ECMlink Infuriating Idle

[Ludejim]

Hello all,
This car is too much fun! and now that I have had some fun, I really want to get this thing dialed in to perfection. I am following this guide here: Tune Speed Density W/O MAF.

If I am reading this document correctly, after setting up ECMLink and all of my Speed Density (SD) sensors the first thing to do is tune my idle. I cannot for the life of me get this thing to have a steady idle. Short Term Fuel Trims (STFTs) and Long Term Fuel Trims Lo (LTFT) are constantly on the move. I adjust my deadtime and global fuel to compensate, but it seems they end up fluctuating too much to lock in to what I would consider stable.

When the car first starts, the idle seems to be stable (running open loop because it just started), but it is not hitting the target Air Fuel Ratio (AFR). The AFR it is hitting isn't dangerous per se, but it is not what is set in ECMLink. Once the car has warmed up, 190 degree coolant level AND it switches into closed loop, the idle really starts seeking. You can see my O2 sensor oscillating, and you can see my STFTs follow the oscillations. The ECU seems to be doing exactly as told, but the idle RPMs vary greatly and the STFTs seem to change too much to allow a stable idle. My first question I guess is, should the O2 be oscillating faster?

To answer the first response that is going to be thrown out there, yes I have read Terry's Talon Troubleshooting Tips #3 - Fast Idle (aka. Idle Surge) and I have performed the following as instructed by that guide and else where:

• Boost Leak Test - tested up to 30 PSI. I am fairly sure this is not the issue.
• Cleaned Throttle Body (new gaskets in between TB and intake manifold and between TB and intake piping)
• FIAV block off plate
• Throttle Cable Adjustment
• TPS adjustment
• Vacuum lines deleted.
• EGR Deleted
• Adjusted BISS
• Purchased new Idle Speed Controller (ISC). No change. Tested ISC coils, tested good! The newer ISCs resistance should be somewhere in the 40 ohm range from what I have read, that is where my new one sits. Here is a video of the ISC operating, honestly looks broken as hell to me, but I don't know what a working one looks like and cant seem to find a video of a working one.
• Pulled ECU to check ISC drivers and corrosion. Everything seems to checkout. I have attached photos for inspection.
• I've gone to the extent of pulling the timing cover off to verify that the crankshaft, oil pump, and cam timing marks align when cylinder 1 is at TDC.

I am out of ideas, unless that new ISC is bunk and I need another new one. I have attached a data log showing what the car is doing at idle. There is certainly more information I can give. If there are any questions that need answered, I am ready to answer them....

Regards,
James



///////////////////////////////////////////////////////////////////////////////////////////QUESTIONAIRE////////////////////////////////////////////////////////////////////////////////////////

1). Any boost, vac, or exhaust leaks?
Boost leak tested 5psi above your max boost with no leaks? My boost is set at 16 PSI, I tested up to 30 PSI without any leaks

2). Verified mechanical timing? Yes, Cam sprockets dowels at 12 o'clock and timing marks line up in the middle. At the same time, the crankshaft and oil pump timing marks are lined up perfectly.

3). Verified base timing? Yes, 5 degrees BTDC

4). Ignition system
COP or Stock Coil: Stock
Wire brand and Age: NGK, Less than 2,000 miles on them
Spark Plug brand, type and Gap: NGK BPR7ES .028

5). Motor health (Compression Test)
Cylinder 1:160
Cylinder 2:145
Cylinder 3:160
Cylinder 4:160

6). Performed basic throttle body adjustments?
Idle Switch: Simulated in ECMlink, performed TPS adjustment and ECMLink sees .63 volts at 0% throttle and 5 volts at WOT
Throttle Cable: Yes, just enough slack that the throttle doesn't stick
TPS: Yes, as mentioned above, I performed the TPS adjustment within ECMLink
BISS: Tried, but can't get ISCPosition to stay at 30. Grounded diagnostic pin in using ECMLink and set the BISS to get the ISCPosition as close to 30 as I could. ISCPosition wanders trying to keep idle stable.

