R1100S idles way too slow and tries to stall when hot - Why?

[roger 04 rt]

...

Johnny, there's nothing holy about the friggin' blue paint. If you have and can use a good DVM to check & adjust the TPS (several on-line articles are available, here, on the R.A. tech pages, and in the ADVRider pages), and you know how to balance the throttle bodies, there is nothing to be lost by just doing it. Heck, it may even be as simple as backing out the BBS's another eighth- or quarter-turn each. But Roger is right too in that there's no substitute for the scientific method of seeing exactly what's happening in there.

What I think is important about the blue paint is that alpha-N fueling like the Motronic uses is very sensitive to throttle angle for the first few degrees of throttle rotation (see the +/-17% power variation for +/- 1.5 degrees throttle error in the chart below). Therefore it is important that the same air flows through each TB. Our vacuum sync is something of a proxy for the amount of air flowing but isn't a direct measure.

When the factory aligned the throttle stops they had the benefit of flow equipment, which even the Bing rebuild service said they lacked when I asked them.

When you use a procedure like zero=250, with clean throttles and clean BBS channels, we're, in effect, using the bypass channels as an airflow reference, hoping that each channel passes the same amount of air at the same number of BBS turns open. Of course the vacuum draw from each cylinder is probably different so even if the bypass hole in the TBs are equal, you still may not be able to get equal flow at small throttle angles.

For these reasons, I think the BMW recommendation to leave the throttle stops as set is smart.

 

[pauls1150]

Well explained (as usual!) and no argument here.
But the factory did it on an assembly line, and the procedure involved at least one human...

At least three of the big shops here had no qualms about re-doing the TPS (followed by a careful rebalance) on bikes that had idle or vibration issues. Of course they knew that the Faktory preached not to, but nobody wants to send their throttle bodies to Germany for a few weeks (not to mention Customs delays). Sometimes it helped, sometimes it didn't. But they used bright red paint to show that they had been dinked with.

Without a GS-911 and a handy PC out in the garage, it would be a Good Thing to at least measure the Idle voltage between pin 1 of the TPS (rear-most, white wire w/ a red runner) and the closest ground.

 

[montliz]

Thanks, Roger and Paul, and thanks again for the great advice, tips, and specs.

I see advanced electronic learning in my immediate future .

I don't have access to (or know-how about) a GS-911, but I do have a good DVM and can check the voltages as you suggested, and I will do that this week. I love learning new things and this is a wonderful unintended consequence of owning this bike.

Roger, two questions: generally speaking, would the AF-Xied tend to or try to compensate for a defective O2 sensor? And do I understand correctly that the fuel injectors should stay dry and not spray or drip any fuel during the test in which we pull them and observe them while cycling the fuel pump?

By the way, I checked on the price of new fuel injectors just for fun. Those beauties go for a cool 150 USDs each, so I think my first move will not be to buy a fresh pair of them. Ditto the O2 sensor at 265 dollars. Yikes!

Paul, I'm hearing from you that fuel and idle issues are (How's a nice way to put it?) not uncommon on these bikes, so at least it's not just me. And I appreciate and agree with what you said about the throttle stop adjusters.

This afternoon I'm headed off to Houston to stop by our BMW dealer to ask their head fuel man and oilhead fan what he thinks about the idle and what they would do. That'll give me another hundred miles and let me run through another tank of fuel on this beautiful warm afternoon.

Best,

Johnny

 

[rxcrider]

r1100s idles way too slow and tries to stall when hot - Why?

Things that I don't think have been checked based on a thread search :

Fuel pressure
- my friend's R1100S did all sorts of odd thing when his "U" fuel hose was failing.

Spark plugs
- when I swapped the used Bosch twin electrodes for a set of iridium autolites, my idle climbed 200 RPM

Voltage
- what if system voltage is low at idle...?

Timing
What would happen if the cup for the hall sensors came loose and the motronic timing was advanced? I know there were bikes that wouldn't run because of this.



Sent from my iPhone using Tapatalk

 

[roger 04 rt]

...

Roger, two questions: generally speaking, would the AF-Xied tend to or try to compensate for a defective O2 sensor? And do I understand correctly that the fuel injectors should stay dry and not spray or drip any fuel during the test in which we pull them and observe them while cycling the fuel pump?


...

The AF-XIED can't compensate for a weak O2. If anything it would be telling your bike that the exhaust was lean. Possibly that would add too much fuel but it has not caused that issue in the past. The fuel injectors should not spray or drip at all unless the engine is turning over.

