BMW F800GS engine tuning

Discussion in 'Parallel Universe' started by DakarBlues, Dec 17, 2009.

  1. mousitsas

    mousitsas Long timer

    Joined:
    Oct 10, 2008
    Oddometer:
    3,349
    Location:
    Athens, Greece
    If you have an AFR 14.7:1 up on top, then you are heading for a very short engine life. Better try 13:1 .
    #21
  2. zumohamr

    zumohamr Adventurer

    Joined:
    Mar 15, 2010
    Oddometer:
    24
    Location:
    The South
    A pipe and air filter would give these 800s some much needed sound. My old TL1000s was unrideable until i used a fuel controller from yoshimura. No piggyback junk like a fuel commander. Plug it in and adjust fuel settings for each individual cylinder. (90V, a real engine) Take for a ride and repeat until bike is correct. it takes awhile but worth it. Someone could surely make something similar for these BMWs.
    #22
  3. LukasM

    LukasM Long timer

    Joined:
    Nov 15, 2006
    Oddometer:
    5,461
    Location:
    On a RTW ride - currently touring the U.S.
    +1.

    People that think that 14.7 is optimal for anything other than emissions are clearly clueless about engine tuning.
    #23
  4. The Griz

    The Griz Long timer

    Joined:
    Nov 26, 2008
    Oddometer:
    3,851
    Location:
    Minnesota
    Well, soooooorrry, there Lukas! I guess I'm just fucking clueless then!! :photogI meant that I gravitate more towards the stoi ratio of 14.7:1. I actually found myself achieving the best results on my KLR's with about 14:1 using a lambda sensor and meter kit (http://www.innovatemotorsports.com/products/lc1.php and http://www.innovatemotorsports.com/products/g2_gauge.php). I had the exhaust header drilled and a bung welded in for the sensor. It's common knowledge that, of course, stoi 14.7:1 is hardly achievable on any engine that constantly changes RPM and load, like a motorcycle or car engine. Now, that said, on a generator motor or something of similar nature that maintains a constant RPM and load? Hell yeah, you can achieve 14.7:1 safely and effectively.

    And btw, 14.7:1 is completely safe. In fact, contradicting such a fact would be disagreeing with decades of hard scientific evidence regarding the internal combustion engine! I'm with Joel on this one, guys. Anyone feeling the need go any richer than 14:1 is looking for spark plug fouling, damage to catalytic convertors, a reduction in fuel economy, etc. But, yeah, you will increase HP slightly.
    #24
  5. LukasM

    LukasM Long timer

    Joined:
    Nov 15, 2006
    Oddometer:
    5,461
    Location:
    On a RTW ride - currently touring the U.S.
    No offense Griz, but yes, it would appear that you are clueless. :D I'm far from an expert myself but what you are saying just doesn't make any sense.

    I am not sure why "it is common knowledge that 14.7 is hardly achievable", and also why you would even try to tune for that - unless for emission reason as I mentioned above?

    A properly tuned engine will target around 12.6 under full load for max power, and then lean out to about 15.4 under light load to cruise economically. With correctly set up EFI this is entirely achievable.

    What did you use to tune your KLR then?
    #25
  6. The Griz

    The Griz Long timer

    Joined:
    Nov 26, 2008
    Oddometer:
    3,851
    Location:
    Minnesota
    It is hardly achievable for any sustained period of time, due to fuel quality, load, and RPM change.

    Click on the links in my above post ^

    And read this while you're at it:

    http://en.wikipedia.org/wiki/Air-fuel_ratio
    #26
  7. itsatdm

    itsatdm Long timer

    Joined:
    Jun 19, 2004
    Oddometer:
    5,371
    Location:
    Nor Ca.
    Sounds like this is the reason for the the slight stumble, when slowly trying to ease over things just off idle. If it is, than doesn't sound like there is much that can be done about it.
    #27
  8. LukasM

    LukasM Long timer

    Joined:
    Nov 15, 2006
    Oddometer:
    5,461
    Location:
    On a RTW ride - currently touring the U.S.
    Way too much editing of your posts going on Griz... Next time just make a new one instead of changing what you said (optimal results with 13.5 or was it 14.0?) and adding stuff 5 times.

    So you tuned your KLR with the G2 gauge then? Did you do it on a dyno? Any graphs?

    If you are not very well versed in a topic, maybe you shouldn't always make these absolute statements like "And btw, 14.7:1 is completely safe. In fact, contradicting such a fact would be disagreeing with decades of hard scientific evidence regarding the internal combustion engine!". Keep an open mind man.

    Whether 14.7 is completely safe depends on a multitude of factors, for example the compression ratio, fuel octane, ignition timing, induction type, etc. I would like to see you run a high compression drag racing engine or even a low compression one with 20 lbs of boost at your "completely safe" 14.7:1 on standard gas. You'd be picking up your pistons in pieces.

    In case of a stock F800GS, this is unlikely to be an issue. But then again, if you were going to reprogram your EFI to compensate for the free flowing exhaust, air filter, etc why not make it optimal instead of just safe? You know, like, rich at full load and lean for cruising? More power and still get better mileage?
    #28
  9. upweekis

    upweekis Long timer

    Joined:
    Jul 22, 2008
    Oddometer:
    3,101
    Location:
    Indiana-We Tax Everything!
    I read somewhere that a guy is making a new throttle roller(gear, whatever the part is on the handlebar grip) that is cammed, so that at less than 1/3 throttle, it moves slower, and then picks up to normal after that. I think that would cure a lot of off-road problems. Of course, I can't find it now...
    #29
  10. johngil

    johngil Reseda, CA

    Joined:
    Jun 28, 2007
    Oddometer:
    3,815
    [​IMG]
    #30
  11. johngil

    johngil Reseda, CA

    Joined:
    Jun 28, 2007
    Oddometer:
    3,815
    [​IMG]
    #31
  12. LukasM

    LukasM Long timer

    Joined:
    Nov 15, 2006
    Oddometer:
    5,461
    Location:
    On a RTW ride - currently touring the U.S.

    Do you mean G2 Ergonomics? They make some nice stuff, I am using their changeable cam system on my KTM.

    http://g2ergo.com/index.html
    #32
  13. Y E T I

    Y E T I Unpossible

    Joined:
    Apr 26, 2007
    Oddometer:
    8,335
    Location:
    San Diego
    Like Lukas said, it's a G2 Throttle Tamer. I installed one on my bike and wrote it up HERE. Sam at G2 is a great guy too. It doesn't toally fix the problem, it's still a bit jerky right off idle, but it is much more controllable once you've got the throttle open.
    #33
  14. Dolly Sod

    Dolly Sod I want to do right, but not right now

    Joined:
    Aug 18, 2003
    Oddometer:
    15,275
    Location:
    Richmond, Va
    :clap

    It's not the first time.
    #34
  15. upweekis

    upweekis Long timer

    Joined:
    Jul 22, 2008
    Oddometer:
    3,101
    Location:
    Indiana-We Tax Everything!
    Yep! that's it.
    #35
  16. JoelWisman

    JoelWisman Long timer

    Joined:
    Apr 22, 2009
    Oddometer:
    1,392
    Location:
    STL, MO, USA
    (quoting what I said about no fuel adaption at idle)

    actually what I said was much to simplified.

    There are three types of adaption, "additive", "multiplicative" and "extrapolated".

    Addative adaption occurs on the fly as soon as the engine is warm. The F800GS uses a standard narrow band slow oxygen sensor and it is too far back in the exhaust to be useful without causing idle to hunt and peck all over the place. Addative adaption occurs once the sensor is warm at just above idle all the way to just short of wide open throttle (wot).

    Multiplicicative adaption occurs over time and is remembered and incorporated as a conversion factor of the base fuel map. Multiplicitave is remembered from start up cycle to start up cycle and changes slowly over miles and large fractions of an hour of running.

    The F800GS does use multiplicative adaption for idle fuel mix as well as part throttle fuel mix.

    Lastly there is extrapolated adaption. The F800GS's O2 sensor is not capable of measuring fuel mixtures that are significantly different then lambda, and yet for a variaty of reasons INCLUDING emissions you want a pretty rich mixture at WOT. As such, the F800GS engine management computer enriches the mixture at near full throttle through WOT by a certain percent. This enrichment is based on the modified multiplicitive fuel map and is therefore called extropolated adaption.

    The idle fuel mix is adapted but not well, it takes it some time to be optimal when the energy or oxidizer content of fuel changes. When I have gone from oxygenated to unoxygenated fuel it took about 20 minutes for the o2 sensor to start crossing.

    The real reason for jerky transitions from idle is a lack of acurate sensors such as an air mass sensor or at least MAP sensor and the fact that BMW has so far failed or does not wish to damp this transition with industry techniques.

    My guess is that the snapy throttle sells bikes after a test ride to typical F800GS customer, even though it is sub optimal off road in technical stuff.

