GuzziDiag Stelvio idle AFR issue

[resom]

Hi Guys,

I'm having an issue with the remap of my 2014 Stelvio NTX using GuzziDiag.  I've used GuzziDiag in the past to do a full remap on my Griso 1200SE, so I have quite a bit of experience with it.

About the bike - Low mileage (1100 miles).  Everything is in check (valve clearance, TB balance, etc.).  Agostini Black-X slip-on (no dBK).  No O2 sensors (open loop).

About me - been twisting wrenches on bikes for over 30 years, so I get it.  Software - not so much, when there's an issue.

I recently installed an Agostini Black-X slip-on (no dBK).  After the install I ran the bike using the stock map (closed loop) just to see how it would run.  The bike started normally, but I could tell it was very lean while riding.  I only put a few miles on the bike while in this configuration.

I removed the O2 sensors and installed a new map using GuzziDiag IAW5xWriter V0.21.  I got the map from another member whom I worked with on my Griso, which also has an Agostini slip-on (no dBk).  I modded the map to my liking using TunerPro, making the fueling similar to my Griso.  I knew it wouldn't be perfect, but thought is should be a good starting point, which it turned out to be.  The bike runs great throughout the rev range, BUT there's one problem...... cold starting and idle AFR.

I can NOT get the idle to richen!  I've went from 3600uS in the idle cell (4.6tps, 1200rpm) all the way up to 6500uS!!!  The idle AFR does not change, it reads 15.7 AFR no matter what 'uS' I input into that cell.  I've taken this reading from 2 sensors...  I have an Innovative LC-2 on the left cylinder and an Innovative MTX-L digital gauge on the right cylinder, both of which read the same 15.7 AFR at 1200rpm idle.  As soon as I increase the throttle the bike goes rich to between 12.0-13.0 AFR, which is appropriate for a starting point, but as soon as the bike settles into an idle, it goes right back to 15.7 AFR.

I went thru all the different tables and everything looks correct.  There are no wild "modifiers" in any tables.

I originally thought the "Stepper" wasn't operating properly, but it turns out it is.  I ran the GuzziDiag test and it cycled on and off properly.  I can also hear it cycle after shutting the key off.

I'm now thinking there's something in the .xdf (GuzziDiag V1.29 Two Lambda) that's not allowing me to change a modifier that's changing the TPS 4.6 column...?

I have tried everything I can think of, including running the dBK, which only increased the AFR by 0.3....checked for vacuum leaks, etc.  Which wouldn't make sense anyway because the bike starts and idles fine with the stock map and O2 sensors connected. 

I did remember to un-check the O2 sensors to change the new map to open-loop, so that aspect of the map is correct.

Any experts out there that may have an idea what's up...?

I emailed Christian at GuzziDiag to see if he has any ideas, but haven't heard back yet.

Thanks

[Wayne Orwig]

I can NOT get the idle to richen!  I've went from 3600uS in the idle cell (4.6tps, 1200rpm) all the way up to 6500uS!!!  The idle AFR does not change, it reads 15.7 AFR no matter what 'uS' I input into that cell.

Once the bike has run for a time, does this wild idle cell show up?
I was thinking the "warm-up" section is used near idle for a minute, after startup. (this has nothing to do with the lambda warmup)

[resom]

The bike stays lean from about 90 seconds on after startup.  When the bike starts it's at 13.5 AFR, then gradually goes up to 15.7-15.8 AFR, so the warm up table is working properly.

When the bike is idling when warm the ECU is only inputting approximately 2700uS at 4.6tps and 1200rpm.  I got this number from ScanM5x.  My map is telling it to input 6500uS.  I've tried from 3600uS to 6500uS.

Again, all my tables don't have any heavy modifiers that would kick my max. setting of 6500uS down to 2700uS.  I'm thinking something is buried in the map that I can't see...?  Maybe the XDF file (V1.29) on GuzziDiag isn't displaying all the tables..?

[pauldaytona]

Hi Resom,
 funny, all is exact working as designed.

let me explain. The ecu has a number of operating modi. When you start the bike, it knows that, and the first time it adds the warmup values to the mix. The same way it knows the when it is idling of having max throttle.
Now iddle does have one map, you found it, the idle rpm at a temperature. So as long as the engine knows it is in idle status, it does one thing, it tries to rum the rpm that belong to the actual temperature. It does that with changing injection time, the bypass valve and also ignition.  If you let it idle with guzzidiag connected you can see it happen. 

The moment you blib the throttle, it looks at the fuel table because the idle is not needed anymore.

 When bike is stock, the lambdas make it have the right CO amount, but lambdas off you need to set it yourself, you can adjust co for that. And because you changed the exhaust, it will need more fuel because less backpressure. Try what happens with Co at +5 or +10

[resom]

When bike is stock, the lambdas make it have the right CO amount, but lambdas off you need to set it yourself, you can adjust co for that. And because you changed the exhaust, it will need more fuel because less backpressure. Try what happens with Co at +5 or +10

Hi Paul, and how exactly do I change the CO?

I have already doubled the fuel input at idle using TunerPro.  I changed the idle cell at TPS 4.6 and 1200rpm from 3000uS to 6500uS.  No change in AFR.

Resom

[pauldaytona]

Hi Paul, and how exactly do I change the CO?

I have already doubled the fuel input at idle using TunerPro.  I changed the idle cell at TPS 4.6 and 1200rpm from 3000uS to 6500uS.  No change in AFR.

Resom

 you didn't read/ understood what I wrote, you can't set the idle that way.

With guzzidiag, same page where you do tps reset, you find the co set. needs warm and running engine

[Wayne Orwig]

you didn't read/ understood what I wrote, you can't set the idle that way.

With guzzidiag, same page where you do tps reset, you find the co set. needs warm and running engine

I have to say that I don't get it either.

The ECU know it is at idle. The ECU knows that it must tweak the CO trim, and tweak the injector time, to get the proper idle speed and mix. But I still don't understand where the 'base' value is, that the ECU tweaks?

[Xlratr]

I had the same experience when I started mapping my Stelvio. Even the most extreme adjustments to the fuel and delta maps at the idle point had virtually no effect on AFR. On top of that, the right cylinder was leaner than left with Lambdas off. I don’t know to what extent the values are ignored or compensated for via the stepper motors or ignition, but I found it impossible to fine tune idle AFR with just the fuel map.

As Paul said, the CO trim on the other hand “will” let you adjust the mixture. I agree this is strange, because basically you’re also just adjusting the amount of fuel, but for whatever reason, this is a change the ECU can “accept”. Unfortunately the CO trim will also affect the AFR at higher engine speeds (although to a lesser extent as RPMs rise). My solution was to hit my target idle AFR with the CO trim and then adjust the rest of the map to fit. That’s why all the maps I made for my bike all require a +10 CO Trim. This is only a practical solution if you are able to log the AFR on both cylinders otherwise changes to fuel and delta values are too hit and miss.

Now, to make matters more mysterious, I discovered that after I had the roller conversion done (adjusted mapping required), the bike has now actually become responsive to changes in the fuel cells at idle. With the flat tappets it was impossible to get an identical AFR on both cylinders with Lambdas off. Now I can fine tune it with the fuel and delta maps and make them the same. I have no explanation for why that is now possible. And even stranger because yours is a 2014 which should have rollers and should therefore also be the same. Sometimes I think that each bike is a little bit individual. 

At the end of the day, getting the AFR to where you want it at different engine loads/speeds will be a mix of CO Trim and fuel/delta adjustments.
John   

[resom]

I had the same experience when I started mapping my Stelvio. Even the most extreme adjustments to the fuel and delta maps at the idle point had virtually no effect on AFR. On top of that, the right cylinder was leaner than left with Lambdas off. I don�t know to what extent the values are ignored or compensated for via the stepper motors or ignition, but I found it impossible to fine tune idle AFR with just the fuel map.

As Paul said, the CO trim on the other hand �will� let you adjust the mixture. I agree this is strange, because basically you�re also just adjusting the amount of fuel, but for whatever reason, this is a change the ECU can �accept�. Unfortunately the CO trim will also affect the AFR at higher engine speeds (although to a lesser extent as RPMs rise). My solution was to hit my target idle AFR with the CO trim and then adjust the rest of the map to fit. That�s why all the maps I made for my bike all require a +10 CO Trim. This is only a practical solution if you are able to log the AFR on both cylinders otherwise changes to fuel and delta values are too hit and miss.

