Lean spike in middle of AE pulse

[dggt6]

Hi Guys,
seems it is my turn to ask about a "lean spike".
I think I have my tune sorted and have been adjusting the AE all over the place and have finally found some settings that allow the car to drive/accelerate well.
However the log shows a short lean spike in the middle of the AE pulse regardless of how long or short I make the pulse. I have tried every setting from 0.1 to 1.0 seconds but I still get a similar spike. Usually the AFR goes rich after the spike and any adjustment to lean the AFR only seems to increase the spike!
I have attached a few photos of the log graphs with the spikes circled in blue. (The pink circle shows a small lean "bump" for a similar TPSdot figure that previously gave a large lean spike!!)
Any thoughts on what these spikes mean would be appreciated. (Do I need to adjust lag time in something? Is this where EAE comes in??)

Many thanks,
Regards,
Doug

[nathanhardy]

A lot of the time ae will just confuse you if you tune by data logging it's much more how the car drives and how the afr's transition then that of what a data log might show if it runs well I'd leave it since most cars can have quite a broad range when tuning ae. As long as it isn't a rich miss fire.

[slow_hemi6]

One of the best things I ever did early on was put my well tuned carby on my car with MS installed.
I did datalogs with it and thanks to that I got to see how poorly the carby (that drove great,didn't dirty the plugs and tailpipe showed no signs of over rich) maintained mixtures under acceleration. Always a compromise on acceleration of going over rich to a bit lean to settling into the main circuits.
That showed me that you want a fairly decent initial hit of extra PW that then tapers away. You seem to be going for a given amount that stays the same for it's duration then very little taper. Still a slight lean for less than 0.3 of second is not the end of the world either.

[2swe]

I think the lean spikes are not in the middle of AE pulse, but in the beginning, don't forget AFR signal is about half a second retarded. depends on RPM, Load and distance between exhaust valves and sensor..

[Yves]

I think, like SWE, that it's because of the delay in exhaust pulses. You can get the AE to come in sooner if you either use map ae to some extent or use the WOT function in AE.

[kaeman]

if the lean spike occurs when your pw is large then its most likely that you are having a rich misfire, I used to have that then realized that it was a misfire and reduced the fueling in that area of the table and no more lean spike. you might be getting a little to large ae pw. reduce the amount of fuel in the ae settings and see if that makes the lean spot go away.
Good Luck

[muythaibxr]

The ones talking about "delay" have it right.

[dggt6]

Many thanks for all your responses.
Following the suggestion that "delay" is what I am seeing, am I correct in assuming that the lean spikes during the AE period is what the AE is designed to reduce and not a "problem" as such? It just takes "that" long for the lean exhaust gases to travel down to the O2 sensor?
Is there a setting somewhere to adjust this "delay/lag"?
I had been assuming that as the graphs of TPSdot and the Acel PW were aligned (kicked in at the same time) the AFR should start to rise at the same time and therefore my lean spikes were a "lack of fuel" during the AE time period.

So, if I were to "pick up" the AFR graph line and move it backwards so that the lean spikes were inline with the beginning of the rise in the AFR and TPSdot graphs, THEN adjust/tune the AE TPSdot curve to remove the AFR spike. (with much shorter periods of AE.)

Many thanks again for all your continuing help,
Regards,
Doug

[muythaibxr]

MS2? MS3? Sequential fuel? If not, squirts per cycle? Lag factors? AE trigger thresholds?

[slow_hemi6]

You can use decell fuel cut to gauge your delay. The log will show the pulsewidth drop out and compare to when the O2 shows lean. Obviously there is more delay when pottering about then when you are up in the rev range.