1200/1250 GS Watercooled Engine Tuning

[Deleted account 191119001]

In this thread please do not mention the sponsors name on top of the page, otherwise this thread will be removed to the sponsors own section.


So as I am looking into possibly tuning my bike in a while when my warranty runs out, can some of the guys that have actually done things to their Watercooled GS1200/1250 chime in and say technically what they think has worked for them, and why, I have seen AX-FIED that looks good, I have seen that TPS RESET seems to help, is their anything real out there that I have missed, something new maybe.....

 

[malcopopolo]

Very good.

Do we can understand more...

How does your bike run when going up a hill?
What's it like in top gear?

Sent from my Lenovo TAB 2 A10-70F using Tapatalk

 

[Deleted account 191119001]

Very good.

Do we can understand more...

How does your bike run when going up a hill?
What's it like in top gear?

Sent from my Lenovo TAB 2 A10-70F using Tapatalk

Going up a hill it pulls ok, but in top gear it just feels that it runs out of puff, and could give a little more, but I need to explore if anything really can be done, or just accept it as this is the way the bike is, and I aint buying the new 1250 for a bit more torque and power, there must be a way.

 

[fredaroony]

TPS reset won't really do anything unless the calibration has wondered. You usually get symptoms of poor running and actual issues, it's not a performance thing.

I'd find a place that will tune the bike per cylinder before the cat. This type of tuning usually takes a while to get right and takes numerous dyno runs

 

[Deleted account 191119001]

TPS reset won't really do anything unless the calibration has wondered. You usually get symptoms of poor running and actual issues, it's not a performance thing.

I'd find a place that will tune the bike per cylinder before the cat. This type of tuning usually takes a while to get right and takes numerous dyno runs

Thanks for that food for thought, I do have a friend who really is a shit hot bike builder and tunes race bikes, when I spoke to him, he said nothing much can be done with the Watercooled engine the ECU is clever enough, and learns all the time, the more you ride the more it learns, he says the only guys that actually know enough of the insides of them ECU,s are BMW themselves, and that is part of their trade secrets, hundreds of millions invested in R&D

 

[spoonz]

Having a pc3 on my other bike which works well, that is my obvious choice with a map per cylinder as Fred mentioned.
But my potential issue with that beyond warranty ( ends Oct ) is that with exposed injector wiring etc I can imagine it looking a bit of a dogs dinner and maybe being a bit vulnerable.

 

[Deleted account 191119001]

Having a pc3 on my other bike which works well, that is my obvious choice with a map per cylinder as Fred mentioned.
But my potential issue with that beyond warranty ( ends Oct ) is that with exposed injector wiring etc I can imagine it looking a bit of a dogs dinner and maybe being a bit vulnerable.

Yep seems like I am looking for something in my dreams that does not exist.

 

[G&T]

I’m non technical, and whilst I understand that an ECU can learn, my riding style can change one day to the next depending on my mood and motivation. So, what is it learning - that I’m a broad church with some hang ups? And therefore what is the bike giving me back?

 

[Deleted account 191119001]

I’m non technical, and whilst I understand that an ECU can learn, my riding style can change one day to the next depending on my mood and motivation. So, what is it learning - that I’m a broad church with some hang ups? And therefore what is the bike giving me back?

If my understanding is correct, the bike is learning what fuel or air to give in a certain situation, and rightfully as you say our riding methods change, so the ECU has to adapt.

 

[tanneman]

If my understanding is correct, the bike is learning what fuel or air to give in a certain situation, and rightfully as you say our riding methods change, so the ECU has to adapt.

No, there are a set of parameters and the ECU only calculates the best parameter for the given situation on which it relies from data inputs from sensors. Think of it as a 3D map where the lowest and highest extremes represent the design criteria and everything inbetween is a variable that satisfies a calculation done at any given time using that data collected from the sensors.

To anticipate what the inputs might be, that is learning, and you will find that tech in the control systems of autonomous cars that use GPS data overlayed on a map to predict what action to take at a particular place in the journey. The bikes we ride cannot 'see' what is happening around it and the inputs to the ECU is reactionary rather than predictive.

 

[Roger 04 RT]

I’m non technical, and whilst I understand that an ECU can learn, my riding style can change one day to the next depending on my mood and motivation. So, what is it learning - that I’m a broad church with some hang ups? And therefore what is the bike giving me back?

Fueling adaptation is independent of riding style, except that after an ECU reset if you ride at steady speeds in a variety of gears and engine loads, the Adaptation can happen faster. The basic process is that any time the ECU can enter Closed Loop, it immediately starts building short term trims called Lambda Control Factors. Then over time it moves the LCFs to long term trims called Multiplicative and Additive trims that are applied to the entire fueling map. Again the values of these trims are independent of riding style, they just insure that fueling is as BMW designed it.

The AF-XIED takes the approach that since the O2 sensor is the reference for the ECU, if you change the O2 sensor switching value toward a richer value the ECU will use its algorithms to add fuel across the map, which it does. That might sound like brute force but a little more fuel makes a nice improvement to most O2 sensor based engines.

On my 2017 R1200 I get: stronger take off with less throttle, smoother auto shift performance, better roll on acceleration, I usually ride one higher gear, 5th gear is now fine even at 40 mph and I’m always in 6th gear by 50. The exhaust is also slightly cooler. I run on setting 8 which is 6-7% more fuel but because of running one higher gear locally don’t see an MPG difference locally but may lose 1 MPG on the highway.

 

[Deleted account 191119001]

No, there are a set of parameters and the ECU only calculates the best parameter for the given situation on which it relies from data inputs from sensors. Think of it as a 3D map where the lowest and highest extremes represent the design criteria and everything inbetween is a variable that satisfies a calculation done at any given time using that data collected from the sensors.

To anticipate what the inputs might be, that is learning, and you will find that tech in the control systems of autonomous cars that use GPS data overlayed on a map to predict what action to take at a particular place in the journey. The bikes we ride cannot 'see' what is happening around it and the inputs to the ECU is reactionary rather than predictive.

Thanks that is nice and clear, makes easier understanding for me......

 

[malcopopolo]

I hear there is a company that in some circumstances can give you better fuelling with 10 to 20bhp power increases and even supplied dyno information.

Been thinking about this.... (dangerous)

We have been advised that you can clone an ECU. So it would be possible to extract this data as at the end of the day it is computer code and in its most basic form it is ones and zeros. If the file size was say 1 megabyte. and you altered the ecu code slightly to say add more fuel in certain scenarios.... the file would be virtually the same and unless compared side by side the the original it would appear invisible

I think all that happens is tweaking of the original code. Tuning companies will not confirm their methods as it would be quite easy to clone a modified ECU in my simple opinion

 

[Roger 04 RT]

I hear there is a company that in some circumstances can give you better fuelling with 10 to 20bhp power increases and even supplied dyno information. ...

I hate to rain on everyone’s parade but here are some reasons why I don’t ever expect ECU fuel tuning to deliver 10-20 additional horsepower to a 100 HP BMW engine:

1) Since HP is torque times RPM, peak HP is by definition in the upper band of RPM at WOT.

2) Every BMW motorcycle I’ve measured with a Wideband O2, ahead of the cat, shows WOT, high RPM fueling in the range 12.7 to 13.0:1. That is already at or near what’s called BEST POWER MIXTURE—meaning in that vicinity you don’t get more power from just more fuel, it just reduces the exhaust gas temperature and makes a dirty burn.

3) Engine designers often don’t advance the ignition timing to peak HP at WOT. They leave it less advanced than optimum HP to create a safety margin against detonation and preignition. A tuner can eat into the safety margin, but the HP increase isn’t 10-20%.

4) To add HP at WOT to an engine you have to add more fuel AND more air and then adjust timing to optimum, with a safety margin for temperature and barometric pressure and engine aging.

Below WOT for 70% of the map, a tuner can change fueling values in the map, but Closed Loop Adaptation returns the fueling to BMW’s design targets. There are some improvements that can be made with spark advance but it’s a lot of work and takes a lot of time. For that 70% of the map, if you want more fuel with connected O2 sensors you have to change the O2 sensors switching point one way or another.

 

[spoonz]

From a gs LC point of view extracting the data from the ecu is not the hard bit but writing back any alterations is, as there is protection. So without exception all tuning options appear to involve around changing the values fed to the oem ecu code, not altering the code itself.

It’s also possible given the right gear to use a hex comparator which will highlight any changes whatsoever to code occupying a chipset should anyone want to. As in the whole chipset not just the bit used by an oem.

 

[Deleted account 191119001]

Fueling adaptation is independent of riding style, except that after an ECU reset if you ride at steady speeds in a variety of gears and engine loads, the Adaptation can happen faster. The basic process is that any time the ECU can enter Closed Loop, it immediately starts building short term trims called Lambda Control Factors. Then over time it moves the LCFs to long term trims called Multiplicative and Additive trims that are applied to the entire fueling map. Again the values of these trims are independent of riding style, they just insure that fueling is as BMW designed it.

