• Welcome, Guest! We hope you enjoy the excellent technical knowledge, event information and discussions that the BMW MOA forum provides. Some forum content will be hidden from you if you remain logged out. If you want to view all content, please click the 'Log in' button above and enter your BMW MOA username and password.

    If you are not an MOA member, why not take the time to join the club, so you can enjoy posting on the forum, the BMW Owners News magazine, and all of the discounts and benefits the BMW MOA offers?

Oilhead surging

35751

Badger
I've owned a 2004 R1150RT twin spark (which I never should have sold!) that never had a surging problem. Also owned an R1100GS that surged like a chronic asthmatic; that was pretty much eliminated with the installation of a "Power Commander." I'm now in the market for another R1150RT or an R1100RT. Do the single spark R1150RT or the R1100RT models have a surge problem? Thanks.
 
All the single spark Oil heads can surge to some extent. How much of a problem it is depends on the bike and how sensitive you are to it.

In specific response to your question, I had a single spark 1150rt and it surged. I changed the cat code plug (donÔÇÖt remember to which one, and changed plugs to non-stock and she was fine after that.
 
I've owned a 2004 R1150RT twin spark (which I never should have sold!) that never had a surging problem. Also owned an R1100GS that surged like a chronic asthmatic; that was pretty much eliminated with the installation of a "Power Commander." I'm now in the market for another R1150RT or an R1100RT. Do the single spark R1150RT or the R1100RT models have a surge problem? Thanks.

Many riders report that the single-spark R1100 and R1150 have varying degrees of "surging", which is a broad topic. The dual spark 2004 R1150RT to a much lesser extent. There are a class of Lambda-shifters like the Innovate LC-1 which can remove the surging. It may also be that some plug 'n play devices will turn up to do the same job as the LC-1 but much more simply.
 
It was worse with the 1100s but they both have the issue; how badly depends on how well the assembly line (and then followed up by a competent tuner) set up the valves, the TPS (including the baseline Zero), and sync'd the throttle bodies. The cure is a combination (total summation) of tuning corrections. There are a jillion and a half articles and posts on various sites about this already; search this site with the window at the top right ("throttle body", "TPS"), or if the history got nuked with the last upgrade, go over to the Internet BMW Riders and look through the Oilhead Tech articles. http://www.ibmwr.org/
 
I have a '96 r1100rt. I am the second owner. No surging problems and frankly didn't know there were any until every once and awhile I see something posted on this web-board.
I have every single dealer and previous owner maint. logged from day one.
Straight stock, nothing has been altered or modified because of surging.
It has 72,000 miles.

Does anyone know when this started or was it just an anomaly with certain years (or half years) or models? Something must have changed.

mike wex
'96 r1100rt
 
Mike- earliest year oilheads (which your bike might well be) had a CCP (cat code plug) which created a slightly richer mixture, which effectively eliminated the surging. That was changed by '97 or so. Surging is generally understood to be a "hunting" by the EFI to achieve the correct A/F ratio. What functionally happens (others, probably Roger04RT, will give deeper analysis) is that the EFI adds more fuel, sees that it's too much, then gives too little, then has to add more, ad infinitum. Happens most at consistent throttle settings. Whe riding hard it does not surge as you are steadily adjusting/adding fuel yourself, or when rolling off as you are asking it to deliver minimum fuel anyway.
 
True,....and not quite. My 94 RS (built in Oct 93) had no CAT code plug at all. The very early Oilheads ran a bit more rich before the CAT code plug system was added. I never had any surging problems at all.

That said, I have ridden other Oilheads and never noticed a surging problem. But they do respond well to a Power Commander or Booster Plug or Techlusion, to richen up the mixture slightly.
 
Almost two decades of surge threads on the internet, starting on the "Big List"...What a crack up:rofl

IÔÇÖve owned 5 Oil heads; my surge dance card played out as follows;

96ÔÇÖ 1100RS, surged very bad at speeds between 30-50mph making around town stuff very unpleasant
99ÔÇÖ 1100S, no surge
01ÔÇÖ1150GS, no surge
03ÔÇÖ1150RT, minor surging, cure was easy
04ÔÇÖ 1100S, no surge
 
I've owned a 2004 R1150RT twin spark (which I never should have sold!) that never had a surging problem. Also owned an R1100GS that surged like a chronic asthmatic; that was pretty much eliminated with the installation of a "Power Commander." I'm now in the market for another R1150RT or an R1100RT. Do the single spark R1150RT or the R1100RT models have a surge problem? Thanks.

I would simply say, yes, they do. However, the fix is so easy that I would not hesitate to purchase the model you like best. And when you fix the underlying cause of surging, you get a big torque pickup between 2000 and 4000 RPMs.

Apart from normal maintenance, including a throttle body balance two more steps transform the R1100s and r1150s:

Clean the injectors and have them flow tested for balance .