    As for stumbles going from part throttle to idle. This shouldn't happen if everything has had time to adapt to fuel and conditions but does on some bikes and I have no idea why.
    #36
  17. eddyturn

    eddyturn Wannabe

    Joined:
    Oct 27, 2009
    Oddometer:
    2,655
    Location:
    Hesperus, CO
    God Joel... I love that kind of talk. :norton
    #37
  18. The Griz

    The Griz Long timer

    Joined:
    Nov 26, 2008
    Oddometer:
    3,851
    Location:
    Minnesota
    No need for a dyno. I read the meter while riding under actual conditions. I adjusted fuel air mixture with needle size and jet sizes until I was sitting at 14:1 on the meter. 14:1 felt and worked the best for economy and power. IMHO, a fuel/air mixture gauge is far more accurate than a dyno. Dyno's and charts aren't realistic. I want to set up the fuel/air mixture while riding under loads and conditions that I will likely encounter during use. Not in a lab on a treadmill.

    Exactly what I said earlier. Repeat.

    Agreed.:thumb Optimal and safe sometimes fall at the same ratio though.
    #38
  19. martinbywater

    martinbywater n00b

    Joined:
    Mar 26, 2010
    Oddometer:
    3
    Hi, a few have mentioned the vibration at 80mph, I have used a larger front sprocket ie 17th, available from touratech and others, makes it far more comfortable for fast touring, did also try the smaller rear sprocket ie 41th, this made 5th gear identical to 6th prev, but I felt too highly geared to me, but gives in effect a 6th gear overdrive.

    I would highly reccomend the 17th front for touring, goes on easily with the std chain lenght.
    Martin
    #39
  20. LukasM

    LukasM Long timer

    Joined:
    Nov 15, 2006
    Oddometer:
    5,461
    Location:
    On a RTW ride - currently touring the U.S.
    Your offer for truce is nice, but that's not how discussions work. I am not attacking you personally (well, maybe poking fun a little bit :D) but mostly addressing the incorrect information that you are posting. If anything I am saying is wrong I would be glad to have it pointed out to me so I can learn something, too. But if you keep posting bullshit, I'll keep calling you on it.

    As for the editing function on ADV - this can be both a blessing and a curse, which is why you will see many forums that don't have it or at least put a time limit on it. It can be very useful if you realize too late that you made an honest mistake, like posting a wrong part number or torque value, and don't want to mislead people. Or maybe just to correct a spelling mistake. But even if "the tools are provided" you should have the integrity to stand by what you said and not go back 5 times and change things as the discussion goes on. Simply make a new post for new information, or if you really must, make an editing comment on the bottom of your old post to clarify what you meant. This is the only way people will be able to keep track of who said what. Then again, can Booni and I really expect you to actually THINK BEFORE YOU POST? According to you, apparently not. How long would you like us to wait before we read your posts and reply to them?

    For being an engineer (I read that in another post of yours) you sure talk funny. Did you ever hear the saying that data should be observable, measurable, and repeatable? You call a dyno and charts "not realistic", yet your butt dyno seems to be so finely tuned that it is able to produce this type of data. Amazing. Or, it simply confirms once more that you are clueless about engine tuning. Did it ever occur to you that every dyno has an air/fuel meter that is used as a basis for any adjustments made? Maximum power doesn't just happen by coincident, it is simply a matter of physics. Saying that "14:1 felt and worked the best for economy and power" is just very narrow minded. You should be adjusting your mixture dependent on load and the other parameters I mentioned before: timing, compression, N/A or forced induction, octane etc), which in turn will give you the results you are looking for.

    And while your idea of "setting up the air/fuel mixture while riding under loads and conditions that you will likely encounter during use. Not in a lab on a treadmill." is cute, it somehow makes me doubt that you ever used the G2 gauge. I specifically asked you - twice - how you measured AFR, and you said that you "read the meter while riding under actual conditions". Nowhere did you mention the word data logging. Surely you must realize that using a readout gauge - especially a digital one that will cycle between values very fast - is pretty much useless while you are riding the bike. How the hell would you completely focus on it to get any kind of useful info across the whole RPM range, while "riding under actual conditions"? For what it's worth, even on a dyno the use of a gauge is limited. While the lab conditions - you know, like, not having to look where you bike is going at increasingly fast speeds - will allow you to monitor AFR at various throttle positions at least to a certain degree, you will still miss spikes in either direction. This is why you always need data logging, and why these "not realistic" graphs are essential....
    #40