RomRaider

Hi,

I have an ADM 09 STI (AZ1G202G) with the following mods:

- K&N Filter
- X-Force TBE 3" / High Flow Cat
- Custom Tune
- 205KW
- 98 RON (Shell V-Power)


ISSUE

The car runs great, it makes 205KW at the wheels, good on fuel and learning view FLKC is all zero's

However, there is a funny surge / flat spot at light throttle and light load that I need to solve.

When you reproduce it kind of surges slightly then backs off a bit and surges slightly again. (The is no observable speed increase as it is very slight)

I have researched extensively the so called "GR Stumble" that is here and on other sites and tested many of the solutions. None of which have worked which is why I am posting.

My tuner and I road tested and even put the car back onto the dyno where we could reproduce the problem but still couldn't diagnose it.

We have been able to "tune" around it with reasonable success but I would like to solve it as it makes the car less enjoyable to drive.


SYMPTOMS

The problem seems to happen right on the CL to OL transition, Once in OL it is fine and further down the rev range in CL it is fine.

It is like the ECU can't make up its mind if it should go to OL or stay in CL.

It feels very similar to an old school 4 barrel carburettor that opens the secondaries without adding fuel from the accelerator pump.

It also seems to make a slightly different noise from the intake(?) when it happpens

NOTE: I never noticed anything like this in the stock tune.


STEPS TO REPRODUCE

The problem is reproduced under the following conditions:

- Around 20% Accelerator (Held constant under light load)
- Between 1.10-1.60 g/rev load
- Between 2500 - 2900 RPM
- 3 or 4 gear is best


WHAT HAS BEEN CHANGED IN THE TUNE

The base tune is actually very simple with only the following changes (AZ1G202G), No scaling has been changed and any table not shown below is the same as stock.

Modified tables: 17
(P0420) CAT EFFICIENCY BELOW THRESHOLD changed
Max Wastegate Duty changed
Initial Wastegate Duty changed
Target Boost changed
Rev Limit (Fuel Cut) changed
Primary Open Loop Fueling changed
Primary Open Loop Fueling (Failsafe) changed
Boost Limit (Fuel Cut) changed
Advance Multiplier (Initial) changed
Base Timing Primary Cruise changed
Base Timing Primary Non-Cruise changed
Base Timing Reference Cruise (AVCS related) changed
Base Timing Reference Non-Cruise (AVCS related) changed
Knock Correction Advance Max Non-Cruise changed
Intake Cam Advance Angle Cruise (AVCS) changed
Intake Cam Advance Angle Non-Cruise (AVCS) changed
Checksum Fix changed


WHAT HAS BEEN TRIED

I have tried many different changes on many tables to try and determine where the issue is. All of these changes were made individually and settings moved both ways to determine if there is any impact. I have been able to alter the characteristics of the flat spot but not change the rev range it occurs or get rid of it.

- Changing Intake AVCS both ways
- Changing Primary Open Loop Fueling
- Changing CL to OL Transition with Delay (Throttle)
- CL to OL Transition with Delay Throttle Hysteresis
- Changing Requested Torque (Accelerator Pedal) SI-DRIVE Sport Sharp
- Changing Requested Torque Base (RPM)

There have been two changes that have worked best:

- Changes the tuner made to the CL to OL Transition with Delay (Throttle) and CL to OL Transition with Delay Throttle Hysteresis.
- Changes I made to the DBW tables

The changes to the OL to CL tables worked OK but used a lot more fuel. (Basically kicking it into open loop sooner)
The changes I made to the DBW tables worked well but the throttle response was similar to Sport up to about 60% Accelerator.

The DBW changes were tried as after reading throttlehappy's post http://www.romraider.com/forum/viewtopic.php?f=15&t=4911&hilit=gr+stumble&start=225


LOGS

I have attached a log of a few pulls, The data that was outside the criteria was stripped and the runs that are left are all basically when it was happening.





I would really appreciate any suggestions that have worked on similar issues on the 08+ STI, Or any comments on the logs. It is a minor but annoying issue!

Cheers...

Timing is way too low in that range which is the likely cause for rich fuel mixture in transition. Hopefully you have some logs from the original to get an idea of the timing curve.

Compare the stock Target boost to stock, My bet is the final few columns are the only ones changed?
Open the WGDC tables, the column tables will also be stock, right?

In the Target boost table, find which RT column wastegate boost pressure is reached, say around 200RT

Change the 3rd column value in WGDC to 200, 1st column to 0psi at 3000rpm RT value(around 160) then 2nd column to 180, rescale the 4th to final column as needed

Zero out the 1st column completely, everything from 2400rpm to rev limit in 3rd column is zero'd out on initial and 8 in max, everything from 3000rpm to rev limit in 3rd column is zero'd out on initial and 16 on max.

RT and Target boost is the issue

Thanks for the feedback, However I`m not sure I understand what you think the issue might be.

The way I read the tuned tables at the problem RPM / Accel Pedal is below:

DBW > RT S# is between ~100RT and ~200RT. (2400 - 3000 @ ~20% - 25% Pedal)

Target Boost at ~200RT is ~0 and under (Vacuum).

WGDC (Both) start at 270RT and the value is 0

The appears to indicate that the engine should still be under vacuum (or 0) at this point which is indicated in the logs.

What are your thoughts as to what the engine should be doing?

Thanks again


I have posted the stock tables below as I can`t post the tuned tables (Tuned tables are scaled the same as stock)


Stock Initial WGDC

The tuned table is 0 value in the first column and 65 for all other values



Stock Max WGDC

The tuned table is 0 value in the first column and 70 for all other values



Stock Target Boost

The tuned table is much the same to 200RT, 250 is ~3.5, 280 is ~6.5, 300 is ~9.4, 310 is ~12.2, 320 is ~15.1, (All RPM values in each colum are similar), 330RT and above all values are identical for all RPM`s @ ~18

I am starting to think it might be timing related as well, As I would assume it would be using the cruise timing table at that point I might try replacing the tuned Base Timing Cruise table with the stock table and see if that changes anything.

Otherwise I might flash the stock ROM back in and run some more logs.

What would you expect the timing to be in that range?

Thanks for the feedback

This issue has been resolved with some simple timing changes.

The Base Timing Tables are the same as the tuner had set and I added some more (~6 degrees) to the Knock Correction Advance Tables between 1.00 and 1.60 Engine Load (g/rev)

Learning View is all still zero's after a 500km drive which included "A Lot" of 0-1X0 km/h runs testing Launch Control and Flat Foot Shift so it appears to be fine so far.

Thanks for all the help.

Good to see you fixed it, but have you fitted the fuel line mod? Fixed my hesitation.

I did quite a lot of research regarding the FPR issue but could not find any evidence that this problem was present in my car.

AFR's held rock steady on the dyno and on the road with a wideband so if there is a fuelling problem it doesn't show up.

The other reason I discounted the FPR was that flashing back in the stock tune eliminated the issue.

Having said that the FPR mod seems to make sense and might be worthwhile but it just didn't seem to be the cause of this issue.

I have also made some more changes to the AVCS settings and the car is running really nice now but I have purchased another complete GRB motor for a new E85 / Pistons / Rods / Heads / Cams / Turbo build so this tune will pretty much stay as is until the stock engine is removed.

Thanks for the feedback!