GS-911 Error Code Question - Lambda Sensor Related

[roger 04 rt]

So if it is sorted, can someone some up simply what the problem was, and how it was fixed...because I have not got my head around this thread, but I want to.

Andy

 

Trail'R purchased an AF-XIED in order to improve the driveability of his 2001 R1150GS. Several months after installing it he reported the following:

 

I hooked up the GS-911 to my 2001 R1150GS and pulled the following error codes:

 

304: Oxygen (Lambda) Sensor Circuit Malfunction, Short-circuit to Earth.

This fault occurred 69 times.

The fault is not present now.

4400: Lambda-Control, Upper control limit reached

This fault occurred 72 times.

The fault is not present now.

 

Any thoughts on the cause of the errors?

 

I do have an AF-XIED at setting 7 that I installed this winter. I will also note that I have a 2004 R1150RT and the GS has always had a small amount of surge in low gears compared to the RT. I have always thought it was a lean surge so after reading Roger's threads I installed the AF-XIED in hopes that it would help. Recently I have noticed my GS fuel economy really going down. Unfortunately I installed the AF-XIED at the same time as a larger Aeroflow shield, so I am not sure about root cause on fuel consumption or if it is even related to the error codes.

 

Your thoughts are appreciated.

 

I hope that Trail'R can add some color to this but here is what I see:

 

1) During the course of debugging he found a hole wearing in the fuel pump. This might have added to the issues but perhaps not. In any event a future failure was avoided.

 

2) The O2 sensor was old and tired and not able to output a high enough rich voltage to provide a proper signal to the AF-XIED. This led to the 304: Oxygen (Lambda) Sensor Circuit Malfunction, Short-circuit to Earth error.

 

Although the O2 sensor would have continued to work for Closed Loop Trail'R discovered a good, inexpensive replacement, the Bosch 15109.

 

3) The idle speed on the motorcycle was raised just high enough that a large part of the Mixture Adaptation budget for the Idle-Air-Leakage Long Term Trim was used up correcting what the Motronic saw as leaking air. This caused a few problems:

 

a) It directly caused the 4400: Lambda-Control, Upper control limit reached error. When the AF-XIED looked for more Mixture Adaptation, the Motronic was unable to adjust further.

 

b) The high Idle RPM (due to BBS setting) led to leanness during cold starting. This appears to have been made worse when the Yellow Coding Plug was installed. Over time, I believe that Mixture Adaptation would have corrected the lean Cold Start.

 

Summary

This debugging has led to an awareness that in addition to proper setting of the TPS, it is equally as important to set idle speed to the "book" value. The combination of TPS registration and idle speed are used as the baseline for the Motronic in its low-throttle-angle calculations.

 

The logic goes something like this:

 

--The Motronic assumes that the TB throttle stops were set correctly at the factory at 0.32 degrees open (0.32 is what the GS-911 reports for closed throttle after learning).

 

--It controls the mixture to an AFR of Lambda=1 and "commands" you to adjust idle speed to 1100 RPM. By doing this it knows the fuel its using and makes you tweak the air correctly for 0.32 degrees throttle.

 

The result is a full alignment of Idle and Small Throttle Angle.

RB

 

[AndyS]

So, The fitting of the AF-XIED was fine however the O2 sensor was bad and the idle was set too high!

 

Wow, well done to all involved. It seems to have taken a long time to get to the resolution.

Also I guess that the GS-911 was a crucial tool (and moreover, understanding what it was telling you).

 

[roger 04 rt]

You're right on all counts. And it did take a while but Trail'R was thorough and systematic. If he hadn't been we would not have found the root cause. Without a GS-911 I don't see how we would have gotten there either.

 

For about a month he worked through the usual suspects. Then about a month ago, the lean cold-start had me wondering about excess air.

 

Because the Mortronic doesn't report any Short or Long Term Trims to the GS-911, you can't tell when the short term is soft-faulting and ratcheting up the long term trim. However, I can now see if all sensors are working normally, and in Closed Loop you get Lambda closed-loop control dropouts, that's what's happening. That's a big bit of new knowledge.

 

If the 4400 fault occurs you also know that a trim is faulting.

 

The second deduction is that when you adjust idle speed higher than 1100 RPM (and there is no way to tell this), you are using up the Motronic's trim range. Who would have guessed that?

[Trail'R]

Roger's summary is right on the mark for all of the steps taken. The first symptom that brought all of this to light was very high fuel consumption, possibly brought on by a decaying oxygen sensor coupled with the AF-XIED. The original oxygen sensor would produce a steady oscillation, so I thought initially we are good to go. With Roger's knowledge he was able to determine that is was "lazy" as the amplitude was lower than normal. It also turns out I was working against more than one issue, so that made it difficult to diagnose as well.

 

In regards to time, I needed to be very methodical and make a single change before a new cold start measurement. This helped to isolate the issues as I continued to make cold starts. The cold starts turned out to be a precious commodity as I could make changes to the bike faster than I could log due to the cool down time. I would usually do a start early in the morning and then after work to log data.

 

What was really puzzling to me was when Roger was convinced that there was an air leak. I have been all over the bike and was very confident that there were no leaks in all of the usual places. As it turns out those big brass screws act like an air leak and created the symptoms Roger was noticing in the log. I was convinced I was looking for an electrical issue due to the closed loop dropping out, glad it turned out to be something much easier to correct.

 

What is still amazing to me is that an idle setting of 1200-1250 creates the closed loop drops noticed in the Motronic. Keep in mind that everything else at this point was adjusted to factory spec with no modifications to the bike. I kept mentally reviewing parts of the system that may be out of spec keeping in mind that the TB's are perfectly balanced through all RPM ranges, there are no air leaks, new O2 sensor, stock exhaust, new air filter, injectors cleaned and balanced, TPS is set to factory spec, air sensor is measuring correctly...everything was as it should be other than the idle level. The RPM input in to Motronic appears to be the root cause of the closed loop drops. It is one of the inputs where "close enough" just does not apply for my bike. Maybe this is specific to the single spark GS.

 

I have owned the GS-911 for 6 years and this was the first time I have needed to put it to use. I would check for errors occasionally and after finding none I would stash the device back into the toolbox.

 

I am working to put in some miles to make sure the bike is running as it should. I will follow-up after some significant riding time.

 

A big thanks to Roger for taking the time to share his knowledge. My learning curve was very high and I now have a greater appreciation of the dark arts of Motronic. There were some new lessons learned that I hope will help another rider down the road should they be facing the same issue.

 

Now, on to my other project where I can apply my newly acquired knowledge to the R1150RT, which is what brought me to this fantastic forum.