RomRaider

Sure. Just let me know what you would like me to do/log.

Thanks for the explanation. I guess I need to make sure I tune without blinders on and concentrate on the big picture.

Messed with this some more. Although no testing yet to verify on Casey's car, I believe that the minimum throttle enrichment is causing the early transition to OL. It is very similar to the 32-bit ECU, so the below also generally applies to both ECUs:

1. The "Minimum Active Primary Open Loop Enrichment" and the "Primary Open Loop Fueling" map(s) would typically determine the CL to OL transition as explained in the original post.
2. However, the "Minimum Primary Open Loop Enrichment (Throttle)"(not currently defined for the 16-bit ECU) is applied after the fuel map enrichment value above is determined. And then the "Primary Open Loop Fueling Compensation (Coolant Temp)" is applied. This final value is what is used in the decision process.
3. For an 02 WRX, the "Minimum Primary Open Loop Enrichment (Throttle)" has an effective AFR of 12.97:1 at about 89% TPS and above (due to interpolation it would ramp richer than stoich from above 70% TPS with a separate minimum enrichment of about 13.5:1).

So, for example, even though Casey called for no enrichment in that load/rpm range of the fuel map (14.7:1), the minimum enrichment based on TPS was applied when hit the throttle (the minimum TPS based enrichment is determined AFTER the primary fuel map value). Because the final decision includes this final value -> if no enrichment, then remain in closed loop, any enrichment switch to open loop, which is why you see the unexpected switch to OL. I'll be adding this table to the next ecu defs for the 16-bit ECU.

It was mentioned in another thread, that if you are not sure about the accuracy of the O2 sensor (thinking specifically of those of us that tune older vehicles), that setting all the delay values to zero might not be ideal.

In this instance, would it be better to rely on just one table (e.g. ' Closed Loop RPM') to determine when to transition to Open Loop. If so, does anyone know which delay value to leave (and zero the rest), so that in effect only Closed Loop RPM and Minimum Active Primary Open Loop Enrichment values are used to determine whether the car enters Open Loop or not?

You need to re-read the original post. The primary fuel map enrichment will always be the final key in the transition, regardless of whether there is a delay or not. The RPM table is one of those that clears the current delay. The delay is really the time "period" that the primary CL to OL with delay TPS or load thresholds have to be exceeded and THEN the primary enrichment value is checked to determine if the transition from cl to ol takes place. If the delay is zero, then only the primary fuel map comes into play. If it is non-zero, then the primary cl to ol with delay tps and load thresholds are used. If the delay in your map is non-zero, you use the auxillary tables (RPM, veh. speed, etc.) to control when the primary TPS and load thresholds are ignored and simply skip to the primary fuel map check. That is, they are extremes where you would want no required load TPS threshold to be exceeded in order to switch from cl to ol (only a fuel map enrichment value to be checked).

Great. Thanks for looking into this.

The original explanation has been updated for clarity and to add some additional info (such as the minimum throttle enrichment).

Awesome.. I was wondering about this as well with my car.... thanks for explaining why this happens....

02 WRX. I basically wanted the car to go into OL soley based on the Open loop fuel map so I raised all of my throttle, rpm, and load ranges...it will still flip into OL when it hits 100% throttle....even though I made the cut off 102%. maybe you already know this.

my delay table has been zeroed, but it still appears that the other tables are in play. I will try zeros across the board and see what happens. after reading the first post again...I think this is what you said will happen?

Also, I log on a daily basis, so if you have some new defs you want to try for an 02 A4SGE01C, I'll do it.

The other tables are not in play if the entire CL to OL Delay table is zeroed. Post logs including cl\ol status, load, and rpm and post up your rom (as long as it is not a commercial rom). Remember that there is a minimum throttle enrichment (not currently defined for the 16-bit ECU) that will cause you to transition to OL with heavy throttle regardless of the fuel map (as explained in the OP that I recently updated). I'm going to add this table for the 16-bit ECU to the next definition update.

so I forgot to hit "log" this morning, but with everything zeroed, it worked the way I wanted it, but you are right, when I go to 100% throttle, it still switches to OL. this may have been my issue all along, so I will have to go back to a previous map and try again without going to 100%. if I find something to the contrary, I'll post the logs.

The other day my wastegate came off because I lost the little clip to hold it on(long story). The maximum load is basically 1.0 g/s untill you reach the higher rpm and it will start to rise. regardless, my OL AFR table calls for no enrichment all the way out to 1.34g/s. When going WOT, it does just as described above...it will max the AFR to about 13.5 and not match the OL map unless it requires something less than 13.5. At least it didn't knock, so I know I have some head room on the rich side(no fuel knock)

Forgive me if this should be obvious, but I drive an 08 LGT 5MT. In my rom the CL to OL Delay table is a single cell set to 750. Is there more logic involved in the CL to OL transition than previous years? Or does zeroing this single cell have the same effect of allowing the primary fuel map enrichment table to be the only determining factor in going to OL?

Yes, that is the primary delay, so setting that table to zero will have that effect. You should also have another delay based on SI-Drive intelligent that if non-zero will use that delay in intelligent mode instead of the primary delay. That is something you can experiment with if you want to achieve better gas mileage in intelligent mode.

Had a similar question about the 08 sti.
I believe the rom was stock unmodified and the delay was set to 0.

would this mean that subaru does not have an openloop closed loop delay on the 08 sti's? it just follows the fuel enrichment values.