HELP O2 sensor removal and Y pipe

Discussion in 'GS Boxers' started by CaptRick, Jun 29, 2007.

  1. bananaman

    bananaman transcontimental

    Joined:
    Jun 20, 2007
    Oddometer:
    9,689
    Location:
    Madison, Wisconsin and/or Panama, Panama
    If a year ago you guys had told me that in The Spring Of '10 I'd be obsessed with O2 sensors, Dutch fault-code tables, and phantom vacuum leaks manifesting as mythological limp-home modes programed by clairvoyant German engineers...
    #41
  2. Poolside

    Poolside Syndicated

    Joined:
    Apr 1, 2003
    Oddometer:
    11,887
    Location:
    Silicon Beach, CA
    <BR>I feel ya brother! Welcome to my world. :D It's often difficult in knowing exactly what happened and when. Or if not exactly, at least knowing what is possible and/or probable.

    <BR>
    #42
  3. vintagerider

    vintagerider Long timer

    Joined:
    Nov 20, 2005
    Oddometer:
    1,976
    Location:
    the west
    Bananaman, any U-Toob before and after videos on the perceived performance increase from replacing the lambda sensor?
    We want to believe but...

    A fault in the TPS, but not the lambda sensor can trigger limp home mode. I write from from personal experience with a damaged TPS on an r1100 w/mo 2.2. A sampling of incidents and ecu theory reported in various forums worldwide also suggests that it is unlikely that a bad lambda sensor would limit your engine rpm or top speed. The fact that you are now able to reach redline (are you?) after simply replacing the lambda sensor is coincidental. Perhaps you had a plugged tank vent that cleared itself or a mouse nest in the airbox that finally got sucked past the valves? What was the max rpm before the lambda change and how many rpms do you perceive to have gained by replacing the lambda sensor?

    For benefit of the less technical here, there are certain things we ought to suspect first when a single spark oilhead experiences loss of power. We know enough about the operation of motronic 2.2, as well as the non-ecu issues that we can now quickly resolve common r11xx performance problems without buying unnecessary parts or labor. Faulty lambda sensors do not cause sudden, dramatic loss of engine power. At best they cause declining fuel economy for those riders that spend a lot of time cruising at steady state.

    Non cat bikes masquerading as cat-equipped bikes:
    One inmate queries from Mexico and another from Oz: "Why do I have a cat but no o2/lambda?"

    A; NO, you do not have the cataclysmic avertor device on your R1100GS. You have a factory collector/pre-muffler that LOOKS like a cat, but is not a cat. It is an empty can, perhaps with baffles but not with rare earth metal catalyst. Should you buy a lambda sensor? Perhaps you are concerned about the plight of drowning polar bears. If that is the case, by all means, order a complete oem U.S. spec exhaust for your non-U.S., non-cat 1100. Hey, I'll trade you mine and throw in a brand new lambda sensor! Please re-cycle your CO-pot and old oem non-cat exhaust by sending it to me. I promise to keep it out of the landfill.
    Lambda sensors are required to prevent damage to the cat. No cat> no lambda sensor required.

    Limp Home re-visited:
    To understand what limp home mode feels like, simply unplug your TPS. Compare that to a disconnected o2 sensor.
    #43
  4. Poolside

    Poolside Syndicated

    Joined:
    Apr 1, 2003
    Oddometer:
    11,887
    Location:
    Silicon Beach, CA
    <BR>
    That'd be a good test for your bike bm. It will not hurt anything. Nothing needs to be reset afterward. You can unplug and/or replug the TPS while you are riding along if you like.

    <BR>
    #44
  5. johnjen

    johnjen I've Been Resigned

    Joined:
    Nov 28, 2001
    Oddometer:
    16,676
    Location:
    Seattle
    Uh just to clarify&#8230;

    Turn OFF the bike then plug or unplug a sensor. But not while the engine is running&#8230;

    This has been a public service announcement courtesy of the ADV technical department&#8230; :gerg

    JJ :evil
    #45
  6. Poolside

    Poolside Syndicated

    Joined:
    Apr 1, 2003
    Oddometer:
    11,887
    Location:
    Silicon Beach, CA
    <BR>
    What do you mean, are you thinking that will damage the ECU? I can tell you that as far as an ECU input is concerned, the absence of a signal is the same as the presence of a signal. 'No signal' merely reads as 'no signal'.

    A variable signal, including a zero signal, on the TPS and O2 sensor inputs is perfectly normal. The Motronic handles those signal changes live, just like when the throttle is moved.

    Maybe you're thinking that it's generally a good idea to not reach down and unplug the ECU while you're riding? I can get that as a good sort of caution. But heck man, how else can I find out what happens until I try it? :evil So far I haven't crashed from unplugging the TPS and O2 sensors, while riding.

