Having experimented with the "new" AFC hack, and read all the advantages of it over the "old" hack (involving check valves, etc.) I have to ask the same question that DBman posed several times on another thread:

Why does this work better?

I have recently experienced some of the problems that the "new" style hack supposedly cured. One is bogging at the change from negative to positive boost during closed loop operation. My A/F guage goes completely lean right there, then jumps up after about 1 second of bad hesitation. Resetting the ECU usually cures this for awhile, but certain conditions (not sure what those are yet) seem to bring it back.

Another problem involves going from 0 throttle to part throttle boost. Once again, the A/F guage will go dead-lean for a moment and the car will hesitate some, then regular closed loop operation resumes. I have heard this situation described as a "lazy O2 sensor", but that theory can be refuted by the fact that resetting the ECU usually cures the problem for awhile, or at least affects it somewhat.

So, what could be happening AT the ECU which would cause this sudden lean condition at light throttle or the switch from negative to positive boost. Has anyone managed to work through this yet and diagnose it properly? I think the 2 conditions are linked somehow, and IMO the issue does not involve slow response from a tired O2 sensor. Something else is going on here, I just don't have the knowledge to figure it out... Has anyone else had better luck with this?

 

I'll bet that if you amplified your TPS signal so that you go into open loop earlier, say at 50% throttle it would help.

 

What would be the best way for that? The "TPS Relay tcick"?

- Slater

 

Not familiar with the TPS relay trick but I can imagine using a relay to switch +5 volts to the ECU TPS input at a certain throttle position would work.

I was thinking along the lines of using an op amp to boost the output voltage of the TPS.

Now that I think about it, I'll bet that if you changed the reference voltage of the TPS from +5 volts to +12 volts that would do the trick. Not quite sure of what the ECU does when it see's more than +5 volts on the TPS input but it's a high impedance input so it's probably OK.

 

So, the hesitation may be related to information from the TPS sensor, interpreted incorrectly by the ECU? I wonder what happens after that second of hesitation and the ECU suddenly starts "behaving"... This correction occurs even if we never hit the open loop point on the TPS. Is it possible for the ECU to switch to open loop in spite of low TPS readings?

Here's a theory: Maybe part throttle boost is causing conflicting calculations due to the odd relationship between the TPS and manifold pressure readings. The ECU is temporarily "confused", but the TPS information is quickly overridden by voltage readings from the MAP sensor. Or, the ecu jumps to open loop. Is this scenario even possible given my ignorance in the workings of the ecu?

 

so you've successfully run the new v2 hack? how did you set it up, i cant seem to get mine to do much of anything lol...

 

The ECU should transition directly to open loop if the MAP voltage changes quickly enough. TPS function is, to my understanding, limited to an accelerator pump type action when throttle is activated and to kick the ECU into open loop when it goes over 4 volts.

If you correct your ignition map, bucking is a thing of the past.

 

As I understand it the TPS is used to trigger open loop at a certain percent throttle (not sure what it is exactly) and also to trigger a spurt of fuel richness when the throttle opens quickly.

Perhaps the ECU in closed loop just can't react quickly enough to the changes in manifold pressure when you are part throttle boosting.