Richen the closed loop AFR target with a Wideband O2 sensor like he LC-1 or one of the plug 'n play narrowband lambda shifters which will come on the market this year. (One of which I'm test riding now.)

RB
 
Last edited:
Many Thanks!

I've owned a 2004 R1150RT twin spark (which I never should have sold!) that never had a surging problem. Also owned an R1100GS that surged like a chronic asthmatic; that was pretty much eliminated with the installation of a "Power Commander." I'm now in the market for another R1150RT or an R1100RT. Do the single spark R1150RT or the R1100RT models have a surge problem? Thanks.

I really appreciate all the great input I got here ref the surging in single spark oilheads. Think I'll get on here more often to garner the vast experience of many riders. Mike.
 
My 2000 R1100RT was fine when new but gradually developed a viscous surge by 20,000 miles. Multiple attempts to get rid of this following other published procedures went nowhere, although I did not try disabling the cat plug or other processor controlled functions. After all, the system worked once, so why couldn't it work again?

My hunch was that surging could be caused by one cylinder (most likely the right one) dropping out from either too rich (unlikely) or more likely too lean operation at near closed throttle. The fuel injector pulse lengths are identical so the problem had to be in the airflow measurement and delivery system. The processor optimizes the fuel injector pulse length with the airflow as measured on the left with the throttle position sensor, but it also has to assume that the right airflow vs left TPS output voltage is close enough to the left side that a combustible mixture is assured on the right side. At high throttle openings this is reasonable to assume, but at small openings it depends on the equality of the BBS settings and the throttle stop screw settings, either of which can be affected by contamination, and both of which are functionally a setting trade-off.

BMW says not to mess with the throttle stops (an EPA certification requirement?), but the relationship of equal airflow to the single-TPS output on each side won't necessarily hold when there are two sources of air to each cylinder.

The acceptable range of combustible mixture is much wider than the 4% number suggested above, as poor carburetors and automatic chokes have proven for many years. According to Marks Engineering Handbook, it is over 2:1 ( i. e. 7:1 to ~19:1 A/F ratio). In other words, in our case, if the L-R airflow inequality exceeds ~2:1, there will be a right side miss - and I contend that is the source of the surge.

This approach completely eliminated the surge by carefully adjusting the intake valve clearances, thoroughly cleaning the throttle bodies and BBS ports, and resetting the throttle stops & big brass screws (BBSs) to equal positions. It was not necessary to disable or change any of the processor controlled mixture functions at all.

In summary, surge in my case was caused by an imbalance in the throttle stop screw /BBS settings created by throttle body contamination.
 
This approach completely eliminated the surge by carefully adjusting the intake valve clearances, thoroughly cleaning the throttle bodies and BBS ports, and resetting the throttle stops & big brass screws (BBSs) to equal positions. It was not necessary to disable or change any of the processor controlled mixture functions at all.

In summary, surge in my case was caused by an imbalance in the throttle stop screw /BBS settings created by throttle body contamination.

Interested to hear how you made sure the throttle stops and big brass screws were set to exact equal settings.

The only way I know of unless you have a flow bench setup is to zero the left side then use the TPS voltage to set it up for idle and then use vacuum to make the right side match the left side. All of this is done with the BB screws turned all the way in so they don't affect the initial zero zero procedure. Then turn them out to the factory setting of 1.5 turns out and again use vacuum to balance both sides followed by a throttle cable adjustment at higher RPM.

I've tuned 1100 engines pretty well using this method but never completely eliminated the dreaded lower RPM lower gear surge.
 
My 2000 R1100RT was fine when new but gradually developed a viscous surge by 20,000 miles. Multiple attempts to get rid of this following other published procedures went nowhere, although I did not try disabling the cat plug or other processor controlled functions. After all, the system worked once, so why couldn't it work again?

My hunch was that surging could be caused by one cylinder (most likely the right one) dropping out from either too rich (unlikely) or more likely too lean operation at near closed throttle. The fuel injector pulse lengths are identical so the problem had to be in the airflow measurement and delivery system. The processor optimizes the fuel injector pulse length with the airflow as measured on the left with the throttle position sensor, but it also has to assume that the right airflow vs left TPS output voltage is close enough to the left side that a combustible mixture is assured on the right side. At high throttle openings this is reasonable to assume, but at small openings it depends on the equality of the BBS settings and the throttle stop screw settings, either of which can be affected by contamination, and both of which are functionally a setting trade-off.

BMW says not to mess with the throttle stops (an EPA certification requirement?), but the relationship of equal airflow to the single-TPS output on each side won't necessarily hold when there are two sources of air to each cylinder.