7). Compression ratio - Stock

8). Any known bad sensors or brittle wiring? uh, don't all DSMs have brittle wiring?? haha in all seriousness, there are some brittle wires.

9). Any DTC/CEL codes? No

10). Electrical system
Car off (not running): Working
Car running: ECMLink battery voltage reporting 14v, drops to 13v when under load (for example when rolling up both windows and they both reach the closed position but your are still holding the buttons)

11). Base fuel pressure and injector values
Base Fuel Pressure: have tried both 37 PSI and 43.5 PSI (of course I adjusted my global fuel in ECMLink when trying both base fuel pressures. I adjusted the fuel pressure using my Aeromotive FPR)
Injector Size (cc/min): FIC 1150s

12). Properly calibrated and configured wideband sensor
Sensor Brand: Innovate LC-1
Calibration Date: N/A
I'll calibrate the wide-band once I can get the car to idle with the stock narrow band O2.

13). Type of fuel
Type: Gasoline 91 octance

14). Watched ECMlink how-to videos? About a million times, My eyes are now immune to them

You must be logged in to view this image or video.

You must be logged in to view this image or video.

 

[Vegas Smith]

Hmm, the only thing I see is that when the idle starts oscillating the afrest is too. Not sure if it's the chicken or the egg. I suspect it's an issue with your tune. The VE map is really bad. The airflowperrev is too low. Put the stock 1g VE map back in and make sure your global and deadtime are accurate as well as injectorbatadjust values. I don't think it has anything to do with the idle system. Everything looks solid there.

Also, don't use throttle body ports for vacuum/boost sources; they're inconsistent.

 

[ATF 2GTalon]

Try setting the Coasting FC offset to 0 and see if that brings your idle down to your target idle.

Nothing really jumps out at me. Most of your settings look pretty good. Fuel trims are in a good range.

 

[steve]

I don't see anything wrong with your ECU or how the ISC is stepping but that ISC in the video doesn't look very new.

 

[Ludejim]

Hmm, the only thing I see is that when the idle starts oscillating the afrest is too. Not sure if it's the chicken or the egg. I suspect it's an issue with your tune. The VE map is really bad. The airflowperrev is too low. Put the stock 1g VE map back in and make sure your global and deadtime are accurate as well as injectorbatadjust values. I don't think it has anything to do with the idle system. Everything looks solid there.

Also, don't use throttle body ports for vacuum/boost sources; they're inconsistent.

I could have sworn I reset that darn Speed Density Table. Anyways, here is what I did, I downloaded the stock settings from this ECMLink Page and applied these settings:

• Reset VE map to stock
• Reset global and deadtime to recommendations on ECMLink page
• Reset OpenLoopMaxOct and OpenLoopMinOct tables
• Reset TimingMacOct and TimingMinOct tables
• Reset InjBatteryAdj Table
• Target idle 750

After resetting everything, I started the car. The idle RPM was much higher than the target 750 RPM. I tried grounding the diagnostic pin in ECMLink to set the BISS, and the screw bottomed out, thats where it sits now.

I let the car idle until closed loop kicked in, adjusted the dead time a little, and then saw that the Front O2 wasn't oscillating and that the STFTs seemed to max around +16.8%. I let the car idle until it got to 190 degrees and then watched the LTFT Lo start to make its adjustments. Ultimately LTFT Lo pulled enough that STFT started to drop and the Front O2 started to oscillate again. From there I made adjustments to global fuel to get LTFT Lo to drop closer to 0%. The global fuel and deadtime aren't perfect, but I figured I would get some feedback before messing around too much more.

The cars idle still audibly ramps up and down. You can see it in the log as well dipping into the 600 RPM range and reaching somewhere in the 800 RPM range.

 

[Ludejim]

I don't see anything wrong with your ECU or how the ISC is stepping but that ISC in the video doesn't look very new.

