"Tuning" and "too much" form an oxymoron.

 

lol,, ,i was getting same thing and said WTF !... hope to get in more runs this weekend.. wondering how many logs it takes to get it right...

 

I think DJ is working on an AT-100 kit which will include 12mm WB sensors, so the sensors alone should be available, maybe even at a reasonable price. If not, why not just add some 18mm bungs to your head pipe and use both NB and WB together? I believe Fuel Moto's head pipes have both.

 

The warm-up enrichment process only lasts for a short time, around 45 sec. or less with my '07, even if the bike is warm. If you start logging after that point you will not see a value there for the duration of the log. Since I always start my logs well after cranking the bike I never see warm-up values. I think they put that parameter in because if your bike is above the minimum ET, it is cranked, and you start the log immediately thereafter you should see values. That would affect the tune if compensation wasn't made for this.

 

I was combining morning (cooler) and afternoon (hot around here) tuning runs to get better or I should say more consistent results when tuning with the stock O2 sensors. The stock sensor tuning seem to vary more as the result of outside temp. I would do one tuning run and see a cell go up by 4 then use the afternoon log and it would go back down by 3. I combined the runs to get "quicker" better averaged results and it seemed to work well. When I started WB tuning, the outside temp didn't seem to cause as much variance from one tuning run to the next.

As to going back and using the stock O2 sensors - as you know they only work in closed-loop and are supposed to take into account air temp and altitude changes. Well, the ECM already uses Intake Air temp and Manifold air pressure in its calculations so even without the O2 closed loop, the ECM accounts for changes in these variables any way. Honestly, I didn't see any benefit to running closed-loop with the O2 sensors re-installed. Some say that the closed-loop with stock O2 sensors account for change in temp and altitude but once the closed-loop takes control it ignores the air temp and MAP values and locks in on stoich 14.6. The more I study this the more it seems like you should run open loop and let the ECM do it's calculations like it did for all the pre O2 sensor bikes...

 

I haven't looked at MLV seriously, so I'm not sure what the graphs represent. Looking at yours, I assume that any movement from the mean (i.e. VE=New VE) represents a lean or rich spike. Is this correct? IOW, at the top of the column (e.g., New VE=110 and VE=94.7), does that represent a lean spike? I assume you're looking for a tidy 45° angle with most points in the graph hunkered along the mean, while some spikes are going to occur. Is MLV getting the data from the CSV file and plotting New VE from it? If so I guess you're also logging VE.

Moving to a variation of this subject, I know what the New VE table represents in the LT, but in the CSV file I'm not so sure. Does it show an on-the-fly change? If so, what parameters are used in creating this value? I've tried to make sense of it, assuming it is RPM & TP vs. Lambda & Set AFR at a given instant in time, but although it sometimes seems to follow this convention it doesn't always do so. Why is this parameter even included if LT is figuring the New VE?

 

That's I've always understood and I'm glad you've come to these conclusions by testing. I would prefer staying with open-loop and will for the foreseeable future, but will keep a close eye on Lambda values as temps fall and I ride in higher altitudes. I'll have to travel far to get much above sea level, so those tests will come in the long term.

 

Ve new = VE * o2 integrator/100 * AFF/100

If the adaptive fuel is turned off then aff is always 100 and this drops to

Ve new = VE * o2 integrator/100

 

I haven't used it much either but was playing with it to analyze this data. The sharper the line the less the VEs are deviating from the VE set in the map. It doesn't represent a lean or rich spike just shows how close the New VE is to the VE in the map table.

That value is just where the cursor happened on the graph to be when I did the screen print.

Yes MLV is getting the data from the CSV file and plotting it. It is picking up the VE and New VE at the particular RPM and plotting it on the graph. A sharp 45 degree line shows that the VEs in the tables are close to what is actually happening.

Whittle posted how the New VE value is calculated and it is what is in the CSV file and is used by LT to come up with the new weighted values that it generates.

 

Yeah, but if you have a New VE that's VE+20 and another right under it that's VE+2, doesn't that mean that a lean spike has caused the New VE to be +20 richer than the VE?


I don't mean at the cursor necessarily, but any point toward the top of the graph.

Okay, then wouldn't a dot at the top of the graph be a lean spike? If New VE is VE+20 it would mean that it is adding fuel in response to a lean condition. I may be totally off-base, but if so I'm confident I'll get it eventually. Be patient.

I asked him to explain "O2 integrator," as that doesn't compute.