Now, to make matters more mysterious, I discovered that after I had the roller conversion done (adjusted mapping required), the bike has now actually become responsive to changes in the fuel cells at idle. With the flat tappets it was impossible to get an identical AFR on both cylinders with Lambdas off. Now I can fine tune it with the fuel and delta maps and make them the same. I have no explanation for why that is now possible. And even stranger because yours is a 2014 which should have rollers and should therefore also be the same. Sometimes I think that each bike is a little bit individual. 

At the end of the day, getting the AFR to where you want it at different engine loads/speeds will be a mix of CO Trim and fuel/delta adjustments.
John

Hi John,

Ahhhhh, great, now I understand.  The "Trimmer" needs adjustment.  Got it!  On my Griso I did mess with the Trimmer a bit, but ended up at "zero" because I was able to adjust everything using the Fuel Table cells.

Thanks John for letting me know you had the issue also, and your explanation is perfect as far as setting the idle first, then setting the rest of the cells because they're affected too.  I get it.  Great explanation. 

If you don't mind me asking, what base map are you using?  Stock or something else?  Would you mind sharing thru email?

If Paul would have included the word "Trimmer" it would have clicked with me.  Probably just a translation thing.  Thanks Paul for your explanation.  I totally forgot about the CO "Trimmer".

So, the CO Trim can be set "live", with the bike running, and you can see the AFR change?  I forget, its been so long since I tuned the Griso.

Resom

[Xlratr]

My "base" map is the original 2 lambda map, but I have made quite a lot of adjustments to main fuel, delta, ignition and warm up, with slight modifications to the phase, and as I mentioned, it's designed to work on +10 CO Trim. Sure, you're welcome to try it if you want. PM me with your email.

Yes the CO trim is adjusted with the engine running and if you have an AFR readout you can see the change. It needs a little while to settle down though. The engine MUST be warm (above 60 deg. Celsius I think) and you can only do it if a map is installed with the lambdas switched off.

John

[resom]

My "base" map is the original 2 lambda map, but I have made quite a lot of adjustments to main fuel, delta, ignition and warm up, with slight modifications to the phase, and as I mentioned, it's designed to work on +10 CO Trim. Sure, you're welcome to try it if you want. PM me with your email.

Yes the CO trim is adjusted with the engine running and if you have an AFR readout you can see the change. It needs a little while to settle down though. The engine MUST be warm (above 60 deg. Celsius I think) and you can only do it if a map is installed with the lambdas switched off.

John

Cool, thanks John, yes, I would like to take a look.  I will PM you.

OK, cool.  Yes I have an Innovative LC-2 (bluetooth) and MTX-L I can monitor for AFR.  I can't remember, is there a "save" button after setting the trimmer to say +10 or whatever it ends up being?

Resom

[resom]

Hi Resom,
 funny, all is exact working as designed.

let me explain. The ecu has a number of operating modi. When you start the bike, it knows that, and the first time it adds the warmup values to the mix. The same way it knows the when it is idling of having max throttle.
Now iddle does have one map, you found it, the idle rpm at a temperature. So as long as the engine knows it is in idle status, it does one thing, it tries to rum the rpm that belong to the actual temperature. It does that with changing injection time, the bypass valve and also ignition.  If you let it idle with guzzidiag connected you can see it happen. 

The moment you blib the throttle, it looks at the fuel table because the idle is not needed anymore.

 When bike is stock, the lambdas make it have the right CO amount, but lambdas off you need to set it yourself, you can adjust co for that. And because you changed the exhaust, it will need more fuel because less backpressure. Try what happens with Co at +5 or +10

Paul, this worked great, thank you for letting me know about this solution.

There's still one problem though.....  Because the idle is controlled by the CO Trimmer I can't adjust "Delta RH" fueling.  I need to add fuel to the RH cylinder at idle.  Do you know how I can do this?

I've tried adjusting the idle fuel cel (4.6/1200rpm) using TunerPro from 2000uS all the way up to 4000uS, but it has NO affect on AFR.  The only thing that changes AFR at idle is the CO Trimmer, but this controls LH and RH exactly the same, there is no "Delta" when using the CO Trimmer.  The rest of the map is controlled without problem using TunerPro.  It's very strange that the idle circuit is not affected by the mapping...!!!!  I don't get it....?

[pauldaytona]

If left and right are balanced, there should not be much difference. But I suppose you are using a 2 lamba lm2?  Try what happens if you change the lambdasensors with each other. I tried also with two lambdas and made me crazy. The were not giving same values back. 

