in the stepper being near one end or the other of its range? Then later, if it needs to make a dynamic adjustment in the same direction, it can’t. And the bike runs poorer than it could have.

 

All of this adds up to why the dealer’s GT-1 diagnostics machine “parks” the steppers in a fixed and equal (but unknown to us DIYers) position prior to doing a cable sync. Presumed to be somewhere in the steppers mid-range. Think of it as the same thing we do when we set the LBSs equally to 1 ½ turns out (or whatever) on an oil head prior to sync’ing the cable on one of those.

 

What we are trying to do with our experimentation is to duplicate that, so the steppers, and thus the air bypass ports are equal, prior to sync’ing the cables. Then later when they are put back on-line so to speak, they are maximized to be able to keep the engine in best sync on the fly.

 

Does all this affect the real world? Well maybe, maybe not. But part of what we are doing is trying to advance the knowledge of what we know about the system and the optimum way to deal with it ourselves.

 

Am I close Marc?

[Albert]

Great explanation Ken. I'm learning all along the way here. I hope you guys keep it up, this is fun! Thanks to all.

[jwurbel]

Boy, do you guys know your stuff.

 

Having recently bought an R12 I was prepared to do my own wrenching as I always have in the past. I have Carb Stix that I have used with my 72 CB 750 to sync those carbs and assumed I could use the same procedure with the BMW. After reading all the posts, I am not so sure. Am I better off letting the dealer do it after I lash the valves? Or are all the ways you folks have listed possible alternatives? In short, I am a little confused.

[MarcS]

Ok, cool.

 

On my bike the relative vacuum decreases as RPMs rise.

 

With an oil manometer or a TwinMax or similar you don't want the vacuum differential to stay the same, you want it to decrease slightly. But how much I do not know...

 

I think we are definately on the same page w/ this procedure though.

[Ken H.]

Boy, do you guys know your stuff.

 

Having recently bought an R12 I was prepared to do my own wrenching as I always have in the past. I have Carb Stix that I have used with my 72 CB 750 to sync those carbs and assumed I could use the same procedure with the BMW. After reading all the posts, I am not so sure. Am I better off letting the dealer do it after I lash the valves? Or are all the ways you folks have listed possible alternatives? In short, I am a little confused.

Well there's a couple different trains of thought / opinions floating around.

 

One says - pretend the stepppers aren't there. Hook up your favorite sync. measuring tool, pick your favorite RPM, hold the bike there, then adjust to in sync. with the right throttle body cable and you're done.

 

The other train of thought, the one we're (mostly) discussing in this thread, is that you can't ignore the steppers. Like the factory procedure you have to some how set them the same before sync'ing the bike with the the cable(s). Clear Marc and myself are in this camp.

 

The dealer's diagnostic machine, the GT-1 when connected to the bike parks, holds, the steppers in a specific position while the cable adjustment is done. Given that few us can afford to buy a $30K GT-1 (even if we could) the ongoing research / discussion / debate is how best to simulate what the GT-1 does. Or for that matter, to figure out - does it matter?

 

Clear as mud?

 

One thing is for sure, you CAN sync. a hexhead yourself. What procedure to do so is still a bit open for discussion.

[Emoto]

Guys, I still don't see the need for this, in terms of real world results. Can someone clue me in please?

Well here's the thing / theory...

 

First you have to accept that the stepper controlled air bypass ports affect sync at any RPM, not just idle. And history with the LBSs on the oil heads support that they do. A change in the position of a LBS on an oil head will affect sync. at any RPM.

 

With the steppers we know that they are adjusting sync on the fly based on inputs to the BMS-K module, primarily from , we think, the O2 sensors.

 

The challenge becomes when we try to adjust above idle cable sync. especially if we are doing it at higher RPMs (like 3-4K) and the steppers are trying to make adjustments at the same time we are. It is certainly possible that the steppers are trying to move sync. one way, while we are trying to move it the opposite way with the cable adjustment. Especially as we rev. the bike, return it to idle, repeat, airflow amount change, the engine goes in and out of closed loop, etc., all changing the values the BMS-K is trying to use to set sync. Which I think accounts for a lot of the “hunt” and inconsistency that we see when trying to sync. a hex head ourselves.

 

Another potential issue is that the stepper only has so much adjustment range to work with. What happens if the cable position for best sync. we happen to end up with also results in the stepper being near one end or the other of its range? Then later, if it needs to make a dynamic adjustment in the same direction, it can’t. And the bike runs poorer than it could have.

 

All of this adds up to why the dealer’s GT-1 diagnostics machine “parks” the steppers in a fixed and equal (but unknown to us DIYers) position prior to doing a cable sync. Presumed to be somewhere in the steppers mid-range. Think of it as the same thing we do when we set the LBSs equally to 1 ½ turns out (or whatever) on an oil head prior to sync’ing the cable on one of those.

 

What we are trying to do with our experimentation is to duplicate that, so the steppers, and thus the air bypass ports are equal, prior to sync’ing the cables. Then later when they are put back on-line so to speak, they are maximized to be able to keep the engine in best sync on the fly.

 

Does all this affect the real world? Well maybe, maybe not. But part of what we are doing is trying to advance the knowledge of what we know about the system and the optimum way to deal with it ourselves.

 

Am I close Marc?

 

Thanks, Ken. I guess what I was wondering is whether going through all of the bother was worth it. I suppose if you synch it the old way (say, at 4k using carb sticks and BTW I use a throttle lock so it doesn't keep dipping back to idle) and then ride it around and see how it feels, THEN re-do the synch in this new complex manner and then ride it again and see if it feels any different...

[russg]

Emoto, etc..

 

I think one of Marc/Ken's primary theories here is that even if you got it wrong to some degree doing it the "old way" the electronics via the steppers would compensate. Therefore, unless you really screwed up, you likely couldn't tell except in open loop mode.

 

Marc/Ken, I have a very good understanding of how modern computer controlled engines work and I believe your theory is correct. Many modern systems are completely fly by wire. (for those that don't know, your foot on the peddle isn't connected to a cable to the carb)

 

I would absolutely agree that adjustment would have to be done while the steppers are parked. I also have the same worry as Ken. In my estimation, the steppers would have limits each way. It would be best to have them parked midway when setting up?

 

During closed loop operation, the electronics are going to constantly adjust the fuel/air mixture.

 

Standard disclaimer: I've been known to not know crap...

[Ken H.]

It would be best to have them parked midway when setting up?

Yes. The challenge is that we DIYers don't have a way to know where is, or move them to; midway. Or for that matter, we don't even know where the dealer's equipment parks them.

 

There is about a year old thread here somewhere when I spent a bunch of time (and money ) trying to build a stepper motor controller to manually move them. While I was able to get to where I could move one, I still didn't know where to move it to / "park it." Then while experimenting with one of the steppers on the bench I drove the plunger out the end and ruined it. Had to buy a new one. I threw in the towel on the experiment.

 

What we really need is a cooperative dealer that would park them for us on a bike, let us then remove one and measure the position of the piston. But for some strange reason (can't imagine why ) dealers seem reluctant to let you "play" with their mega-bucks GT-1 though!

 

Absent that, based on Marc's observations and my measurements, parking them by disconnecting them after a shut down, and before starting the bike to do a cable sync, seems to be an approximation of midway in their total travel range. At the very least we now know that when doing it this way they are set equal.

[Jim VonBaden]

 

Thanks, Ken. I guess what I was wondering is whether going through all of the bother was worth it. I suppose if you synch it the old way (say, at 4k using carb sticks and BTW I use a throttle lock so it doesn't keep dipping back to idle) and then ride it around and see how it feels, THEN re-do the synch in this new complex manner and then ride it again and see if it feels any different...

 

I agree it is very complex, and I have so far had good results with doing it the old fashioned way, minues the idle adjustment.

 

However, I am willing to concede that there might be a better way, and am willing to try it to find out.

 

I wasn't able to do it yesterday, as I was recovering from a weekend electrical tech day, but will give it a shot on Friday when I am off.

 

Jim

[Emoto]

 

Thanks, Ken. I guess what I was wondering is whether going through all of the bother was worth it. I suppose if you synch it the old way (say, at 4k using carb sticks and BTW I use a throttle lock so it doesn't keep dipping back to idle) and then ride it around and see how it feels, THEN re-do the synch in this new complex manner and then ride it again and see if it feels any different...

 

I agree it is very complex, and I have so far had good results with doing it the old fashioned way, minues the idle adjustment.

 

However, I am willing to concede that there might be a better way, and am willing to try it to find out.

 

I wasn't able to do it yesterday, as I was recovering from a weekend electrical tech day, but will give it a shot on Friday when I am off.

 

Jim

 

Jim, I look forward to hearing what you think. So many times, it seems that all of the theory imaginable doesn't add up to much in the real world. This is not to say that the theory is incorrect.

[MarcS]

Theory? Why don't you try it and report back. Not counting time to warm your bike up, it'll probably take 20 minutes start to finish.

 

If your synchronization is correct per BMW's specifications, you shouldn't need to adjust anything at all.

[jwurbel]

This post is getting more facinating by the day. But for a new BMW purchaser, could you tell me where I can find the "old fashioned way" your refer to? Thanks

[Guest]

Here is an experiment to try.

 

Leave the stepper motors hooked up and start the engine.

 

Using a fan and throttle lock, hold the engine at 2k rpm.

 

Now adjust the right side throttle cable in one direction until you can disturb the balance (lets say the "in" direction. Now count turns and adjust the cable in the out direction until you disturb the balance again. The mid point back "in" should be the center point for the steppers.

[Ken H.]

Here is an experiment to try.

 

Leave the stepper motors hooked up and start the engine.

 

Using a fan and throttle lock, hold the engine at 2k rpm.

 

Now adjust the right side throttle cable in one direction until you can disturb the balance (lets say the "in" direction. Now count turns and adjust the cable in the out direction until you disturb the balance again. The mid point back "in" should be the center point for the steppers.

Sorry, have to disagree, Adjusting the right cable will affect the sync no matter what the steppers are doing. They don't react that fast. Remember they're primary input to measure in sync is the O2 sensors. They may attempt to adjust for your miss-adjusted cable over riding time, but the dynamic stepper range seeking you are advocating unfortunately won't work. Good thought though!

[MarcS]

You think they are driven mostly off the O2 sensor? What's your evidence?

 

I initially assumed they were driven primarily by engine temp. The initial position at startup has to be engine temp since there is no other useful sensor data, unless the startup mixture is altered ONLY through the amount of fuel injected -- but what about fast idle?

[MarcS]

Just like any other bike except slightly different -- you sync at 4k instead of idle/just off idle.

 

Hook up vacuum gauges, rev engine to 4k, hold it there with a throttle lock, adjust the right side throttle cable length until the vacuum gauges show that both cylinders have the same vacuum.

[Ken H.]

You think they are driven mostly off the O2 sensor? What's your evidence?

Evidence, evidence? Don't need no stinkin' "evidence"!

 

Seriously though, when the system is trying to adjust sync, it is trying to adjust balance between the two cylinders. What do we have that can measure the "product" of each cylinder independently, and then compare them to each other to determine what adjustment the steppers need to make? The only things I can think of on the hexhead are the two O2 sensors and the two knock sensors.

 

The single engine temperature sensor will not give the BMS-K module information on what each cylinder is doing. I suppose some argument could be made that if the engine is out of sync the engine temp will go up (or down?) but it's going to be doing that anyway from other factors. So I think that it uses this for "am I in or out of syncronization?" sensing, is a stretch.