The acceptable range of combustible mixture is much wider than the 4% number suggested above, as poor carburetors and automatic chokes have proven for many years. According to Marks Engineering Handbook, it is over 2:1 ( i. e. 7:1 to ~19:1 A/F ratio). In other words, in our case, if the L-R airflow inequality exceeds ~2:1, there will be a right side miss - and I contend that is the source of the surge.

This approach completely eliminated the surge by carefully adjusting the intake valve clearances, thoroughly cleaning the throttle bodies and BBS ports, and resetting the throttle stops & big brass screws (BBSs) to equal positions. It was not necessary to disable or change any of the processor controlled mixture functions at all.

In summary, surge in my case was caused by an imbalance in the throttle stop screw /BBS settings created by throttle body contamination.

But did you do a flow test on your injectors first? Dirt can easily affect flow & pattern. AMHIK
You say the surge developed so I assume it was not there originally. One injector not matching the other could easily cause your symptom. I would do that LONG before I messed with the stops. They are set at Bing on a flowmeter.
Don't encourage messing with those screws as most that lack the experience would be tempted to adjust them and most probably make matters worse. Especially if they did not have the required instruments.
 
...

The acceptable range of combustible mixture is much wider than the 4% number suggested above, as poor carburetors and automatic chokes have proven for many years. According to Marks Engineering Handbook, it is over 2:1 ( i. e. 7:1 to ~19:1 A/F ratio). In other words, in our case, if the L-R airflow inequality exceeds ~2:1, there will be a right side miss - and I contend that is the source of the surge.

This approach completely eliminated the surge by carefully adjusting the intake valve clearances, thoroughly cleaning the throttle bodies and BBS ports, and resetting the throttle stops & big brass screws (BBSs) to equal positions. It was not necessary to disable or change any of the processor controlled mixture functions at all.

In summary, surge in my case was caused by an imbalance in the throttle stop screw /BBS settings created by throttle body contamination.

Surging is an outcome that can have many causes. The description in the prior post is a good description of one half of the problem of light-load power imbalance. I will be the first to acknowledge that L/R air balance is an important part of getting the L/R cylinders to be power balanced. But it is only half the story, and it is the half we most focus on because we can measure it (intake manifold vacuum balance) and adjust it (BBS, throttle stops, valve clearances).

But the other half of the equation is fuel balance. This is easy to ignore because there is no way to easily measure it, and nothing to adjust. And it is wrong to believe that because the injector pulses are equal, that the amount of fuel or the quality of its atomization is equal.

At the speeds where surging is often experienced, throttle angles are low leading easily to air imbalance. But the injector pulse widths are also small, on the order of 2/1000's of a second. Of that 2mS, half of the time is the time to open the injector and the time to close the injector. So 1 mS is Open/Close and 1mS is spraying fuel. Of the 1 mS when the injector is spraying fuel, there can be different rates and different atomization pattern due to slight fouling and orifice sizes. In addition, as injectors age the open/close, or "dead time" as it's known, changes. This can easily lead to significant differences in fuel L/R and therefore power.

The notion in the prior post that combustion can take place between 7:1 and 19:1 is true, but wrong-minded. If the difference between cylinders was merely (let alone 7:1 or 19:1) 13.2:1 (roughly best power mixture) and 16.4:1 (roughly best economy), a twin cylinder motorcycle would be unrideable. But that gets to my real point which is why adding 4-8% more fuel that a narrowband sensor dictates makes such a significant difference.

In simple terms, you can get the power between cylinders to be close to equal by air balancing alone, IF there is enough fuel in the mixture to consume all the available oxygen. Once there is enough fuel to consume all the O2, then a bit more fuel in one cylinder doesn't effect the power balance L/R. The other thing to do would be to get perfectly equal fuel side to side and then add enough air so that all the fuel was consumed.

There are no adjustments to make fuel equal side to side so we have to balance air side to side and then, for a really smooth running motorcycle, add enough fuel so that all the O2 is consumed. The stock mixture is 14.7:1 and it turns out you only need to add 4-8% more fuel (given typical injector mismatch) to consume all the unburned air. So although 4% doesn't sound like much, it very often does the job. And the best part, of the 4% fuel that gets added (or 8% if you like) half of it is converted to power so not much is wasted.

The last consideration is the O2 sensor itself. The prior post mentions that there is only a TPS on one of the two cylinders. It should be noted that there is only one O2 sensor for both cylinders. Therefore if you perfectly balance the air but the L/R fuel is imbalanced by 6%, one cylinder is running at 14.4:1 and the other is running at 15.0:1 this too is a large imbalance. The greater the injector mismatch, the more additional fuel you need to add, hence my estimate of a 4-8% range.