    <BR>
    #46
  7. bananaman

    bananaman transcontimental

    Joined:
    Jun 20, 2007
    Oddometer:
    9,689
    Location:
    Madison, Wisconsin and/or Panama, Panama
    Maybe I'll tape an interview with Jason at Mischler's BMW. So that he can clarify what he meant by "bad O2 sensor." My O2 sensor wasn't non-functioning. My O2 sensor was mal-functioning, resulting in the engine running full-rich.

    In neutral, under no load, I could redline. Under load, 1st gear, redline. 2nd gear, redline. 3rd-5th gear, 65mph maximum speed.

    Maybe I could re-install the bad O2 sensor... P
    #47
  8. johnjen

    johnjen I've Been Resigned

    Joined:
    Nov 28, 2001
    Oddometer:
    16,676
    Location:
    Seattle
    It was meant as a caution for those who may not know that it is better to be safe than sorry (as in not tempt Mr. Murphy and possibly kill the ECU). Pulling a sensor cleanly out of the circuit usually isn't a problem (there are exceptions), but if done in a 'dirty' manner, well it isn't pretty…

    Besides you (and I) are technologically aware and proficient. Fools such as us take much for granted and if we fry a component, well, we chalk it up to experience. :norton

    And yeah it probably won't result in a dead engine but if it does you aren't going anywhere until the moronic brain has a transplant ($$). :huh

    JJ :evil
    #48
  9. rattis

    rattis Long timer

    Joined:
    Apr 3, 2010
    Oddometer:
    1,924
    Location:
    East of the seas of Kattegat
    not rellay to do with principal subject of this thread but rather with a comment in one of the replies.
    The one whos' more concerned about having a catalyic converter on his bike and may loose the odd hp than global heating should have his head examined and loose his right to combust anything .I'd better start a new thread....
    #49
  10. vintagerider

    vintagerider Long timer

    Joined:
    Nov 20, 2005
    Oddometer:
    1,976
    Location:
    the west
    Bananaman, the detatails help us to understand your experience. However I am still having difficulty imagining how any malfunction of the lambda can limit top speed, or what is meant by "engine running full rich" (from lambda sensor input). The motronic 2.2 is designed to ignore out of range signals from lambda by defaulting to a slightly richer idle and steady state cruise. Clearly you value and trust the opinion of your shop. Is it possible that while the lambda was faulty this was only coincidental to your complaint of limited top speed? Did you try disconnecting the sensor before replacing it? The point of this is to assist others in understand how motronic 2.2 uses different sensor inputs and what to check first when encountering very poor performance such as you described. TPS and HES sensors are critical, lambda is not. Did you discuss this opinion with your shop? You can always disconnect the new lambda sensor and observe the effects.

    Jim: Are you are saying that the life of the 1100 catalyst (U.S. models) will not be shortened if the lambda is disconnected for extended operation? I have not claimed to know the answer, instead, I posted that there was debate on that issue. Non-U.S. spec 1100 owners often mistake their pre-muffler/collector for a cat and ponder the absence of a lambda on their machines. Non-U.S. spec 1100 owners say they value the clarification because they find some of the motronic threads confusing.
    #50
  11. bananaman

    bananaman transcontimental

    Joined:
    Jun 20, 2007
    Oddometer:
    9,689
    Location:
    Madison, Wisconsin and/or Panama, Panama
    Tomorow I'll disconnect the sensor and report what happens.

    Quick question: does disconnecting the sensor achieve the same re-set as disconnecting the battery or pulling the fuse?
    #51
  12. vintagerider

    vintagerider Long timer

    Joined:
    Nov 20, 2005
    Oddometer:
    1,976
    Location:
    the west
    No, but that raises a good question. It has been reported that the ecu will retain "I have an 02 sensor config" until the power is disconnected. Upon power disconnect-restart ecu supposedly checks just once for an O2 or CO pot and also checks to see if a CCP is installed and what jumper setting. The re-set instructions were mainly for pp switching over to CO pot from lambda but are normally followed for other mods and any sensor changes.

    That you are willing to take the time to do this test and report your result will be appreciated by other 1100 owners. If you want to simulate what the shop says occurred, then no, do not re-set the ecu power after disconnecting lambda. The ecu should generate a "bad 02" fault code. Maybe ask your tech which lambda fault code he read? Shorted to ground?, Open? ect. We do not know at this point what type of lambda fault the dealer detected. The ecu is capable of storing several diiferent types of lambda fault codes. My understanding is that regardless of the kind of lambda fault detected by the ecu, its response should be the same which is to ignore the sensor.

