rawkus

iTrader: (16)

Hello all,
Hopefully my title is clear. I am having issues of my car going way rich when transitioning from decel (out of gear) with the injectors off to injectors turning on right before idle. The injector pulse width is much higher than needed to idle, so the car goes rich and nearly dies. The ROM is 90550001. Attached is an example of a log. Disregard the high desired idle, I am trying all sorts of different options to fix this problem. Anyone have any insights on how to make changes in how the ECU controls this transition?

Thanks!

Dan

rawkus

iTrader: (16)

Anyone have similar issues with another ROM perhaps?

Dan

mrfred

iTrader: (50)

Sounds to me like the ISCV tables need adjusting. Have you by chance fiddled with the "decel fuel cut" maps that Mattjin has reported?

Spec-Ops1

iTrader: (38)

To combat this problem I would set target idle where you want it, then adjust the BISS screw in SAS mode to around 800rpm... log ISC steps as well and shoot for around 30 ISC steps at idle while not in SAS mode, see if this helps your scenario... you may have to perform BISS adjustment more than once to achieve desired results...

Jonesey I7

iTrader: (10)

Mine does this too... I WISH I could figure out a way to fix it.

The only thing I do to combat it is when decelerating, I leave it in gear for a couple 3 seconds and THEN let in the clutch. This seems to make it not want to die, and doesn't go sooo rich.

rawkus

iTrader: (16)

This is what I currently do... I will look into the advise above and report back. Thanks!

Dan

Jonesey I7

iTrader: (10)

This is a quote from another thread...

"The idle dip, from my experience, has been related to the 0 load cell timing and the decel fuel cut delay tables. If you remove the 0 load cell the engine tries to interpolate where it should be. The conditions are always different during decel so the ECU never has a series of timing cells to go by. Also, if the decel tables are too high or too low there is issues on run down.

I see that a lot of "tuners" remove the 0 load cell, and several other drivability region loads. I do not recommend this."

Also....

"Look at a stock timing map. Notice a strange bump in timing thats completely out of place in the 0-1000rpm/60load area? This is for decel catch. If you have cams, large injectors or both, this area can SOMETIMES be adjusted to aid a bit as the fuel is being re-applied by the ECU. ?

I am going to try adjusting the ISCV tables a bit, the decel fuel cut tables, and adding some timing to idle load cells....as this seems to be the only "fix" working for most.

Jonesey I7

iTrader: (10)

Also, here is a compilation of ALL the pertinent things I have found over the last two days of searching...

The idle dip, from my experience, has been related to the 0 load cell timing and the decel fuel cut delay tables. If you remove the 0 load cell the engine tries to interpolate where it should be. The conditions are always different during decel so the ECU never has a series of timing cells to go by. Also, if the decel tables are too high or too low there is issues on run down.

I see that a lot of "tuners" remove the 0 load cell, and several other drivability region loads. I do not recommend this.


Look at a stock timing map. Notice a strange bump in timing thats completely out of place in the 0-1000rpm/60load area? This is for decel catch. If you have cams, large injectors or both, this area can SOMETIMES be adjusted to aid a bit as the fuel is being re-applied by the ECU.


Raise 1000rpm/ 60 load cell to 18* timing.


HERE IS THE ANSWER!!!!!

The reason why there is rev hang, is there is a table that adds a LOT to your idle demand (the vertical axis on the idle stepper lookup table) and causes the stepper position to go really high.

The tables I used to completely erradicate throttle hang were under the "ISCV Control" category:

- ISCV Demand RPM Adder [rpm > 4500, load > 170, speed > 12 mph]
- ISCV Demand RPM Adder (Moderated by Baro, IATS) (FFFF6EA8) (sub_204B6) [sub_1EFC0]

These two tables add up to 100% on top of your idle demand percentage and makes the lookup value much higher in the idle stepper lookup table.

Zero these two tables out, and log your iscv steps in evoscan.
You'll notice instead of going up to 120 or so when you accelerate hard, the iscv steps will only go up to around 70 instead, which will not induce rev hang. The iscv steps will also decay very quickly, since it won't reach such a high value like before. I haven't noticed any reduction in idle quality or part throttle driveability.

You may have to play with your Decel Fuel Cut Delay tables as well, but I don't know what the difference is between tables 1, 2, 3, and 4 are. I just made all my tables the same values as the ones in table 2 (this one has the lowest values and is the same as table 4).

Sorry for the long explanation, but it kind of makes it hard without pictures!

AND THE FINAL FIX IS TO PUT ALL 4 TABLES FOR DECEL FUEL CUT DELAY AS FOLLOWS.

LOAD/UNITS
10 / 7
20 / 36
30 / 50
40 / 56
50 / 59
60 / 61
70 / 61

rawkus

iTrader: (16)

Thanks for the posts. Unfortunately, none of these tables, including the Idle Stepper Table have been properly defined for 90550001. :/

Dan

Jonesey I7

iTrader: (10)

**UPDATE**

CHECK YOUR BLOW OFF VALVE!!! Holy crap, ALLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL these issues I was having like yours were all cured with a blow off valve swap. I had an AP bov, I switched it out to the APS.......... everything.... EVERYTHING!!! is gone. No more rich at clutch in. No more idle dying when rolling to a stop. AFR'S completely back in check.ALL PROBLEMS, All gone. Unless you're running an APS or Synapse...... I would say I'm 99.99% sure that's your issue too, and would consider a swap SOON if you'd like to get rid of this issue.

one1fastevo8

I have a tial bov and it does just fine. There r a few people using this one with no issues it in the tuning somewhere. im sure its just going to take time to find the right tables

rawkus

iTrader: (16)

UPDATE!
Turns out my IAC was disconnected!? Many problems solved, but not yet perfect. Will report when I have more time to look into this.

Dan