MSefi.com read only forum (Jan.15,2018)

I read in the manual that normal AE isnt disabled when x-tau is enabled. Is it necessary to have both on and tune both or should the normal AE be disabled by setting 0ms as enrichment or perhaps a high threshold and leave all the acceleration enrichment to X-Tau?

To me it sounds easier just having one type of AE active with less risk of messing everything up.

Any thoughts on this?

X-tau is a bit more difficult to understand intuitively what values to put in. We have a feel for what a 10% increase in pulsewidth will due, but no feel for what an increase in tau of 20 ms will do. So if your car needs a lot of accel enrichment, it might be easier to back off say 30 % on you existing AE and make it up with x-tau, then continue to phase out the existing AE and replace it with x-tau. But sometimes people find that a combo works best for them.

Another consideration fo rx-tau. On the beta list we have a greatly revamped x-tau code which people seem to like much better and find easier to tune. So you might want to consider going to that code before getting deep into x-tau.

Read a bit regarding the beta code. Does the 2.870 'released beta' contain the synced MAP sampling and changed x-tau or is it in later versions as 2.871t3?

Did some more and downloaded the ini file and answered the q myself. Im going to flash 2.871t3 and try to have a go w that.

Downloaded the 2.871t3 betacode in to the MS earlier today. After some adjustment of the x-tau parameters it runs very sweet. Accel enrichment was very easy to dial in to get a consistent AFR through the entire transient. Very good. I havent played that much w the deceleration parameters yet so there is some lean running after I release the throttle. Normal AE is zeroed out. One strange thing is that the x-tau seems to oscillate some times when im running on very light load an close to closed throttle. Might solve it by changing the treshold for MAPdot. Time will tell.

The synced sampling and possibility to offset the start of injection definitely increased the running quality at idle and low PW runnning. I set the offseted the injection start by 50% to ensure I didnt inject during overlap which helped.

Looking forward to the implementation of tempcompensation in x-tau.

These features should definitely make it to the next relase of the code.

I have noticed the oscillation in xtau correction between 99-100%, but I don't know how to get rid of it other that average it, which I hesistate to do because it might slow things down. As best I can tell from a simulation, this oscillation is a result of integer arithmetic. There is fluctuation in PW from many things like map fluctuation. If you put a lower lag factor on map (more smoothing on it) that might lessen x-tau. But I wouldn't want to do this unless it was possible to feel the oscillation and/or it was causing a similar high fluctuation in AFR. A 1 or 2% fluctuation in xtau correction shouldn't cause more than a 1-2 % change in AFR, which is hard to detect in a responsive afr.