[dleach1407]

Hi everyone, I dont post here much but I normally post over on EFIDynoTuning. I havent been able to get this one particular issue resolved and I thought i would try here as well. I have a 95 mustang using T4M0 ECU and Deciphas base tune T4M2 and TPRT. All of the basics are already done. injectors, maf, timing all of that is done and the car has been running for a couple years. It was already put on a dyno and tuned. My main issues now are dashpot and going lean after a shift. I was told to the issue with going lean after a shift is my manifold volume. I have tried lowering and raising manifold volume which does not seem to resolve the issue. I can affect the initial tip in by changing the manifold volume which now is not going rich or lean initially but I still get the lean spike after a shift. The lean spike is pretty significant to the point it causes misfires when cold. I believe I need to extend the accel time constant table but it is basically maxed out in some cells so my thought was to double the xmul for the time constants and then reduce the accel and decel time constants to give me more adjustability. Should I be looking at something else instead? Thoughts?

[cgrey8]

My initial thought is to look at your KAMRFs and LAMBSEs before and after the shift.

When running Closed Loop, the EEC modulates the LAMBSEs to maintain stoich (more correctly, to keep the HEGOs switching). And if you have adaptives on and learning, the KAMRFs should be adjusting to keep the LAMBSEs closer to 1.0 (or 14.64 if you are viewing them as an AFR). However if you have adaptives disabled (globally or disabled in the RPM/Load area before the shift) or not giving adaptives enough time to adjust/learn, then it's possible you could have a situation where higher RPMs (right before a shift) are working the LAMBSEs up to a lean value (pulling fuel) as the RPMs are rising in that gear. Then the shift happens and the engine suddenly finds itself in a completely different RPM/Load range, where that LAMBSE value may simply be too lean. Theoretically, if the LAMBSEs are working their way high, then the KAMRFs learn-out that high adjustment. But if that's not happening, then you'll need to find out why.

However if this is happening, then another option is to disable the KAMRFs, clear them, and work the tune until that high RPM lean-out of Closed Loop isn't there anymore. My gut feeling is this is being caused by Injector Parameters that need some tweaking. However it could just as easily be adjustments needed on the MAF curve. Read over the Tech Doc on this forum that talks about tuning Injector Slopes. You will need datalogging ability AND a licensed copy of EEC Analyzer. But if injector parameters are the problem, having the tools to help you identify the problem and correct for it in the right way will be worth the money.


Now if you are not suffering from an issue where LAMBSEs are working their way lean, then after the shift are "too lean", then this could just be a transient fueling issue. It just depends on how long after the shift the engine stays lean. If it's just for a second, then transients may be enough to take care of this. However if you have to witness the LAMBSEs work their way down/rich before the leanout goes away, then I suspect your problem is the prior issue, and thus not something that is going to be solved by transient enrichment, since transients are meant to be VERY quick enrichments (decayed and gone within a second), not enrichments that persist over multiple seconds.

[dleach1407]

My maf is dialed in and my fuel error is within a few percent of perfect. I shoot for a fuel error of 1 to 1.03, never anything negative. This happens in open or closed loop with KAMs cleared or not. The higher the rpm the worse the issue is. My injectors are siemens 80s from a reputable supplier and I used fords information for my injectors running 39psi. I recently upped my pressure to 60psi and changed my injector data based on fords information. The exact same issue with the same severity existed before and after the change. I am confident I got the injector data entered correctly. I have TunerPro RT which has datalogging.

[cgrey8]

I'm confident you probably have the data entered correctly too. However I am not confident that the data, as entered, is consistent with how they are behaving in your application. To know that, you'd need to run an injector analysis through EEC Analyzer using a datalog that this leanout issue is occurring in. But the fact that this happens worse as the RPMs rise does further my suspicion that your injector parameters may benefit from some tweaking.

Or I could be completely wrong and it be something else. But as I mentioned, if the leanout only persists for a brief moment after the shift, you may be able to smooth it over by adding some transient enrichment on a sharp increase in Load, which is what happens during an up-shift.

[dleach1407]

I'm confident you probably have the data entered correctly too. However I am not confident that the data, as entered, is consistent with how they are behaving in your application. To know that, you'd need to run an injector analysis through EEC Analyzer using a datalog that this leanout issue is occurring in. But the fact that this happens worse as the RPMs rise does further my suspicion that your injector parameters may benefit from some tweaking.

Or I could be completely wrong and it be something else. But as I mentioned, if the leanout only persists for a brief moment after the shift, you may be able to smooth it over by adding some transient enrichment on a sharp increase in Load, which is what happens during an up-shift.

So apparently I posted this info last year too and was hoping you could tell me how I could add the enrichment on sharp load increase. Unfortunately, I never saw your reply but I am unsure how to make these changes. Im having a hard time getting any info about which transient tables I should be adjusting.

[cgrey8]

If this were GUFx, I could give you the PID names. But since this is CBAZA, I don't know what the accel and transient settings are, what's available, or what values generally increase or decrease enrichment. I'll have to leave that to one of the guys familiar with CBAZA.

But I can clarify the difference between accel and transient.

Acceleration Enrichment is increase in fuel enrichment purely due to increased throttle based on TPS movement. The amount of enrichment is a function of both how much movement the TPS experienced and how fast it was moved. But once the TPS stops moving, enrichment also stops. This is mimicking the accelerator pump action on a carb. It's also worth note that this only ADDs fuel. It does not take away or reduce fuel.

Transient Enrichment is enrichment AND enleanment based mostly on changes in Load. There is a temperature component to it as well that is important to be aware of since the amount of fuel needed for transient enrichment when the engine is cold is different than when the engine is warm. But for the most part, I'm assuming you'll care more about the warner end of the spectrum. Transient enrichment differs from accel enrichment in the following ways:
-The algorithm adds fuel when Load increases as well as removes fuel when Load decreases
-There's a time/decay component involved. Even after the change in Load has stopped, transient will continue to add/remove fuel for a very short period of time (0.5-2.0 sec depending on how much change in Load just occurred).

The theory behind transient enrichment is that in Port Injection, not all of the fuel injected goes into the cylinder. Some of the fuel sticks to the walls of the intake port. The amount of fuel that sticks is dependent on a lot of things such as wall temps, amount of fuel being fired each cylinder fire, injector timing, etc. Once the engine is running stable, the amount of fuel that sticks to the wall and is aspirated off the walls stabilizes. But during a transient where the amount of fuel the injector fires is different (less or more), the wall wetness will either increase or decrease.

During accelerations, more fuel will be sprayed. So for a small amount of time, more wall-wetting will occur. Thus fuel intended to go into the cylinder will instead stay on the walls causing that cylinder to fire slightly lean. Transient enrichment intends to compensate for this. Wetting the walls can take multiple cylinder fires to stabilize, but it suffices that each successive cylinder fire after a major load change will consume less and less to wet the walls as the walls become saturated.

Likewise on a drop in load, suddenly the injectors are firing less. So the wet walls will start drying up. The result of this is cylinder fires during this period will be receiving more fuel than was injected making the burn a bit rich. This is why transient enrichment will take fuel away to compensate for the over-enrichment that will happen during this time period. But running rich, particularly on a decel is hardly a tuning problem most people care about. It rarely causes driveability problems. It's usually the tip-ins where issues are experienced and thus the tune needs to be tweaked. But if you happen to have momentary decel issues, keep in mind transients do play a role here too.

[dleach1407]

nevermind... I got the info.