[resom]

If left and right are balanced, there should not be much difference. But I suppose you are using a 2 lamba lm2?  Try what happens if you change the lambdasensors with each other. I tried also with two lambdas and made me crazy. The were not giving same values back.

I already swapped the LAMBDA's, they're accurate. 

I just spent an hour balancing the Throttle Bodies also.

Did a compression check too.  RH = 200psi, LH = 190psi, this was taken with the motor hot.  So the RH cylinder is a little better.

The RH cylinder does act a little strange.  At cold startup it's exactly the same as the LH cylinder, reading 12.5 AFR, but when the engine is hot the RH cylinder goes lean.  When hot the LH cylinder reads 13.2 AFR and the RH cylinder reads 13.8 AFR, this is at @1200rpm

Thoughts?

[Meinolf]

Hi,

I have been researching the 15M and found that it has 2 air temperature trim tables, for the left and right cylindre. In the 15M BINs the values are different. This led to the deviations you are observing.

I don't know if this applies to your ECU as well, though.

Cheers
Meinolf

[resom]

Hi,

I have been researching the 15M and found that it has 2 air temperature trim tables, for the left and right cylindre. In the 15M BINs the values are different. This led to the deviations you are observing.

I don't know if this applies to your ECU as well, though.

Cheers
Meinolf

Interesting.   I've not sure if the 5AM has the same tables available within the .bin...?  I think the .XDF file controls what we can see within the .bin, but I'm not sure how to find a particular table within the .XDF...?  I've heard there's a lot more data within a .bin that can be displayed, but most users choose not to dig that deep into a .bin.

Do you know how to search for the different tables contained within a .bin?  And add that table to the .XDF file?

Cheers!
Resom

[beetle]

I'll send you a how-to, Mike.

I'm wondering if it might be a mechanical thing. How are your valve clearances? We know that the roller tappets clearances open up slightly as they get hot. We also know the 8V is very sensitive to such things.
The RH clearances might be opening up at a different rate. Re-check your valves.

[Meinolf]

Hi Resom,

Interesting.   I've not sure if the 5AM has the same tables available within the .bin...?  I think the .XDF file controls what we can see within the .bin, but I'm not sure how to find a particular table within the .XDF...?  I've heard there's a lot more data within a .bin that can be displayed, but most users choose not to dig that deep into a .bin.

Do you know how to search for the different tables contained within a .bin?  And add that table to the .XDF file?

yes, I do know how to search for tables and scalars within a BIN, but I have the advantage of having the dissassembled code of the 15M/RC, which simplifies it a lot. Adding a table to Tunerpro is not difficult, it's well documented in the Tunerpro help file. The time-consuming part is the verification of the function of a table or scalar. I have built a test bed, using the 15M, for that purpose. See here http://wildguzzi.com/forum/index.php?topic=67653.0, it has evolved over time. I share my findings with Paul, Beard and several others and publish mostly in the German Guzzi-Forum.

I am sure there's plenty of knowledge about the 5AM spread around. Why don't you start a community effort to gather this knowledge and enhance the XDFs? It's worthwhile, as just tweaking the fuel maps is only scratching at the surface.

Cheers
Meinolf

[resom]

I'll send you a how-to, Mike.

I'm wondering if it might be a mechanical thing. How are your valve clearances? We know that the roller tappets clearances open up slightly as they get hot. We also know the 8V is very sensitive to such things.
The RH clearances might be opening up at a different rate. Re-check your valves.

Hi Beetle,

Yeah, everything is in check.  I always check everything (valves, sync, compression) before digging deep into the mapping.  I just double checked the sync with a different gauge and it was perfect.  The only thing that's off is compression in the RH cylinder is about 10psi stronger (engine hot).  190psi vs 200psi.

I will say that the AFR is VERY sensitive at idle.  If I turn the RH bleed screw 1/10 of a turn clockwise the AFR comes in as equal @13.0 AFR (This may seem rich to some, but with the open Agostini I'm running the bike seems to like a little richer AFR at idle).  If I turn the RH bleed screw to where the balance is perfect (1/10 turn counterclockwise) the AFR is thrown out by almost 1.0 AFR (13.0 LH, 14.0 RH)!  The balance @ 4000rpm remains perfect with the RH bleed screw in either position.

Resom