R1100s Starts Fine, But Soon After Runs Awfully

[Tim Stammers]

My R1100s starts and runs fine, but within five minutes (New York summer, 90deg) it won’t idle (slowly fades and dies) and when pulling away from a stop misses and farts very badly. And then when I get up to about 3,000rpm-ish, it runs fine, with full power, no problems. Then at the next traffic lights, the same behavior again.

 

I’ve done some preliminary checks, but not found the problem (see below).  I know I shouldn't clutch at straws, but I think the sensor at the back of the airbox (the O2 sensor?) may have failed. I’ve read that the signal from this sensor is only used at low RPM, so that would fit the symptoms.

 

QUESTION: Is it OK to test by running the bike with that back-of-airbox sensor disconnected?  I want to try this before starting the usual fuel/ignition/compression check list.

 

Background for anybody who wants to read further or make other suggestions:

 

·         2001 R1100s, 30,000 miles. No engine mods except canisterectomy and quick disconnects on fuel lines about 3,000 miles ago.

·         Oct 2021: No spark. Replaced coil. Problem fixed.  (Mentioned because the current problem could be classic warm-coil failure, although doesn’t seem to fit with the good running at higher rpm.)

·         Jan 2022 – Preventive maintenance: replaced Hall sensor. Bike ran fine after.

·         April 2022: Preventive maintenance: replaced clutch and fuel rail/regulator valve (broke original rail during the clutch job, replaced with a second-hand one.) Bike ran fine after.

·         June 2022: Preventive maintenance: replaced u-bend hose and fuel filter inside tank. Then the terrible running started.

·         Opened gas cap. No change.

·         Throttle-cables are sitting properly above throttle body. 

·         Tank off, checked all electrical connector in case not connected properly. All *look* fine.

·          Replaced spark plugs (a whim.) 20-minute test run, problem apparently fixed. Phew.

·         At next outing the same terrible low-rev running returned, within five minutes of start.

·         Checked injector spray pattern while cranking with starter. I’ve never done this before, but they looked like I guess they should – an even cone, same on both sides.

·         Went back inside fuel tank. All looks fine, except broken wire to low-fuel sensor (clumsy me.) Very small amount of what looks like sand at bottom of tank. Mesh sock is clean. Fuel filter has no direction arrow and is symmetrical, so I don’t know which way it should point. When I fitted it, I blew through both ends, couldn’t feel any difference either way.

·         Blew through vent pipes (the ones that went to the long-since removed canister). They were clear.

 

Despite the problem starting after I’d been inside the tank, I don’t think it’s a fuel starvation, because at higher rev’s it runs fine. As above: Can I test for failure of airbox sensor by running the bike with it disconnected? Any other suggestions before I start the usual air/ignition/compression checks? (I could share the check list I’ve drawn up, but this post is already very long.)

 

Cheers

Tim, Long Island, NY

[dirtrider]

1 hour ago, Tim Stammers said:

My R1100s starts and runs fine, but within five minutes (New York summer, 90deg) it won’t idle (slowly fades and dies) and when pulling away from a stop misses and farts very badly. And then when I get up to about 3,000rpm-ish, it runs fine, with full power, no problems. Then at the next traffic lights, the same behavior again.

 

I’ve done some preliminary checks, but not found the problem (see below).  I know I shouldn't clutch at straws, but I think the sensor at the back of the airbox (the O2 sensor?) may have failed. I’ve read that the signal from this sensor is only used at low RPM, so that would fit the symptoms.

 

QUESTION: Is it OK to test by running the bike with that back-of-airbox sensor disconnected?  I want to try this before starting the usual fuel/ignition/compression check list.

 

Background for anybody who wants to read further or make other suggestions:

 

·         2001 R1100s, 30,000 miles. No engine mods except canisterectomy and quick disconnects on fuel lines about 3,000 miles ago.

·         Oct 2021: No spark. Replaced coil. Problem fixed.  (Mentioned because the current problem could be classic warm-coil failure, although doesn’t seem to fit with the good running at higher rpm.)