    If you disconnect then you will be simulating the open circuit fault. Idling should be affected to some extent, followed by the ecu slowly adapting to running w/o lambda. That isn't the issue. You just want to see if the top speed limit returns. Not many expect this to happen but who knows? Perhaps ride a few days w/o lambda so the ecu can optimize its "dead lambda" compensation settings. Once you finish making observations w/o lambda and it is reconnected, then definitely re-set the ecu to clear fault codes so that the ecu can again recognize that a good lambda is present. Just as before, you will have to ride for a few days under varying load conditions for the ecu to optimize.
    #52
  13. Poolside

    Poolside Syndicated

    Joined:
    Apr 1, 2003
    Oddometer:
    11,887
    Location:
    Silicon Beach, CA
    <BR>
    &#8194;&#8194;:lol3&#8194;&#8194;+1 fool&#8194;&#8194;+1 take much for granted


    <BR>
    #53
  14. Poolside

    Poolside Syndicated

    Joined:
    Apr 1, 2003
    Oddometer:
    11,887
    Location:
    Silicon Beach, CA
    <BR>You know, if run that way for a while, a disconnected TPS sensor will cause an 'O2 sensor out of range' fault code.

    <BR>
    #54
  15. johnjen

    johnjen I've Been Resigned

    Joined:
    Nov 28, 2001
    Oddometer:
    16,676
    Location:
    Seattle
    Ah and the thought plickens…

    JJ:evil
    #55
  16. Jason1202GStime!

    Jason1202GStime! Been here awhile

    Joined:
    Feb 25, 2010
    Oddometer:
    253
    Location:
    Singapore
    Hi Guys,

    How do i know the O2 sensor is not working? I use the 911 to check, it appear that the mili volt reading from the 2 sensors are very different.
    #56
  17. johnjen

    johnjen I've Been Resigned

    Joined:
    Nov 28, 2001
    Oddometer:
    16,676
    Location:
    Seattle
    If the O2 sensor is 'out of range' the fuel injection brain will log a fault. This can be read either with a GS-911, a BMW engine diagnostic tool or (I think) the RID.

    If they are just different but not out of range then either there is something wrong with the engine or one of the O2 sensors has a problem. It's difficult to tell from this side of my keyboard.

    JJ
    #57
  18. Ozav8r

    Ozav8r Been here awhile

    Joined:
    Oct 26, 2012
    Oddometer:
    159
    Location:
    Sunbury, Australia
    I was wondering if any of the guru inmates can tell me; can you get access to the O2 sensor connection under the fuel tank by merely lifting the tank at the back (after loosening it) or does it require a full removal with the associated draining of fuel etc. I had an unsuccessful (the freakin' thing was Locktite'ed in or sumthin') go at removing the O2 sensor last week in LA, but had to abandon it and put it all together again as I was running out of time for my flight back to Oz. I am going back to LA tomorrow and will have another crack at it, but will do a pilgrimage up to Sears (rather than Harbor Freight) for some serious wrenches (rather than the adjustable wrench i wrecked trying to get the plug out) and a butane torch, to heat the freakin' thing up before haaving another go at the big heave. Just thinking it would be better to get the whole catcon assembly disconnected away from the bike before I try again, as I am likely to topple the pig over myself if I don't:eek1 Also, as I was using an adjustable wrench last time - it looked like a 22mm plug - can anyone confirm that?

    Thanks folks
    #58
  19. Poolside

    Poolside Syndicated

    Joined:
    Apr 1, 2003
    Oddometer:
    11,887
    Location:
    Silicon Beach, CA
    <BR>I don't know about access to the connector on the 1200. But the tank doesn't need to be drained to remove it.

    The sensor is 22mm across the wrench flats. After the harness is disconnected and fished through the frame, I think use a 22mm combination wrench and a 3 to 4 pound sledge to tap on it. 'Dead' blow hammers are the wrong tool to create a 'live' inertia impact with.

    Hammers from Harbor Freight
    3lb: http://www.harborfreight.com/3-lb-drilling-hammer-with-hardwood-handle-67817.html
    4lb: http://www.harborfreight.com/4-lb-drilling-hammer-98258.html


    <BR>
    #59
  20. Ozav8r

    Ozav8r Been here awhile

    Joined:
    Oct 26, 2012
    Oddometer:
    159
    Location:
    Sunbury, Australia
    Thanks you so very much Poolside - I really appreciate your knowledge and reply. I should have mentioned - my bike (I got it only a couple of months ago) is a 2000 R1150GS with 14K miles on it. I will take a trip up to Harbour Freight on Carson/Woodruffe and get the 4 pound sledge you suggested and get stuck into it when I get to Long Beach on Friday. Thanks again - ride safe!
    #60