I had a relocated OEM IAT sensor in the ASM arm and it didn't seem to have a huge effect on the stuttering at 2000 rpms. I now have a GM IAT sensor in an aftermarket TI arm and it seems better. The car always seems to transition at 2000, even if it doesn't stutter. I am going to post my calibration in the morning, maybe we can all compare and see what is simmilar and different in our calibration files.

Jeff moved me from the modeled fuel mode to traditional during my tuning process, I didn't really understand why. Also I watched some of his training videos on his subscription site and his tune of my car seems to contradict the advice given there.

 

def sounds like a tuning related issue. Link ECUs do not like using any wideband other than a Link wideband. Without data logs or a car in front of me, its hard to offer assistance.

 

I’ve heard that from several people. I’ve also heard that relocating the IAT sensor from the intake manifold to the intake arm would help. However there are several people on here that have both the LINK wideband and relocated the IAT sensor and still have the same issues.

I’ve sent a couple people my logs and they are reviewing them but at this point, I’m pretty much done with it after dealing with non stop issues for two years. Car drives perfectly with factory ecu it’s nice! Slower but nice

 

Are the link widebands not just a bosch wideband in link packaging? That's what I seen at a local link distributor

 

Yes, the actual sensors are Bosch, as are the majority of the wideband sensors used by various companies (high end standalones like Motec often use NTK sensors).

The main difference between the different wideband systems are the wideband controllers not the sensors themselves.

 

Ah ok, thanks for the info. My tuner recommended getting a wideband to use with the G4X so I'll make sure to go for the link one when I do.

 

Do people have these issues with Hondata KPro or Doctronic?
I have a 2004 drive by cable and I want to have a tune on my car.
I’m leaning towards going with Hondata and replacing the parts needed or Doctronic.
Any information is appreciated.

 

Hi,
I noticed references to this thread in the Link Forums that was posted, a few of us over there had been reviewing the S2000 logs and trying to provide assistance.
Apologies for the long post..

I thought it might be worth providing some updates as the original poster - Sonic9? - seems to have left the discussion over on the Link forums.
Adam from Link had found that in a few of the LinkECU logs provided (from users here), that there is a correlation to erratic AFR traces likely contributing to the "bucking" or hesitation issues whenever the Purge solenoid is activated at a certain IAT threshold.
Along with this, while the Purge solenoid is active, the logged AFR value barely ever tracks the intended AFR Target, so Closed Loop Lambda was sporadically trying to correct the logged AFR as a result.

We weren't sure if the issues were related to Wideband o2 configurations or wiring issues, faulty sensors, or another auxiliary output issues, masking the root cause of the problem.

We had requested some S2000 users provide log files with the Purge function disabled in the LinkECU software to determine whether AFR would track target correctly - or there were more issues at play;
Overnight, "TexasRoller" has provided some updated logs both with Purge disabled and Closed Loop Lambda control enabled, and Purge disabled with Closed Loop Lambda disabled.
I've noticed now the AFR vs Lambda target is much more consistent now that the Purge solenoid is deactivated and logged AFR tracks AFR target well.
While "TexasRoller" latest logs were taken at lower IAT where the bucking issue wasn't prevelant - the latest logs at least indicate that the bucking issue should be drastically reduced, if not resolved entirely.

What now needs to be determined is why the Purge solenoid is having such a drastic impact on AFR, and how to resolve it correctly.


You can see here in - I believe Sonic9's log file from March 25th - the bucking issue is apparent when the solenoid valve activates in orange.
For the first 3 seconds, AFR tracks normally against AFR target - then by the 4th second the Purge solenoid activates (170° F IAT), AFR goes mega rich to 9.44, and Closed Loop Lambda immediately tries to pull 15% of its maximum allowable fuel to get AFR back onto the 14.7 AFR Target.
You can see the "bucking" occurring for almost 7 seconds as the Purge solenoid rapidly cycles on and off and Closed Loop Lambda is pegged trying to pull 15% of fuel out..


S9 bucking


Conversely in TexasRoller's examples overnight - with Purge disabled and Closed Loop Lambda enabled - AFR is almost always on Target, with CLL still doing a fair amount of work in yellow, adding roughly 3-9% of fuel


TexasRoller CLL on

With TexasRoller's Purge disabled and Closed Loop Lambda also disabled - AFR is still consistent, but its a fair way off target in some portions - at sections 15% leaner than Target AFR.


TexasRoller CLL off


I also noticed the same "bucking" issues correlate to the Purge solenoid activating in other S2000 owner in the Link forums who provided logs - "MK4derek".
At least now we seem to know what looks to be the root cause, hopefully over the next few weeks it can be resolved.

Hopefully some of this info helps others here.

 

Hi, just a few more general comments..

We now know the issue is correlated to the Purge solenoid/valve activating - the Purge function looks to be enabled in the S2000 AP1 LinkECU Base Map, so its likely its an OEM requirement from Honda.
The bucking only seems to occur above about 170°F when Purge activates - not all Tuner's will encounter see this issue while dyno tuning, or remotely tuning via E-Tune if its a cold day and IAT's never reach 170°F where Purge would activate.

This seems an odd choice to go back to Traditional from a Modeled fuel equation;
Modeled is useful if you are looking to scale to larger size injectors in future, or make a switch to Flex-Fuel with E85, as the VE (volumetric efficiency) of the engine is already mapped out in "Fuel Table 1",
the ECU can do 99% of the work automatically when you then enable "Modeled Multi Fuel" for E85 Flex - it literally saves you hours tuning "Fuel Table 2" from scratch to suit E85 under the Traditional "millisecond" based fuel equation.

This is untrue - there have been a number of users that use all types of Wideband o2 sensors successfully with Link G4+ and Link G4X ECU's;
Haltech WB1 / WB2 wideband controllers, Spartan "14point7" controllers, AEM X-Series CAN and AEM UEGO Analog wideband controllers..

I've had both an AEM X-Series wired to a Link G4+ and G4X via CANBus (AEMNet) and an AEM UEGO Analog 5v based wideband wired into the Link ECU's;
Both operate extremely well and allow me to use Closed Loop Lambda functions in the LinkECU's without any issues at all.

The Link CAN Lambda's are recommended because they interface well with the LinkECU's - have heating control for the Bosch LSU 4.2 and 4.9 sensors - and feed error and updates states back to the LinkECU's over CANBus in the event of sensor failures

 

@jdniss hey im TexasRoller from Link forums. Im going to the track this weekend and we are supposed to have warmer weather, so ill be able to do a back to back log of the Purge on / off to see if it bucks, and then stops.