 

There has been some discussion / speculation that the BMS-K engine management module on the hexhead uses crankshaft position sensor pulse duration variances to determine if each power pulse is equal, and thus is the engine in sync or not. I'm not sure the math works though. Given that the frequency of the output of the crankshaft sensor is several multiples higher than that of the power pulses. E.g. - there are multiple notches on the crankshaft sensor ring, how can a single side's power pulse difference vs. the other side's, multiplex across the frequency of the crankshaft sensor? So personally I have a tendency to discount that theory also.

 

Which brings me back full circle to the O2 and possibly the knock sensors.

 

I think it's more likely that the engine temp is used along with the O2 sensors to determine when to put the system into closed-loop mode. At initial start, before the O2 sensors become operational (remember these are not pre-heated O2 sensors) I believe, like many engines, it is in open-loop mode and the steppers are out of the loop. Or at least parked equally in an initial position that combined with a wider, enriching fuel delivery pulse, gives you fast idle. Then indeed the BMS-K is probably looking at only the engine temp sensor. Once the bike starts to warm, and the O2 sensors are to operating temp, the system goes to close-loop, idles down, the steppers come on-line, etc.

 

But then I'm no engine management systems expert either, and BMW isn't talking. So this is indeed all speculation.

[MarcS]

Right, ok -- brain fart when I posted that last post

 

The CPU should be able to ensure that both cylinders suck in the same amount of fuel, and then the CPU can adjust the amount of air sucked in until the mixture is perfect for both cylinders. If it's correct to assume that two cylinders burning the same amount of fuel produce the same amount of power, then we're OK...

 

Will this necessarily account for differences in efficiency due to wear and differences in valve lift, though?

[Ken H.]

Will this necessarily account for differences in efficiency due to wear and differences in valve lift, though?

Well it would seem like that would be one of the advantages of the stepper controlled system. If it is looking at combustion "outputs" (via some method or another) for sensing and adjusting cylinder balance, then almost by accident it should see and correct for other causes of imbalance beyond just say a throttle body cable out of adjustment.

 

Now to what extent, or for that matter over what range of 'other' issues, is anybody's guess. For example if you pulled one of the four spark plugs, it probably wouldn't be able to compensate for that!

 

OTOH, a valve adjustment that is out a minor bit causing one side to breath a tiny bit poor than the other? The system, in theory, it seems to me would see that and adjust to compensate for it.

 

But then I'm not a BMW engineer and I don't play one on TV, so what do I know?

[GelandeStrasseWayne]

One more question. If the servoes are active all the time, then it seems like it does not matter how you have the cables adjusted, as long as they are not so far out of adjustment that the servoes cannot compensate. It seems the computer and servoes will define how the system is synchronized, not the exact cable ajustment (that we all so carefully make).

[Ken H.]

One more question. If the servoes are active all the time, then it seems like it does not matter how you have the cables adjusted, as long as they are not so far out of adjustment that the servoes cannot compensate. It seems the computer and servoes will define how the system is synchronized, not the exact cable ajustment (that we all so carefully make).

Well personal opinion, no. If the cables are out a bit, then yes, it seems to me that the stepper controlled air bypass ports could compensate. But I think there are two things working against an assumption that they could overcome much more than a little bit of cable induced syncronization error.

 

First is that the steppers can only mover so far before they hit the end of their travel. One direction or the other.

 

But more importantly, the ratio of airflow through the air bypass port is so small compared to the amount of air moving through the throttle body's throat itself. Thus changes made to the bypass ports to a point would be able to "fine tune" miss-adjustments to the basic sync, but if the cables are out very far, it would be beyond the ability for them to compensate. They just can't flow enough air if needed and driven fully open, or restrict enough even if driven fully closed by the steppers.

 

Which in my mind is why it is all the more important to "park" them equal, then still do a very accurate cable sync.