·         Jan 2022 – Preventive maintenance: replaced Hall sensor. Bike ran fine after.

·         April 2022: Preventive maintenance: replaced clutch and fuel rail/regulator valve (broke original rail during the clutch job, replaced with a second-hand one.) Bike ran fine after.

·         June 2022: Preventive maintenance: replaced u-bend hose and fuel filter inside tank. Then the terrible running started.

·         Opened gas cap. No change.

·         Throttle-cables are sitting properly above throttle body. 

·         Tank off, checked all electrical connector in case not connected properly. All *look* fine.

·          Replaced spark plugs (a whim.) 20-minute test run, problem apparently fixed. Phew.

·         At next outing the same terrible low-rev running returned, within five minutes of start.

·         Checked injector spray pattern while cranking with starter. I’ve never done this before, but they looked like I guess they should – an even cone, same on both sides.

·         Went back inside fuel tank. All looks fine, except broken wire to low-fuel sensor (clumsy me.) Very small amount of what looks like sand at bottom of tank. Mesh sock is clean. Fuel filter has no direction arrow and is symmetrical, so I don’t know which way it should point. When I fitted it, I blew through both ends, couldn’t feel any difference either way.

·         Blew through vent pipes (the ones that went to the long-since removed canister). They were clear.

 

Despite the problem starting after I’d been inside the tank, I don’t think it’s a fuel starvation, because at higher rev’s it runs fine. As above: Can I test for failure of airbox sensor by running the bike with it disconnected? Any other suggestions before I start the usual air/ignition/compression checks? (I could share the check list I’ve drawn up, but this post is already very long.)

 

Cheers

Tim, Long Island, NY

Afternoon Tim

 

That sensor in the air box is the IAT (Intake Air Temperature) sensor not the o2 sensor. You can run without that hooked up for testing with no problems. As a rule the air box AIT sensor won't cause a problem like you are getting

 

Th o2 sensor is in the exhaust not the air box. You can also run with that disconnected for testing. On your 1100S the o2 sensor is on the front of the catalytic converter between the two front exhaust pipes where they enter the cat (just follow the wire pigtail to the connector). Or a possibility the o2 sensor wire pigtail harness is burning/shorted on the hot exhaust. (inspect for that)

 

It really doesn't sound like the coil but you can't totally count that out either. 

 

Any chance that when you had the fuel tank off for U hose replacement that you got the o2 sensor pig tail harness routed across, or along the R/H side spark plug wire? That can make them run bad at hot idle & at lower throttle openings.

 

It might not hurt to do both a cranking compression test & a cylinder leak-down test. If you have a leaking intake valve they can run like crap at idle & at lower engine RPM's.

[Jim Moore]

That sounds a lot like a bad O2 sensor. The S model is put together a little differently than most oilheads, but on most bikes the connector is under the tank on the right side. You can ride with it unplugged.

[Tim Stammers]

fantastic, thanks guys. When it cools down later today, I will get to work and report back.

[King Herald]

Have you checked valve clearances recently? It sounds unlikely to be that seeing as the problem start suddenly though, but worth a look after all the other work you have done. 

 

My bike started acting just like yours when I mistakenly filled the tank with 80% ethanol fuel in France a few years ago. Horrible at low rpm, but normal above 3000 rpm. 

[Tim Stammers]

Keeping this thread alive: Am waiting for Amazon to deliver proper clamps for pipes in gas tank (plain band type.) Then the tank goes back on, and I will run bike with the O2 sensor disconnected, and do a compression test (which has to be done on hot engine right?)

 

Thanks to King Herald for tappet suggestion. I've adjusted those twice in last 200 miles (wasn't sure I got them right first time.) Will do it again anyway while I wait for the fuel pipe to arrive.

 

The O2 sensor cable was/is nicely routed near the cat box and RHS header, using that cast nub/projection on the engine casing. The cable (double-sheathed - plastic outside, fabric inside) and connector looks fine. Cable was/is routed about an inch under the HT lead, and at closest point is about 1" away from it ( but I don't like that it just sits on the cylinder head.) I've tried to attach a pic of it to this post, but think I've failed.

 

[Tim Stammers]

Looks like  the 02 sensor is the problem.

 

Tank refitted, O2 sensor  disconnected. Rode for an hour with many traffic stops, and the  bike ran fine. Unless I find that the O2 sensor cable is bust (it  doesn't look it) I presume  the best  thing to do next is to replace  the entire  sensor and cable, because according to what I've  seen on the  web, there's no simple  bench  test for  it.

 

[Jim Moore]

Just leave it unplugged . It runs better like that anyway. Cover the bike side of the plug securely. There is still power there. If it gets wet it can cause issues. 

[dirtrider]

9 hours ago, Tim Stammers said:

Looks like  the 02 sensor is the problem.

 

Tank refitted, O2 sensor  disconnected. Rode for an hour with many traffic stops, and the  bike ran fine. Unless I find that the O2 sensor cable is bust (it  doesn't look it) I presume  the best  thing to do next is to replace  the entire  sensor and cable, because according to what I've  seen on the  web, there's no simple  bench  test for  it.

 

Morning Tim 

 

That pretty well (within reason) proves out your o2 sensor as the problem. 

 

Your 1100S actually uses the 1150 Ma 2.4 fueling control system so really should have an operational o2 sensor. The 1100S will run OK with the o2 sensor disconnected but they were not designed to run open loop, especially on E-10 (10% alcohol) fuel as the open loop fueling maps were designed around non-alcohol containing  fuel.  

 

The BMW OEM o2 sensor is about $336.00 (expensive) so a lot of money for a simple narrow band o2 sensor. 

 

Seeing as your o2 sensor pigtail harness seems OK then you can splice on a universal o2 sensor (do not solder the splice connections as that blocks the reference air flow down the wires to the o2 sensor).  

 

Or you can buy a cheaper will-fit but works good o2  from places like  https://www.euromotoelectrics.com/searchresults.asp?Search=oxygen+sensor&Submit=       Or even buy a universal type 02 sensor from Euromotoelectrics.

 

Or even buy a Bosch universal o2 sensor from your local auto parts store (just match to your current sensor) then splice on your present o2 sensor pigtail.  

 

If you install a universal o2 you need to pay strict attention to the o2 sensor circuits not the wire colors as some universal o2 sensors  have the same basic wire colors but the the circuits don't match the BMW wire inputs. 

 

You have a 2001 1100S  so that shouldn't  be an issue as BMW changed the o2 sensor connector sometime late in 2002. Just make sure that what you buy has the same pigtail connector (plug) as your present o2 sensor.  (on the universal o2 sensor it doesn't matter as those use part of your present o2 sensor pigtail & it's connector)

[Tim Stammers]

For the  record/posterity:

 

I replaced the suspect O2 sensor with a new one from Beemer Boneyard for about $80 USD inc shipping. During a 15-mile, 45 minute  ride in stop-start traffic, the bike then behaved perfectly - idled happily at 1.1krpm, pulled hard when I opened it up.

 

So it  looks extremely likely the problem is fixed. Thanks again to JIm Moore and DirtRider for saving me from days of testing and diagnosis.  

 

BTW: You don't need one  of those claw-foot 22mm slotted adaptors to remove and refit the O2 sensor. I did the  job with an open-ended 22mm wrench/spanner, although  this did  mean that I couldn't use a torque wrench, and instead just tightened down the sensor hard. I'll keep checking that it hasn't come loose. 

[Paul De]

21 hours ago, Tim Stammers said:

I did the  job with an open-ended 22mm wrench/spanner, although  this did  mean that I couldn't use a torque wrench, and instead just tightened down the sensor hard. I'

Wouldn’t worry to much about that.  With enough time spinning wrenches, the wrist-elbow torque meter can be pretty spot on.   My brother and I would do bets on who buys beers for torque by feel.  It was surprising how accurate you can be.  Sadly I bought more beers than he did, but we were always no worse than +\-10%.   I wouldn’t do it for torquing a cylinder & head but for a lot of fasteners it is like horseshoes and hand grenades…close counts.