So that's the rest of the story. Graphs and details in this admittedly long thread: http://forums.bmwmoa.org/showthread.php?56990-2004-R1150RT-Wideband-O2-Sensors
 
Interested to hear how you made sure the throttle stops and big brass screws were set to exact equal settings.
I do not have a flow bench setup. I cleaned the throttle bodies & BBS ports with a Q-tip & carb Cleaner. The BBSs were set out 1 1/2 turns from closed. The throttle stops were set using a flashlight from the back side to get just the first indication of opening. Not very scientific but I was especially careful that they both looked the same.

Someone else has advocated a 1/16 turn on the TPS screws from initial contact. Probably about what I had. I then set the TPS to .385 V when the LH throttle was against the new stop setting.

After re-assembling everything, this was all amazingly close to the needed final setting. Maybe I was lucky. I used an oil based U-tube to do a final tweak of the BBSs and to adjust the throttle cables etc.

A lot happens in the first few degrees of throttle opening. The processor has to convert that voltage (plus other info) into an airflow calculation that can be used to update the injector pulse length map. How much depends on the settings of the BBSs vs the throttle stops. Obviously the algorithm must be quite adaptable to a wide range of air flows as the scaling of airflow mass vs millivolts of TPS has to be grossly non-linear as well as wildly different depending on the balance of BBSs setting vs TPS stop setting. Hopefully the slope of this relationship has to be within say a 2:1 ratio for the two sides to prevent a non-combustible mixture error & stumble.

I would like to create a flow bench if someone had a spare throttle body for tests. From it we could establish the relationship between turns of BBS vs turns of throttle backstop to deliver equivalent idle air flow rates.

I did not go after the injectors initially as I discounted contamination since the surge problem for my bike seemed to be come on slowly.
 
The notion in the prior post that combustion can take place between 7:1 and 19:1 is true, but wrong-minded. If the difference between cylinders was merely (let alone 7:1 or 19:1) 13.2:1 (roughly best power mixture) and 16.4:1 (roughly best economy), a twin cylinder motorcycle would be unrideable. But that gets to my real point which is why adding 4-8% more fuel that a narrowband sensor dictates makes such a significant difference.

I don't think power output vs mixture is that critical - especially in a minimum power situation. More critical is what the map optimization does to both cylinders when only one cylinder is dumping excess oxygen into the converter.

I gotta study & think about the rest of your post
NRP
 
I don't think power output vs mixture is that critical - especially in a minimum power situation. More critical is what the map optimization does to both cylinders when only one cylinder is dumping excess oxygen into the converter.

I gotta study & think about the rest of your post
NRP

At every power level, fuel and air play equal roles. They way to take the fuel imbalance out of the equation is either with very good matching of injectors, individual O2 sensors like the R1200 or on our R1100 and R1150 motocycles, add just enough fuel so that all the oxygen is consumed.

The guys at Gami who make precision matched injectors for piston-engine aircraft disagree. They believe, have measured, and produce matched injectors. Their claim is that most "driveability" issues with fuel injected engines have to do with the torque pulses created by power/AFR imbalances between the cylinders. In fact, their view is that lean-misfire is not the cause of the rough-running caused at low fuel flows (lean mixtures).

To prove their point, they have produced custom fueling arrangements that take into account intake fuel scavenging from adjacent cylinders in common manifold engines. Once they get the air and fuel balanced, they can run much leaner mixtures at the same power level.
 
Just as further food for thought, first this from an article from GAMI's web site:

Why do variations in fuel/air ratios cause engines to run rough on the lean side of peak but not on the rich side of peak?

The answer is really pretty simple. But it has nothing to do with what you probably have been taught. Most pilots are taught that engines run rough on the lean side of peak due to some mysterious "lean misfire". Nonsense.

The engines run rough because on the lean side of peak, most engines do not have uniform cylinder-to-cylinder horsepower output. This cylinder-to-cylinder horsepower imbalance causes vibration, which has been often and wrongly characterized as lean misfire. We know [that] if one does precisely balance the fuel/air ratios of all of the cylinders, one can lean the engine to more than 100 degrees lean of peak - even 120 to 140F lean of peak ...
(100 degrees lean of peak is much leaner than our Oilheads, which run at about peak EGT, operate)

Second, just this past week wjg04oh (who is now running and riding with an LC-1) realized that somewhere around 4000 RPM in 1st, 2nd and 3rd gear, his motorcycle was open loop and much leaner than expected. How lean? Roughly 4% leaner than his Lambda setpoint. I reran his test on my motorcycle and found the same conditions and noticed that AFR changed quickly with small changes of throttle input (read the last 4 or 5 posts here). If our bikes had been running the stock setpoint (14.7:1) our measured AFRs would have been 15.3:1 (average of the two cylinders)--one will be even leaner. Referring to the Gami comments above, these conditions would certainly be in the "rough running" (or surging) territory.

I think the more of us that run with LC-1s connected, the more we will understand the combustion conditions that lead to "rough running".
 
Back
Top