Dropping idle mystery: It's back.

[JBBenson]

My uneven idle is back, although now it only drops suddenly without warning, in traffic, or idling in the garage with a cooling fan. It also drops suddenly and more predictably when coming to a stop and/or stopping.

 

Videos here:

 

Steady idle, then sudden drop:

 

http://vid48.photobucket.com/albums/f209/JesseBruceBenson/dropping_idle.mp4

 

Dropping, with throttle blip, steady idle, then sudden drop:

 

http://vid48.photobucket.com/albums/f209/JesseBruceBenson/throttle_blip.mp4

 

Consider this:

 

2001 V11 Sport

6000 total miles.

New intake rubbers

New fuel filter

New air filter

Valves at .006" intake and .008" exhaust

Fast idle cam not interfering

Tightened up exhaust, no popping or surging

Upgraded head temperature sensor (all brass model)

Cleaned throttle bodies, butterflies.

Guzzitech ECU map to match Stucchi crossover and Mistral mufflers.

TPS: 150mV disconnected, 525mV connected

Throttles in perfect balance at idle and off idle 3500+ RPM

 

The bike runs beautifully, but WILL NOT GODDAMN IDLE.

 

I suspect a fuel pump issue. Any other ideas?

[docc]

>Clean the ignition switch

 

>What are you running for relays?

 

>Remove the spark plug wire caps at the coils and inspect for ugliness.

 

>Melting or charred 30 amp fuse?

 

>Remove the brass idle air bypass screws and clean them along with spraying out the passages.

 

>The brass temperature sensor holder gave my Sport fits. I would only consider it an "upgrade" if the plastic one keeps breaking.

 

This links to a thread that traces my odyssey of chasing the Nasty Hiccup in my Sport for over a year (warning: over 120 posts ) which revealed numerous faults and culminated in discovering the TPS had a little glitch that I only discovered with a careful resistance test while rotating it (it did not show up watching the transition in millivolts, at least not with my meter).

 

Also: a synopsis of my experience with various Engine Temperature Sensor holders  (YMMV! )

[JBBenson]

Yes will try cleaning ignition switch. Because everything runs through there?

 

Relays are GEI, mounted to aluminum strip and isolated from vibration with rubber grommets.

 

Spark plug wires at coil....hmmm, I do suspect this is electrical or fuel related, I will check that.

 

Need to check 30 amp fuse. Under seat, right?

 

I have cleaned the bleed screws already but can do again.

 

The brass temperature sensor: This problem really showed up when I installed it after struggling to remove the old crumbly one. I will replace this first to see what happens.

 

It does feel like the ECU is getting the wrong info and changing the mixture suddenly. Its like a switch gets thrown and the idle drops. Maybe the brass sensors are for the V11's with the Lambda sensor, which I don't have.

 

Thanks for the reply.

[emry]

"Guzzitech ECU map to match Stucchi crossover and Mistral mufflers."

 

Via a PC? I couldn't ever get my Scura to idle for shit. MyECU fixed that. I spent years F#%KING with that nonsense. It would seem the range that Guzzi's work with-in is very wide. Some people have no issues, others have never ending. The OEM ECU sucks, or at least its programming does. If you use it as a base, your case is lost before it even began to my experience.

 

Which, the bike ran great for almost 3 years, then it when to hell. Can't imagine that the ECU changed.... But it did.... POS. 

[JBBenson]

Some still-raw nerves there. I hear you. I am starting to feel that way.

 

The ECU was flashed twice by Todd at Guzzitech, I assume via a PC. The bike did run really well when I got it all sorted. But something I did recently is causing some sort of fault.

 

It might be the crap gas we have here in CA, which has up to 15% ethanol, which turns to crystalline sugar with any kind of dry ambient heat. Which we have here in the desert. Or the ECU is a "Monday Model".

 

I am going to revert the head sensor to stock and check the fuel pump, 30 amp fuse, and coil. I am convinced it is electrical or fuel related, or a combination.

 

Almost everything else is new at this point. If it all fails, maybe the ECU is the culprit. 

 

I am tempted to trade it in on a new Stelvio. But it needs to run properly first.....

[moto fugazzi]

Also check the sensor in the airbox, as it's has very thin wires. Did your bike always idle that high? When it drops down, that about where my idle normally is when it's fully warmed up.

IIRC, when the temps sensors go bad, I believe the ECU kind of goes into a default mode, and the bike will run rich. At least that info is according to my trusted MG mechanic.

Ken

[JBBenson]

Also check the sensor in the airbox, as it's has very thin wires. Did your bike always idle that high? When it drops down, that about where my idle normally is when it's fully warmed up.

IIRC, when the temps sensors go bad, I believe the ECU kind of goes into a default mode, and the bike will run rich. At least that info is according to my trusted MG mechanic.

