rusEFI on BMW boxer motorcycle engine

[Flegmatoid]

So, I have 2010 R1200GS motorcycle engine stripped on native electronics (BMS-K, ZFE modules are missing).
I know some 3-d party ECUs support it. All closed-source and pricey - 400 to 5000 USD. Dynojet, Bazzaz,EJK and insanely pricely trijekt

R1200GS Engine Specs:
- 2 cylinder boxer
- DOHC
- per-cylinder air intakes with dedicated valves
- drive-by-wire (no mechanical valve control)

RusEFI seems to be a way to go, however, I am not sure if Frankenso board can support:
- O2 sensor per cylinder (2x O2 sensors in total)
- TPS (with Throttle-Valve servo, BMW calls it E-GAS)
- IAC (also stepper/servo controlled)


Actuators, one for each cylinder
1. Lower Spark Plug Coil
2. Upper Spark Plug Coil
3. Fuel Injector
*4. Stepper type idle air control valve

Sensors:
1. 02 Sensor
2. Oil Pressure Sensor
*3. Cam angle sensor: The BMS-K uses this to determine firing/injection sequence for the left and for the right cylinder, independently.
*4. Knock sensor: Installed into the cylinder (not in the head).
*5. Temperature sensor in the head.
*6. TPS: Three wires
*7. Revolution sensor. It is not to read DTC, (like a hall sensor)
8. Oil temperature sensor
9. Intake air temperature sensor

more pictures https://andyw-inuk.smugmug.com/Motorcycles/R1200GS-BMW-Motorcycle-Photo/R1200GS-Schematics-Diagrams/i-MdWFsbR
and a nice video about its internal structure

Any challenges or risks you see about this conversion are welcome

[AndreyB]

hello and welcome!

- TPS (with Throttle-Valve servo, BMW calls it E-GAS)
- IAC (also stepper/servo controlled)

These are the hard parts. IAC is supported with an external board - see http://rusefi.com/wiki/index.php?title=Hardware:Stepper_motor
e-throttle is pretty much not supported. There were some experiments but noone has used on a real vehicle - see http://rusefi.com/forum/viewtopic.php?f=5&t=592

[kb1gtt]

Do you have a schematic for your throttle body? I'm wondering what amount of current the motor uses, as well it's inductance, and it's general wiring arrangement. To get this to work physically, you may need an additional board added to proto area on the Franenso, or you might be able to use the low side drivers that are on the board. With out additional details I do not know what would be required physically.

I understand that to get this working from a software stand point, that @russian would need to debug and continue to develop some software. I believe we have had one valve working on a bench top, such that it could be opened and closed on command. I recall it was straight up TPS = commanded motor position. I understand that while this would potentially allow an engine to operate, it's probably not the safest approach. I believe most OEM's will predict fuel needs based on RPM, and TPS as a load command. Then the motor moves such you would get the proper AFR for the delivered fuel. Such that if your motor driver(s) break and goes wide open, your fuel doesn't run away and go wide open on you. Currently the fuel algorithm looks at MAF or MAP and calculate fuel. This means that if the driver breaks and your motor goes wide open, you would get run away engine situation.

[Flegmatoid]

Hi again,

here goes the electrical wiring for R1200GS as well as TPS photos.
TPS is a dual 3-wire potentiometer. It fails on each and every BMW motorcycle regularly. So its replacement with a 3rd party-one is a viable option.

Unfortunately, I cannot give the Apms reading for it, but its not critical. Worst case is it will lag a bit, if under-powered.

Also, I'd like to append my previous post. My intention is not to drive the entire motorcycle, but only the engine.
Not sure if rusefi can perform ignition/starter/pump features etc(?).

Those BMW engines are gaining popularity among airplanes, airtrike, autogyro, hovercraft and even small helli enthusiasts due to incredible torque/weight ratio (alternatives are 10x more expensive).
I, myself, do also plan fitting it on an airplane (to perform air-fitting as we call it). Trijekt ECU comes at ~2000eur (more than an engine).
So having sub 500 eur EFI to drive this engine would geometrically increase rusefi popularity.

I haven't looked into the source-code yet, but it will definitely require an "AIR" mode, where engine shutdown due to temperature breach or sensor malfunction cannot be tolerated (in respect to ground vehicles).

[AndreyB]

I, myself, do also plan fitting it on an airplane ... I haven't looked into the source-code yet, but it will definitely require an "AIR" mode, where engine shutdown due to temperature breach or sensor malfunction cannot be tolerated (in respect to ground vehicles).

It definitely sounds like your next step is finding a mechanical throttle body for this engine.

Also see http://rusefi.com/wiki/index.php?title=License:general

* rusEfi is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without
* even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.

For IAT and CLT sensors for example there is some validation which would use a default value in case of an unrealistic value like -200C or +200C, for other sensors it's an open question what kind of validation could be done. I guess once your engine is running on the ground you would have a chance to help us help you.

[kb1gtt]

I know some experimental air craft with a 10+ year history which uses some lawn mowers for engines and blue tarps for wing coverings. I guess at the end of the day people put on those wing suits and jump out of perfectly good planes. I would of course recommend using more proven technology for air stuff. Especially when under 30ft to 40ft. If your engine stalls briefly below that kind of elevation you have no time to recover, and those crashes are typically lethal.

That said, Frankenso with rusEFI has almost everything you need to run that particular engine. The key you are missing is the electronic throttle body. If you are unhappy with the existing TPS, perhaps it would be best to use a completely different throttle body. If you change the throttle body, it would put rusEFI in a far more standard setup, and you are far more likely to obtain success.

Yes currently rusEFI operates an engine only. There has been very little control for other items like transmissions, or other things that are not specifically an engine. While there is room for other features to exist, they require a significant development cycle, which generally requires a significant amount of coding resources from @russian. Those resources are in short supply. For now I understand @russian is trying to focus on the smaller details of the existing code, ensuring it's of good quality and with minimal problems. If you can change the throttle body, it would make it much easier to get support.

[AndreyB]

HNot sure if rusefi can perform ignition/starter/pump features etc(?).

http://rusefi.com/wiki/index.php?title=Manual:Current_Status has the list of features