Things that I don't think have been checked based on a thread search :

Fuel pressure
- my friend's R1100S did all sorts of odd thing when his "U" fuel hose was failing. His bike runs great above idle so it doesn't seem likely.

Spark plugs
- when I swapped the used Bosch twin electrodes for a set of iridium auto kites, my idle climbed 200 RPM

Voltage
- what if system voltage is low at idle...? If the voltage dropped very low, the injector turn on time could extend causing leanness. This is a good thing to check.

Timing
What would happen if the cup for the hall sensors came loose and the motronic timing was advanced? I know there were bikes that wouldn't run because of this.



Sent from my iPhone using Tapatalk

 

[roger 04 rt]

Well explained (as usual!) and no argument here.
But the factory did it on an assembly line, and the procedure involved at least one human...

At least three of the big shops here had no qualms about re-doing the TPS (followed by a careful rebalance) on bikes that had idle or vibration issues. Of course they knew that the Faktory preached not to, but nobody wants to send their throttle bodies to Germany for a few weeks (not to mention Customs delays). Sometimes it helped, sometimes it didn't. But they used bright red paint to show that they had been dinked with.

Without a GS-911 and a handy PC out in the garage, it would be a Good Thing to at least measure the Idle voltage between pin 1 of the TPS (rear-most, white wire w/ a red runner) and the closest ground.

I agree with you on the TPS. No harm resetting it to 340 mV as a voltmeter is a fine reference instrument. As to the throttle stops, I think getting them right is tougher but it is doable if you are careful. One way to eliminate the side to side vacuum difference from the cylinder would be to temporarily swap them side to side, if that is even possible with the throttles disconnected. Then you could take the average of each side.

 

[pglaves]

In the old days we set idle throttle plate openings on multiple carburetor cars with a needle or wire feeler gauge. Setting the opening directly may be superior to using vacuum as a surrogate for airflow as a surrogate for opening cross section.

 

[roger 04 rt]

Interesting idea Paul. Recently someone tried something like that, using thread pitch and a dial indicator from the top on the TB bore as a cross check. The idea was to count the number of turns on the left required to achieve 90 mV TPS movement after a dial indicator started moving. Then duplicate the number of turns and confirm the same dial indicator movement on the right side. He had good results.

 

[rxcrider]

Roger - any Idea of average vacuum (relative to ambient) in the intake manifolds at idle? I have access to a small flow bench and a set of 1996 throttle bodies on the shelf with only 20k on them. Maybe I should make an adapter plate over the winter and get some measurements before installing them on my bike - would give me a baseline for rebuilding my 90k mile set on my bike as well. My right side is mildly clicking away and I had to get creative with idle stops and cable adjustments to get it running nicely. I figure new cables and low mileage throttle bodies along with rings and resealing the base and new head gaskets and a good de-carbon-ing (apparently that isn't a word, but it should be) should make things better yet.

 

[roger 04 rt]

I don't know the idle vacuum level and don't have the equipment to measure it. But if you have the bench you can measure it.

Here's my first thoughts on how to set things up.

Set the left throttle body as follows:
1) Back off the throttle plate stop screw and close the throttle.
2) Set the TPS to 250 mV and temporarily lock it.
3) Adjust the throttle stop screw to 340 mV and lock it.
4) Adjust the TPS to 340 mV and lock it.

Set the right throttle body as follows:
1) Close the left BBS screw.
2) Measure the flow for the left throttle body on your set up.
3) Mount your right throttle body and close its BBS.
4) Adjust the right throttle stop screw until its flow matches the left TB and lock it.

To finish, set both BBS to three turns open so that you have enough air to start. The follow your sync procedure for the throttle bodies. Here's the one I use now: http://forums.bmwmoa.org/showthread...nd-Zero-Zero&p=1003851&viewfull=1#post1003851.

 

[rxcrider]

I'll have to grab a vacuum gauge I trust (not the one on my ancient mighty-vac and get an idea of where my bike runs at idle against the stops. That way I know where to set the bench. I figure I'll get baseline measurements as is and with the BBS closed on the low mileage set before I start dinking with the stop screws on them. This will be a winter project for sure. Despite everything I know I should be doing to it, my bike is running too well to not ride it at the moment. Then when winter comes I'll go snowboarding, so I probably really mean I'll be doing this on a rainy day next April or May. Sorry for the hijack - back to waiting for your next update, montliz.