Ken

 

This makes sense. When the idle drops and the engine lugs, there is a smell of unburnt fuel. I will check the airbox sensor too.

[docc]

I unloaded quite a lot in my post above. 

 

When you say,"The brass temperature sensor: This problem really showed up when I installed it after struggling to remove the old crumbly one." - that sounds like a great starting point.

 

Also, it's easy to unplug the TPS and hook up a sensitive Ohm meter and watch the resistance change as you slowly open and close the TPS. If there are any *jumps*, ditch it for a new one. That was the last straw making my Sport jumpy.

 

Otherwise, being able to run pure premium gasoline has made an extraordinary difference as well.

[GuzziMoto]

I also would start with the brass temp sensor holder. Anything that was done to the bike just before the problem surfaced should be especially suspect.

Also, while the brass holder is more robust it likely has different characteristics and may give a different reading or influence the reading in a negative manor. Some Guzzi's use that sensor holder stock, but they may be set up to use the reading the the sensor/brass holder combo gives. Using the same combo with a map not designed for that combo may yield inferior results. You also may have a failing TPS or temp sensor itself.

Best advice I have is be methodical in troubleshooting this and not just reactive.

[motoguzznix]

How much is the CO at idle?

 

Adjust it to 3 - 3,5 % using the CO trim of Guzzitech.

[JBBenson]

How much is the CO at idle?

 

Adjust it to 3 - 3,5 % using the CO trim of Guzzitech.

My 2001 Sport has no Lambda. I don't think I have that option.

 

Servus aus Kalifornien.

[stewgnu]

Also check the sensor in the airbox, as it's has very thin wires. Did your bike always idle that high? When it drops down, that about where my idle normally is when it's fully warmed up.

IIRC, when the temps sensors go bad, I believe the ECU kind of goes into a default mode, and the bike will run rich. At least that info is according to my trusted MG mechanic.

Ken

 

Yeah, I'd second that, almost 2k for idle?  Mine too idles happy around 1100 - 1200.

Will the guzzidiag diagnose rich running?

[docc]

Even without  a Lambda sensor, the CO can be adjusted by interfacing the ECU.

[Kiwi_Roy]

The second sketch in this topic shows how to easily connect to the sensors from the ECU plug,

 

http://www.v11lemans.com/forums/index.php?showtopic=17469&hl=%2Becu+%2Btest+%2Bpoints

 

 

Just a note about testing the TPS with a multimeter on Ohms, It's not quite as simple as I show.

The reading should be smooth but not quite linear there are some other resistors in parallel, I think it's designed to give you a fixed mV if the variable resistance fails.

The TPS test is best done with the ECU powered up measuring the millivolts as you slowly open the throttle.

 

A tap into the wire going from TPS to pin 11 of the ECU plug with a short wire terminated in a blue crimp splice, this makes a good permanent meter jack for use when setting the TPS or troubleshooting in this case.

 

TPS Slider

    |

    |-----new wire------------------Blue crimp splice Multimeter ------> Chassis

    |

ECU pin 11?????

There's some confusion does the slider connect to pin 1 or pin 11, I will check this evening

 

 

It would be interesting to see if anything happens to the TPS Voltage as the idle drops.

[JBBenson]

I think I found the problem.

 

Turns out, when I did the last “re-tune”, the disconnected TPS was WAY out, around 480 mV. How did this happen? 

 

I thought about it: I would set all the variables at the right settings, and the bike would run well but slowly fall “out of tune”, starting with the idle, then progressing to running shitty in general. This has happened several times, leading me to replace parts in a maddening hunt for the problem.

 

But this time, the “way out” TPS got me looking. The hose running off the fuel pressure regulator was rubbing the TPS to the point that the fuel hose was abraded away. 

 

Because of the way the hose is bent, I think it was pressing ever so slightly  against the TPS, and, as the bike was driven, the vibration and pressure moved the TPS ever so slightly. Since we are talking about fractions of a millimeter here, it wouldn’t have to move it very far to push the engine again out-of-tune.

 

I reset everything to norms after removing and cleaning the TPS: 150mV disconnected, both bleeds out 1 turn, 500mV connected. I then balanced the throttles AGAIN using a new Tec Mate. This time, I balanced at idle using the white knob only. Turns out, when the idle is balanced this way, it is also balanced though the RPM range. It was pretty close to balanced anyway.

 

I lifted the tank and tied up the offending fuel hose so it was no longer touching the TPS.

 

Now the bike seems to idle and run better than ever, even in heavy traffic in the heat. I did replace the head temp sensor with the OEM blue plastic one, but the “way out” TPS has me thinking that the creeping TPS was the issue after all. Sometimes problems have really stupid solutions. Lets see if it lasts.

 

Offending fuel hose and zip tie holding it up: