Replace fuel pressure regulator on R12 -- worth it?

Discussion in 'GS Boxers' started by space, Oct 22, 2013.

  1. space

    space a.k.a. Jake

    Joined:
    Jan 1, 2006
    Oddometer:
    2,662
    Location:
    Los Angeles, CA
    Some R11xx riders have replaced the stock regulator on that bike, rated at 3.0 bar, with adjustable aftermarket ones. End result was that the bike ran smoother and had better throttle response with the pressure set to 3.3-3.5 bar. Inmate hasenwerk nicely describes his upgrade in this thread: http://www.advrider.com/forums/showpost.php?p=17259810&postcount=72

    The R1200GS and HP2 Enduro have stock regulators set to 3.5 bar. I have my HP2 apart right now and have already rebuilt much of the fuel system (including switching to an external fuel pump -- the original in-tank pump broke). I'm now toying with adding one of these:
    [​IMG]
    http://www.summitracing.com/parts/aei-13159/overview/

    Anyone go down this road before?
    #1
  2. ChrisGS1

    ChrisGS1 Been here awhile

    Joined:
    Oct 2, 2013
    Oddometer:
    238
    Just to make sure, you've got a R1200GS, and want to fit a variable fuel pressure regulator? If stock runs at 3.5bar, will jy increase it to say 3.8bar? I've heard of someone that did run it this high, but I think you need other modifications then too, like gas flowing, completely free flowing system and maybe modified air intake (to get enough oxygen in there?) not an expert
    #2
  3. space

    space a.k.a. Jake

    Joined:
    Jan 1, 2006
    Oddometer:
    2,662
    Location:
    Los Angeles, CA
    Yep, that's the idea.

    My replacement fuel pump has a greater flow rating than the stock -- and besides, I imagine the stock pump would also be sufficient. I'm not too worried about air intake, as these bikes are known to run lean. The lambda sensors should adjust the fuel flow (by lowering the duty cycle of the injectors) when in closed-loop operation while cruising, so fuel consumption should not increase too much.

    Or at least that's my thinking. I fully expect Poolside or one of the other local experts to tell my why I'm wrong any moment now .... :D

    One concern of mine: how fiddly are manifolds? Could the output of the regulator go into a tee fitting, which then leads via equal lengths of tubing to the injectors?
    #3
  4. JetSpeed

    JetSpeed Naviator

    Joined:
    Nov 28, 2007
    Oddometer:
    6,975
    Location:
    Coronado, CA - KNZY
    Somebody out there correct me if I'm wrong (and I'm sure they will) but I believe the 1200 has a different fuel injection system in that instead of utilizing a return flow system (like the R11XX bikes) in which a regulator bleeds excess pressure by allowing some of the unused fuel to return to the tank the 1200 uses an electronic fuel control device that adjusts the pumps output.
    #4
  5. ChrisGS1

    ChrisGS1 Been here awhile

    Joined:
    Oct 2, 2013
    Oddometer:
    238
    Having spoken to Roger on this topic (Roger 04 RT), I've gleaned some wisdom for the 1150s, not sure if it applies to R1200. The motronic does maintain preset (for closed loop operation) AFR as it gets feedback from O2 sensor. However, there is 'adaptation' that needs take place when you accelerate; open loop operation. Fore example; if your running lean, there's a time of adaptation before more fuel gets squirted in - delayed response. So, that's one effect of running lean...by the same argument, a delayed response might be the effect of running too rich - when you open throttle to accelerate, more fuel is squirted than is needed (with 3.8bar), too rich a mixture does not lead to more power, but less. Then the O2 sensor picks up that fuel must decrease - delay before the power.

    Roger has got a variable FPR fitted to his R1150RT, and he experimented. Found that 3.5bar gave best result. He tried 3.7bar, but that was too much. RE the air intake mods, It is known that these engines run lean, but that might be due to deficit in fuel, not because there's too much air - getting more fuel AND more air in combustion chamber would be ideal. Lennie's rocket sprockets can help with this, as by advancing CAM timing, the exhaust valve closes earler, trapping more gas that gets compressed, more torque.

    It's been said that the FPR on its own is not enough, you need to adjust closed loop AFR with something like AF-XiED (beemerboneyard). Used in conjunction with FPR, you will be running more rich overall, AND adaptation time will be minimized.
    #5
  6. space

    space a.k.a. Jake

    Joined:
    Jan 1, 2006
    Oddometer:
    2,662
    Location:
    Los Angeles, CA
    You're right that the 12 decreases the pump speed when full flow is not needed (which is to say, most of the time). This is done to increase the life of the pump. But it does have a return line to the tank as well.
    #6
  7. roger 04 rt

    roger 04 rt Long timer

    Joined:
    Nov 9, 2011
    Oddometer:
    3,029
    Location:
    Massachusetts
    Just increasing the fuel pressure does little, if anything. (Okay, the injection mist might be finer at higher pressure.) The reason is that your BMSK learns to squirt shorter injection pulses by comparing the exhaust's residual oxygen to the O2 sensors (you've got two of them, one for each cylinder).

    If you want to richen your mixture, change the O2 sensors to LC-1s or add a pair of BMW-AF-XIEDs to your stock O2 sensors. Terryckdbf has had great results on his R1200GSA by making this change.

    (above) Chris is close but just a bit off on his explanation. The FPR increase affects how long it takes the Motronic to learn a new richer fueling. However, once it has learned it over a few tanks of fuel, there is no delay of the response to throttle with or without a fuel pressure change.

    In my case 3.7 bar is not too rich. It is the setting I use.
    #7
  8. space

    space a.k.a. Jake

    Joined:
    Jan 1, 2006
    Oddometer:
    2,662
    Location:
    Los Angeles, CA
    So once the Motronic learns the new "optimal" pulse width map, it affects the open loop behavior as well? I'd been working under the assumption that open loop was a fixed map.
    #8
  9. roger 04 rt

    roger 04 rt Long timer

    Joined:
    Nov 9, 2011
    Oddometer:
    3,029
    Location:
    Massachusetts
    It's a common misconception that Open loop fueling ONLY uses a fixed fuel map.

    The short answer is that what your BMSK or Motronic learns in closed loop it applies to open loop. (Read up on short term and long term fuel trims.)

    These can be seen in the GS-911 log for F800s and R1200s. The log for 1150s doesn't show them but adaptation values are mentioned (but not explained) in the BMW service manual for Motronic MA 2.4.

    A short explanation, the ECU uses closed loop when it can when the bike is warmed up. When it can't use closed loop such as when the throttle is moving quickly, it starts with the map, corrects for temperature, pressure and battery voltage ... And also applies correction factors learned during closed loop.
    #9
  10. space

    space a.k.a. Jake

    Joined:
    Jan 1, 2006
    Oddometer:
    2,662
    Location:
    Los Angeles, CA
    Thanks, Roger. Very useful information. I'll pass on this mod.
    #10