[ShovelStrokeEd]

Maybe a change in terminology might help clear up what seems to be a fairly foggy issue. On almost every EFI system I have worked on, these stepper motors and their attendant mechanism are called "Idle air control". Their flow rates and range of flow rate adjustment is suitable for controlling fast idle speed, normal idle speed and little else. Passages are small to begin with and occluded by the "needle" that moves in and out of the port. Under no-load conditions, they might provide enough air flow difference to put a water manometer about 1cm out of synch, even at speeds as high as 3 or 4K RPM. With the motor under load, I doubt you would see any difference or, for that matter, feel it. I can see BMW's decision to let them remain active during higher speed/load conditions but they really shouldn't have much effect. I know if I were designing the firmware for this, I'd probably make them inactive above say 10% throttle angle as by that point they wouldn't have any discernable effect on engine running and would, at least, be close to a proper setting when the throttle returned to idle.

[Jim VonBaden]

Well, I tried it, and found essentially no difference.

 

I ended up adjusting it the traditional way, and found it much smoother at 2500 and up by using the Twinmax and adjusting the cables at 3500 rpm.

 

This was after a full 24K service.

 

Jim

 

PS I will be posting a 24K pictorial soon.

[Ken H.]

Well Ed, yes and no. Here's a picture of the stepper servo and its piston. It is significantly bigger than what we might think of as a needle valve.

 

 

Still, I agree, the percentages of air flow via this path is small. But just like the LBS ports on a traditional oilhead TB, it does, IMHO, have at least some effect at higher RPM.

[Emoto]

Well, I tried it, and found essentially no difference.

 

Glad you tried it! I suspected that this would be the result.

 

Bob

[ShovelStrokeEd]

Without a scale for reference I can only make some assumptins but let us say that the pintle is 1/8" in diameter, probably with a taper, and the hole is 1/4". 3 and 6mm if you prefer. So, with the pintle retracted flush with the hole, you are left with an annulous of 1/4" OD x 1/8" ID. That gives a cross sectional area of around 0.036 in^2. Not very big at all. I'll have to fire up Autocad to figure out the area on a 50mm throttle body at an angle of 10 degrees open but I'm betting on an order of magnitude larger at least. Add the difference in depression between an engine under load and one at static RPM, say a 30% difference in vacuum, and I doubt they will have very much effect at all.

 

I do, btw, agree that the things should be parked to do the cable synch. I just don't think it matters all that much where they are parked so long as both are equal and the passages are clean.

[Ken H.]

I do, btw, agree that the things should be parked to do the cable synch. I just don't think it matters all that much where they are parked so long as both are equal and the passages are clean.

Which actually brings up an interesting point. Unlike the air bypass ports on the oil head TBs, the couple of times I've pulled the servo/stepper, on my 12GS the pistons pin and port have been perfectly clean. Haven't thought about why that is, what the difference is in the port location or design, but clearly BMW solved at least that issue.

[MarcS]

Is the piston actually directly exposed to fuel? I would assume not, since you wouldn't want varnish from fuel to gum it up over a few years of normal use.

 

Hey, how did you destroy a servo?

[Ken H.]

Is the piston actually directly exposed to fuel? I would assume not, since you wouldn't want varnish from fuel to gum it up over a few years of normal use.

 

Hey, how did you destroy a servo?

I have really dissected one of the new TBs enough to know, but I'd have to agree with your presumption.

 

I was experimenting with an external stepper controller to move the stepper. In one attempt I drove the piston out the end of the stepper and it wouldn't go back together. No way, no how. That was a $111 boo-boo.

[MarcS]

Wait, what?

 

I would assume the piston is basically a screw that gets turned by a stepper motor, threading itself in and out. You should be able to thread it all the way back in?

 

Do you still have the bad servo?

[Ken H.]

You should be able to thread it all the way back in?

Yeah, one would have thunk it, wouldn't one. Wasn't going to happen though.