When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I have a new issue where, most times, during warmup, typically with CLT between 100 and 135*F; EAE will suddenly drop to zero, or nearly so. Oscillate for a few cycles, then be fine for the duration of the trip.
Car is a turboed 1999 with MS3 Basic running 1.4.1 FW.
No indication of sync loss or spark loss. EAE goes to zero, and so does injection pulse.
I cannot find a root cause. What I have done:
1) With engine running, turned off EAE and back on. That SEEMED to fix the issue.
2) Reloaded the FW. (in case there was some corruption).
3) Made sure fueling was fat (to prevent any AFR shutdown). I also added an idiot light for AFR warning (status6 = 4)
Here is a screen shot: I have also attached the log and tune.
2 warm-ups without issue, but the third, I had the problem again. (That third time, I was not on SD card).
Also, I noted that my EGO gets stuck around 97% at times, even when AFR is lean of target. That I fixed by going away from the AFR Authority Table... not my preferred solution.
I realized that when I re-loaded FW, I checked the "Preserve Sensor Calibrations and Tune Data". Really, for a full re-build, I should un-check that, re-calibrate sensors, and load an old tune. From there I can systematically adjust to my preferred tune.
I'm thinking that if I am going to do a full re-load, I should go ahead to 1.5.X FW (presently on 1.4.1). @Ted75zcar , which version do you recommend?
Bottom line... 1.5.2 is working pretty well. Sure goes to idle quickly. I'll have to fine tune that. EGO is working quickly as well... and properly.
EAE mad oscillation still occurred. Looking at the combination of my temp correction and RPM correction (both of Added to Walls), I surmise that the ATW at certain points, combined with where the Wall Fuel happens to be is just too much.
So I have shaved the peaks off both of those curves and will test / tune until things are reasonable again.
I had used all 3 of the loops in 1.4.1, so I like that I have 5 more, even though I don't presently have a use case.
did you ever catch the event with high speed logging? It looks to me like your TPS is wonky on the TS log above. Time based accel pulses add to wall fuel and can do wonky things, but they also can occur at faster than the 50msec you get with TS logging.
I think I would get the event captured with high speed before really trying to fix it.
I saw your post on msextra where you said you turned off TPS WOT. That feature is awesome, so I would highly recommend going back to it when you get this EAE business figured out.
Ted, when the car is lurching, the right foot does strange things.
I pulled the WOT a it was too extreme as I implemented it. I was getting EGO resets while idling as one LSD of the ADC was enough to trigger that action.
LOG714. At 74 seconds. Happening big time. TPSdot is removed, but EAE curves were copied straight over from the 1.4.1 tune.
LOG715. This is with some mild TPSdot AE. At 78 seconds you can see where the Oscillation is occurring, but not as violently as in other captures. after 83 seconds, goes to normal. I'm pretty sure this corresponds to tune "Shaved", after I reduced the ATW multipliers for RPM and Coolant in the suspect areas, and did return mild TPSdot AE.
for EAE, your problem is clear as day, I should have picked up on this immediately. The total added can never exceed 100%.
The code gets really wonky if you let this happen (same for sucked, but that is 25.5% IIRC). For reference, my base added maximum is 54%. So at the wonky point (these are approx) you have 80% * 1.1 (rpm) * 1.1 (CLT) which is pretty much 100%, especially since the code rounds numbers and uses what looks like a form of fixed point integer math. You can fix this by reducing any of the added curves, or do as I did and scale them all. Or more correctly my EAE tuner software prevents this from occurring.
My advice would be to simply reduce the CLT modifier if it is otherwise working the way you want. That is such a transient operating point for CLT that you are talking about probably less than a minute of less than favorable AE.
For TPS WOT, the actual curve is relative. My approach is to tune the point of greatest interest (presumably off-idle tip-in or near off-idle tip-in) with a flat WOT curve of some low-ish number that doesn't result in noise based events. For me that is around 25% with a 1:4 TPS digital filter setting (also 25). Tune the entire time based curves at that point. You then tune other operating points almost exclusively through the TPS WOT curve. For less AE, increase TPS WOT, for more AE decrease TPS WOT. My TPS dot threshold is 20 and I have MAP/TPS right at 50-50 even though I stopped using MAP. I may go back to it, and I don't want to re-tune TPSdot because I moved the slider. The 50-50 setting means that I am getting half of the adder that you see in the curve.
for EAE, your problem is clear as day, I should have picked up on this immediately. The total added can never exceed 100%.
The code gets really wonky if you let this happen (same for sucked, but that is 25.5% IIRC). For reference, my base added maximum is 54%. So at the wonky point (these are approx) you have 80% * 1.1 (rpm) * 1.1 (CLT) which is pretty much 100%, especially since the code rounds numbers and uses what looks like a form of fixed point integer math. You can fix this by reducing any of the added curves, or do as I did and scale them all. Or more correctly my EAE tuner software prevents this from occurring.
My advise would be to simply reduce the CLT modifier if it is otherwise working the way you want. That is such a transient operating point for CLT that you are talking about probably less than a minute of less than favorable AE.
That is exactly the response I was looking for. Thank you. I have incrementally increased each factor until I went too far. I figured there was a FW or HW limit that I was exceeding.
EDIT: I always thought that the Manual should give some feel for what those limits would be.
DNM
Last edited by DNMakinson; 01-11-2023 at 04:22 PM.
01-07-2023, 07:48 PM
Warm-up EAE wild behavior
0
