Anything that alters the MAF signal can throw off the trims at idle.. If they read much lower, or higher than what the ECU expects will throw it off..

I used to see it pretty often when I first installed my blowthrough setup (because I had the ability to adjust the setting manually and could see that I could trigger it)

if your logging data, watch the conditions under which it happens.. Typically you'll find the trims shift +/- a large amount, then when it resets, the ECU drops to open loop, and a few moments later, the O2 feedback starts compensating again.

I've seen this happen on almost every car that has aftermarket intake parts and has not been tuned.

 

in closed loop the car has an ignition and fuel map it follows also right? is there a reason why the person that made ECUflash didnt add them, or has it not been found yet?

 

Har far beyond stoich should we set the load envelope of closed loop ?

 

they are the maps in ecuflash, they are just being corrected in closed loop

 

I found that the enrich/enlean map affects closed loop greatly. I think it should be renamed Closed loop fuel map.

 

well the high and low ignition and fuel maps are for open loop. There isnt an actual table with load and RPM for closed loop, and what im asking is, has it not been found or does the car use a different method to determine AFR and Timing in closed loop?

 

The ignition maps are for closed and open loop. Only the fuel maps are for open loop.

During closed-loop fueling the ECU uses O2 sensor feedback to keep the burn at a stoich 14.7:1. There may be a table to change this target of 14.7:1, but I don't know if anyone has found it and, even if found, a narrowband o2 sensor only has good resolution around .5 V, which is stoich. The further you get from .5V, the more innacurate the sensor gets. So, what I am saying is that even if we find the table, we may have a difficult time to target a closed-loop AFR too far from 14.7.


Eric

 

14.0-15.0 is the typical range the narrowband swings back and forth, optimum is 14.7, Its unlikely running any leaner than 14.7 will net you any gains for the sacrifices (misfire, knock, etc)

There is a reason why closed loop is set up that way. A member did find the closed loop target table, unfortunately he chose not to share that map with the class, therefore its not worth mentioning. In reality, the closed loop targets arent all that valuable for altering, the only real value it has is to alter the closed loop values at idle to richen the target up a bit if you have trouble..

There are just so many undocumented tables at this point.. I think there are 35 or so that I've found so far, most of which I cannot identify simply because it involves tracing every subroutine it references which I do not have the time to do.

 

3ff5 is the offset for what I believe is the closed loop afr in the 94170008 rom

 

That definitely looks like the right table... I just noticed in the image I posted, I forgot to check off the swap x-y axis checkbox, so the map is skewed..



It certainly looks correct, though I'm not 100% sure the load axis is the correct one..

 

Its interesting, occasionally when my car is idling it swings to about 16.0 or so under load (A/C compressor or whatnot) this can explain my mysterious lean condition.. I never connected the two before..

 

thats funny that it targets leaner conditions in that little area

 

closed loop takes the value in the fuel table and corrects it, the table above is what it trys to correct it to from the fuel table, the timing table is the timing table, this is what i have been saying this whole time
without the open loop fuel table to look at and correct the closed loop fueling would never be able to keep up

 

Wait so closed/ open loop both follow the high and low ignition map? And that AFR map for closed loop AFR, is that not added to Ecuflash?

 

Yep.. The fuel trims are the % of correction required to hit the closed loop target from the open loop fuel map..

Timing is referenced from the timing maps, but I do think there is a bit of fuzzy logic involved in the selection of timing in closed loop.. I see that at idle somewhat.. though it could just be because its shifting between load sites.