 

[montliz]

Thanks to everyone's considerable effort, this thread has turned into a gold mine of good advice and excellent links. Much, much appreciated by me and by others trying to address this particular problem in the future.

Although I am no kind of electrical engineer, and am still somewhat intimidated by the lingo and tools needed to go there, I'm eager to learn and am looking forward to applying these tips and specs to this project.

Update: I had a most interesting talk yesterday with Nick, the service manager and one of the oilhead lovers at Gulf Coast BMW up in Houston, and he was kind enough to listen to the short version of my service history with the bike and then come out into the sunshine and listen to it run and play with the throttle a little bit. He said it sounds perfectly healthy and normal to him except for the disappointing and iffy idle. Then he said (and here I'm going to paraphrase and try not to misquote) that what may have happened and what he has seen happen before is that the throttle bodies over time get gunkier and gunkier (I think that's the technical term he used) and that the EFI gradually adjusts itself to the deteriorating TB conditions more or less successfully until the TBs get so gunky that the engine starts to starve out at idle and won't stay lit at the traffic light anymore. That's the point at which the owner/shade tree mechanic steps in and removes the throttle bodies and airboxes and cleans everything back to its original pristine condition.

The problem then becomes that, although everything is now clean and beautiful, the TPS and its little helpers can't instanly readjust to the new setup and so the idle performance doesn't just snap right back into place like we all hoped it would.

So, a couple of ways to go: Your mechanical repairs appear to be good, so leave the bike here and we'll hook it up to our computer at the earliest possible opportunity and remap and reset and tweak whatever electronics need tweaked, and you'll ride out of here good to go - or - if you don't mind having to help it idle for a while as needed and if you're an experienced enough rider to not let that bother you and you don't consider it a safety issue, then go ahead and ride it for a while and put some miles on it and check the sync every week or so and see if it doesn't clear itself up to your satisfaction. It may very well do that. If it doesn't continue to show progress, or if you get tired of fooling with it and feel like spending a few dollars on the old girl and can get a ride home, bring it in here and we'll make it happen fer ya.

Great conversation; once again made me proud to be a BMW owner. Whaddya think?

Best.

Johnny

 

[mneblett]

Good to hear of a service manager with a good head on his shoulders.

Why not just remove the Motronic fuse for 10 minutes to clear the learned adaptations?

 

[roger 04 rt]

Good to hear of a service manager with a good head on his shoulders.

Why not just remove the Motronic fuse for 10 minutes to clear the learned adaptations?

Based on the service managers theory, that will do everything he can do with his machine. Then remember the key on, rotate throttle fully twice, key off ... To relearn the throttle. Although this may not solve the problem it is a good next step.

Have you tried measuring the lambda sensor voltage at idle?

 

[bikerfish1100]

^ and ^^, that's exactly what I was thinking. I'd sure give that a shot next.

 

[montliz]

Good plan - Thank you sirs.

I will remove the fuse and reset the motronic unit today.

Roger, I have not measured the lambda sensor voltage yet; that will be something which is currently outside of my so-called area of expertise. I'm looking forward to trying that and learning how to do it.

Thanks,

Johnny

 

[montliz]

Question: Do I cycle the ignition or anything while the fuse is in my hand or do I just wait a few minutes and reinstall it and then perform the ignition/throttle reset cycle?

Johnny

 

[mneblett]

Question: Do I cycle the ignition or anything while the fuse is in my hand or do I just wait a few minutes and reinstall it and then perform the ignition/throttle reset cycle?

Johnny

Just wait -- you are allowing the memory to "fade away" and be lost. Then reinstall fuse, turn the ignition on (engine not started), fully open the throttle twice, and turn the ignition off.

 

[montliz]

Hi all,

I performed the Motronic reset procedure today but it didn't seem to make any difference in the idle. I didn't go out on the big road and put any miles on it today because a big thunderstorm came up and I made an executive decision to put that off until tomorrow, but I didn't see any change in the idle during my ten minutes up on the bench.

Mark, I think I remember reading somewhere that the Motronic and TPS in my early R1100S (mfg. 03/2000) don't respond to this procedure, and that this reset procedure is applicable only to somewhat later models. Could that be so? That might explain why Nick up at Gulf BMW didn't suggest that I do it.

Anyway, thanks for the great tip and I'm planning to go for a ride tomorrow and run some more gas through it and see if it continues to improve.

Johnny

 

[roger 04 rt]

Johnny, Every R1100 and R1150 resets by pulling the Motronic fuse. Is it fuse 5 on the R1100S or another?