messiahx

iTrader: (2)

Yes I'm posting again about my AE problems, but it's really what's holding me back from making progress on my tune and actually enjoying the car again. I am using the accel wizard and not EAE because if I can't get the simpler method to do what I want, I have no business tuning EAE.

Basically, 9 out of 10 times, when I step on the gas, my accel enrichments kick in, do what they should for the length of the enrichment (.4 seconds + .4 second taper, a lot higher than most) and immediately after a nice, big lean spike follows. This will happen under nearly any condition. For instance, I can't even step the car off idle without a good amount of AE, and even then, it will stumble in the manner above. The same thing will happen under additional throttle in cruise, and during shifts, all RPMs. I understand that somewhat of a spike will almost always occur, but one that pegs AFRs at 22 should'nt. My TPSdot threshold is low enough that I'll get random AE from noise, too. It also seems like I'm adding in a lot more PW than others do.

I'll be the first to admit that my VE table isn't 100% good yet, but in the idle and cruise areas it's more or less solid with idle usually 13.5-14 and cruise 15-15.5 AFR. You can see in my datalog an example of what's going on.

I know I've posted here about this before (and msextra), but if one of you could give me a virtual smack in the face and point me in the right direction, or aid in troubleshooting, that'd be awesome. Setup is in sig, and I'm running sequential fuel.

richyvrlimited

iTrader: (1)

Your VE table needs to be tuned, the VE table must be 99% before you start messing with other enrichments.

lLook at a log when you stab the throttle, in the load area of the VE table at and just above idle area is where the 'dot' is heading, you need this bit tuned in.

FIWI with sequential you need barely any AE. I actually am running none at the minute.

messiahx

iTrader: (2)

I agree my table still needs work, but is it really that bad, even in the 0-100kPa cruise areas that it would cause that much of a lean spike?

muythaibxr

I notice that you're running seq... do you have the injection timing where it should be? That can make a big difference on AE.

What are your lag factors set to for TPS and MAPdot?

I tend to use numbers between 90 and 100 for both.

Having them too low can cause this as the AE won't respond fast enough. Once you increase these your thresholds for both will probably need to be higher, but it should work better.

Ken

messiahx

iTrader: (2)

I have tried increasing the lag factors before and yes I have to increase the thresholds then. It did help a bit but not enough to eliminate the spikes. With that last datalog I think they were both back down to 60. I will give it another shot.

I searched around for injector timing and couldn't really find anything other than the settings DIYautotune has on their install guide with the sequential module (tested on a '99 miata, I think) which were:

Timing trigger: end of pulse
Fixed timing
Fixed timing 1: -350 degrees
Cranking injection timing 360 degrees

Is there a better timing to use?

messiahx

iTrader: (2)

I changed lag factors for tps and map to 90. Not much of an effect.

One of the datalogs below shows what happens with very slight throttle -- not enough to trigger AE. The other shows what happens when AE is triggered. Note that these are both from idle. And AE still doesn't fix the lean spike under cruising conditions.

richyvrlimited

iTrader: (1)

For that lighter stab you need the VE table tuned in!!

As soon as the TP moves MAP goes up, yet your injector DC goes down... increase the VE in that area.

muythaibxr

I agree... I think you need to turn off AE completely and dial in your VE better.

Also, for the injector timing, finding where it should be is a lot of trial and error.

The way I have been setting it is to do this:

1) Pick a value and tune idle
2) pick a new value 10 degrees away in the idle area. If I have to increase the amount of fuel, then that new timing is bad. If I decrease the amount of fuel, the new value is better (it needs less fuel to get the same AFR).
3) Keep changing the timing by 10 degree values and watching what the AFR does. Keep doing #2 until you have to start increasing VE again to get the AFR where you want it.
4) Keep doing similar for the rest of the table. In my experience, compared to idle, at higher loads you need more timing (higher pressure means it takes longer for the fuel to vaporize) and as RPM goes up you need more timing.

On the 20v 4age, it didn't really seem to matter for AFR where I had the timing, but it did run smoother when you got it right. At other places in the table, it mattered though so I was able to follow the above procedure.

I would also change the -350 to 300 and increase it from there in increments of 10.

Ken

messiahx

iTrader: (2)

...back from the dead.

Two words: Vacuum leak. I had my map line sharing with the WI pressure switch and some other junk. Connected it directly to the manifold after breaking a vacuum tee and sure enough, the car ran super smooth. I didn't hear the leak and the carb cleaner on connections trick didn't reveal it either. This leak is what caused me to abandon my old e-manage setup, too -- I could never get it to respond perfectly either. I feel pretty stupid for not fixing this earlier, but very relieved at the same time.

I do still need to figure out the optimal injection timing, but first I've gotta revamp my VE table. But man am I glad that this is finally working as expected. I thought for sure I screwed up building the MS.