AE Challenged

[dragbug]

I am trying to understand AE so I can tune it a little better. The way I understand it, when looking at the graph, the horizontal area is how quick you press the accelerator pedal and the vertical area is how much extra fuel is added...is that correct? If that is correct, when I am looking at a log and the TPSdot shows a blip to 800 and goes real lean right after that point, then the point on the AE graph at or around 800 should have more fuel added...correct? If both of those are correct, what do I do if I have a TPSdot blip in the log that goes to 1700+ but the AE graph doesn't go beyond 1200? Same question applies for added fuel. The max the graph shows is less then what I think needs added.

If my understanding is flawed, could someone explain it so a 4th grader can understand please lol?

[elaw]

It sounds like you understand the basics pretty well! You're right about TSPdot and the graph... assuming you're using 100% TPS-based AE. If that's not the case then MAPdot will be a factor too.

You can change the scales on the axes of the graphs in TunerStudio by clicking the little button marked "..." at the upper-left of the graph. At least on MS2/extra firmware, it looks like the TPSdot values can go up to 2000.

We could probably provide further help if you upload a log and a copy of your tune file, as there are some other parameters like accel time that might need to be considered.

[dragbug]

Yes I am using 100% TPS based. I will check for that button when I get home this afternoon. Once I get the scale corrected I will make adjustments and if I am still struggling I will certainly post my log and tune. Thank you for the help!

[ol boy]

I've found with the SBFord that you may need to lengthen the Accel TIme, and not add as much PWM amplitude. Changing the "Accel Time (s)" from 0.2 to 0.3 or 0.4 will help cover a larger/longer hole during accel conditions. Adding more time to the "Accel Taper Time" will help with the small dip in AFR near the end of tps/map movement. AE is a pain and takes some time to wrap your brain around. Some might say that having the Accel time higher than 0.2 isn't necessary but if you've ever played with AEA you see it adding large percentage of fuel for quite a bit of time while keeping a stable AFR throughout most driving conditions.

If Accel Time is too short you will begin the see the O2 dip rich during an accel event then spike lean as the PWM was too big but the duration was too short. I'm not sure a fixed Accel Time value is the right way of creating an usable/repeatable AE set up but it's what we have to work with.

Later Ryan

[dragbug]

Thanks Ryan. I already changed accel time (I think) to .3 based on someone else's suggestion. I will have to look at my taper time to verify but I think it is defaulted to 0..?

As long as I have some attention, when reviewing my logs, TPSdot bounces ever so slightly above and below 0 at idle. I mean very little. Is that normal or would that be considered a false trigger? My TPSdot Threshold is set at 50. TPS signal is rock solid.

[ol boy]

That's normal for a small amount of noise on the tps line. As long as the noise doesn't swing past your threshold setting your fine.

[dragbug]

Cool, then I won't worry about it.

So when setting up AE for small, medium, medium high and high pedal, what actually dictates what those values should be? Should low throttle start at 50, 100 or the default 200? Or is it really just up to me to decide randomly? Based on my logs, I have seen TPSdot as high as 1750 so I set the max to 1800 and the low to 100. The middle numbers are 600 and 1000

[ol boy]

That would be a good start. set the first pin at 50 %/s and at 0.1ish PW then run it from there. Small movements will cause a small amount of PW added to the calculated VE PW.

[dragbug]

Thanks again Ryan. I'll start there.

[dragbug]

set the first pin at 50 %/s and at 0.1ish PW then run it from there.

I thought I knew what you meant by 0.1 PW but once I got home and opened TS...I found out that I don't. I have the first dot set on 50 as suggested but I thought the 0.1 PW was the Added% fuel but found out you can't put in decimal points....so I don't know what you meant. Was that supposed to be the End Pulsewidth setting?

[dragbug]

Not sure if this is AE related or something else. The issue I am having reminds me of how a carburetor might act at a transition point. In this case, cruising at such a low load (20 or less) that the throttle is just barely cracked and you just barely press on the throttle a little more and the AFR drops big time and you can feel the stumble. However, if you are cruising at a higher load (30 or higher) and you press on the throttle, you get a nice smooth transition. I also notice this when decelerating. I can find a spot between closed and barely open that seems to cause the car to go rich. At idle this issue doesn't happen. Does this sound like an issue inherent to a TBI system?

[dragbug]

Attached is my latest tune and log file. Sorry, the log was rather large as I was trying to get as much data as I could. I can't really make heads or tails of when and where the issue is happening. Hopefully someone with more experience can. Thanks in advance!

[dragbug]

Anyone?

[Matt Cramer]

The main thing I noticed is that your TPS based AE goes through a very large range, much of which the engine will probably never see. Your top point is 1800% per second, or snapping the throttle fully open in 0.056 seconds. And the next lowest bin is snapping it fully open in 0.083 seconds. Usually there's not much need to change AE settings above 500%/second.

I also think it may be dumping a bit too much fuel and giving a rich misfire under certain circumstances.

[dragbug]

Thanks Matt. I was basing the range on what I saw from my logs. I was hitting above 1750 so rounded it to 1800. I can drop the max back down to 500 and see how it does.

I also think it may be dumping a bit too much fuel and giving a rich misfire under certain circumstances.

Not sure what certain circumstances you are referring to but as I was saying in a couple posts above, there is a spot between closed and just barely open that seems to be dumping fuel in. I don't believe its AE because I can hold it in that spot and watch my AFR stay rich and I can do it on deceleration as well. Could that be the spots you are referring to? Because that is what I am trying to fix. Any suggestions would be great.

[dragbug]

I changed the max AE to 500 as suggested. If I stab the throttle fast enough, I am getting lean AFR (20 to 22) and a stumble. Maybe there is another setting I need? But that's not my main issue. The main issue is this crazy rich problem when I am barely on the throttle. If the computer wasn't detecting throttle movement, I would expect a lean condition due to the throttle plate opening but no additional fuel. But instead I must be getting a ton but I don't understand why. The reason I don't think its AE is because I can decel to the same point and it goes rich. I can even hold it in that position and it will stay rich. Above this magical point and at closed throttle everything works as expected. Originally AE was set at 50% threshold and right now I have it up to 90% threshold and no change. The data log and tune above are still valid as the only changes I have made are the ones I just mentioned. Any help on this would be fantastic. Thank you.

[Matt Cramer]

Try adjusting the VE table at the point where it's too rich.

[elaw]

Yeah I hadn't had time to post back but I was thinking the same thing. Definitely not an AE issue as AE is not active at the time in question. It just seems like the VE table needs some tweaking.

[dragbug]

I can't look at my logs right now but if I remember right, my load is usually around 25 or less. Those areas on the VE table usually produce an AFR around 13.5 or so when I am cruising in that load range normally. If I lower the fuel, I think I might be running to lean if I happen to be cruising in those bins. But, I will try it and see. I can always go back.

[whittlebeast]

In MLV, create a new field named

MAPxRPM and define it as [MAP]*[RPM]/100 in the second box

Now in MLV plot on one trace MAPxRPM and on the same trace have Duty Cycle. If you happen to have a MAF, it will also follow these other two traces.

The traces should run roughly parallel. When AE hits, the Duty cycle should jump up about 10 or 20 percent for about .1 to .2 sec and then settle back back into the normal pattern.

In the trace you posted, there is way too much AE and you are most likely putting out the fire with all the fuel. The spike in the AFR is the result of the fire going out.

Plotting PW on top of MAP will generally show the same pattern.

Hope this helps

Andy