The AF-XIED takes the approach that since the O2 sensor is the reference for the ECU, if you change the O2 sensor switching value toward a richer value the ECU will use its algorithms to add fuel across the map, which it does. That might sound like brute force but a little more fuel makes a nice improvement to most O2 sensor based engines.

On my 2017 R1200 I get: stronger take off with less throttle, smoother auto shift performance, better roll on acceleration, I usually ride one higher gear, 5th gear is now fine even at 40 mph and I’m always in 6th gear by 50. The exhaust is also slightly cooler. I run on setting 8 which is 6-7% more fuel but because of running one higher gear locally don’t see an MPG difference locally but may lose 1 MPG on the highway.

Thank you so much now that is really helping.....

 

[Deleted account 191119001]

I hate to rain on everyone’s parade but here are some reasons why I don’t ever expect ECU fuel tuning to deliver 10-20 additional horsepower to a 100 HP BMW engine:

1) Since HP is torque times RPM, peak HP is by definition in the upper band of RPM at WOT.

2) Every BMW motorcycle I’ve measured with a Wideband O2, ahead of the cat, shows WOT, high RPM fueling in the range 12.7 to 13.0:1. That is already at or near what’s called BEST POWER MIXTURE—meaning in that vicinity you don’t get more power from just more fuel, it just reduces the exhaust gas temperature and makes a dirty burn.

3) Engine designers often don’t advance the ignition timing to peak HP at WOT. They leave it less advanced than optimum HP to create a safety margin against detonation and preignition. A tuner can eat into the safety margin, but the HP increase isn’t 10-20%.

4) To add HP at WOT to an engine you have to add more fuel AND more air and then adjust timing to optimum, with a safety margin for temperature and barometric pressure and engine aging.

Below WOT for 70% of the map, a tuner can change fueling values in the map, but Closed Loop Adaptation returns the fueling to BMW’s design targets. There are some improvements that can be made with spark advance but it’s a lot of work and takes a lot of time. For that 70% of the map, if you want more fuel with connected O2 sensors you have to change the O2 sensors switching point one way or another.

Brilliant so if I was to be picky in point number 2 Roger seems it would be really dangerous to mess with these values, as un-burnt fuel could set the bike on fire maybe similar to what happened to a member of the forum last week, or at the least maybe overheat the bike somewhere....

 

[tanneman]

A rich mixture runs cooler. That is how top fuel dragsters cool the engine (no water jacket for those engines), we are talking liters per second here for the dragsters. A lean mixture runs hot because the burn is more efficient in an abundance of O2 compared to fuel particles. Exhaust gas is about 700 deg C and up so you can imagine the heat buildup of restricting the exhaust with a cat. Maybe worded wrong but for the cat to be efficient it needs to be hot and placing it closer to the exhaust valve increases the efficiency.

Un-burnt fuel would ignite in the exhaust where there is sufficient O2 and heat available. This would cause that crackle and pop you like to hear in a tunnel. However it upsets the pressure wave the engine depends on to scavenge the exhaust gasses and can damage the cat in modern exhaust systems.

 

[Deleted account 191119001]

A rich mixture runs cooler. That is how top fuel dragsters cool the engine (no water jacket for those engines), we are talking liters per second here for the dragsters. A lean mixture runs hot because the burn is more efficient in an abundance of O2 compared to fuel particles. Exhaust gas is about 700 deg C and up so you can imagine the heat buildup of restricting the exhaust with a cat. Maybe worded wrong but for the cat to be efficient it needs to be hot and placing it closer to the exhaust valve increases the efficiency.

Un-burnt fuel would ignite in the exhaust where there is sufficient O2 and heat available. This would cause that crackle and pop you like to hear in a tunnel. However it upsets the pressure wave the engine depends on to scavenge the exhaust gasses and can damage the cat in modern exhaust systems.

Hence why my other older harley with the cat removed, felt way cooler, I am understanding so much more now, thank you guys, this is a nice thread.

 

[Roger 04 RT]

Cats accumulate oxygen during the lean part of the cycle at a mixture of 14.7:1 during closed loop operation. During the rich part of the cycle they use the stored O2 to oxidize any unburned hydrocarbons. When the mixture is rich for a sustained period the stored oxygen is depleted and the cat stops oxidizing and cools down.

When a engine misfires and unburned fuel and unburned oxygen enter the cat it burns and gets extremely hot, damaging the cat.