Good eyeballs, its new in the sense that when I built the car I replaced it. The ISC has less than 2000 miles on it, and honestly I would put it at more like less than 1000. I have not driven the car very much at all since I built it forever ago.

That being said, does the ISC operate like it should?

 

[steve]

That being said, does the ISC operate like it should?

It's stepping in and out like it should. It's been a long time since I've looked closely at mine but it seems like a lot of carbon on it for the miles. How's your PCV and EGR systems?

 

[Vegas Smith]

You need to re-enable tps adjustments. Your isc appears to be working fine and idle looks good.

 

[Ludejim]

I redid my TPS adjustment a couple times. The car seems to want to bounce around .65 and .63 volts, just depends on the cars mood. I'm not satisfied with this idle, I am also not sure what to expect. I want an idle that doesn't wonder ~200 RPMs. Am I asking too much?

I have attached a log that shows the idle wandering. I took this video at the same time this log was taken. If you look closely you can see the RPMs shifting around and you can see the AFRs fluctuating too much for my liking. If you listen to the video, you can hear the idle.

The only thing I changed from in this log was the TPS adjust. It didn't seem to change anything for me. I'm stumped...

I wouldn't expect the RPMs to fluctuate this much in a stable idle. (I did rescale my RPM axis to help exemplify the RPM fluctuations. From peak to peak is about 200 RPMs)

You must be logged in to view this image or video.

EDIT: Let it be known that I have 2 cells in the OpenLoopMaxOct table changed because at idle the ECU dances between those cells sometimes and I want to keep a stable target AFR at idle.

 

[Ludejim]

It's stepping in and out like it should. It's been a long time since I've looked closely at mine but it seems like a lot of carbon on it. How's your PCV and EGR systems?

I deleted the EGR system. I have PCV Setup #1 – The Improved Factory Setup from this tech article. Ultimately blow by was getting the best of me and I also added a oil cap that has a breather that leads to another catch can.

 

[Vegas Smith]

If one of the breathers doesn’t lead back to the intake then thats a possible vacuum leak. Otherwise, there is probably some small leak you’re missing.

 

[Ludejim]

If one of the breathers doesn’t lead back to the intake then thats a possible vacuum leak. Otherwise, there is probably some small leak you’re missing.

What about a leak in my down pipe at the flex portion? I just did an exhaust leak test and air is blowing out of that portion of the exhaust.

I also got tired of trying to find a vacuum/boost leak with the traditional methods (boost leak tester). At the recommendation from a friend I picked up some carb cleaner and started up my car. Before it started to surge on its own, I sprayed carb cleaner at every junction I could see. Intake pipe to throttle body, the entire throttle body, Throttle body to intake manifold, intake manifold to head, oil catch cans, along all connections of intake piping. Nothing.

 

[Ludejim]

I have been reading a lot. Looking at other peoples logs and really digging in. I have found an option that I didn't know about before. O2FeedbackAdjRate on the STFTAndO2Feedback page in the direct access tab. For that matter, all of the setting on that page can potentially help here. What's odd is that I usually see people recommend this when someone is using a wideband to simulate their narrow band and their wide band is mounted further back on their exhaust. My narrow band is not simulated and is right where it should be, just after the exhaust manifold in the O2 housing. I've linked a few threads where O2FeedbackAdjRate is talked about to help stabilize the idle.

https://www.dsmtuners.com/threads/running-rugh-log-attached-inside.342283/

https://www.dsmtuners.com/threads/idle-setup-problems-fuel-trims-maxed-out-need-help.519776/

v3narrowbandsim [ECMTuning - wiki]

www.ecmtuning.com

 

[Lucky13]

Your AirflowPerRev is kinda low at around 19-20....shoot for something between 25-30.
ISCPosition is also a touch low at mid/low 20's but it holds pretty constant.
TPSVolts should be at .63...your at .65....minor....but just pointing it out.

 

[El Cuervo]

I also got tired of trying to find a vacuum/boost leak with the traditional methods (boost leak tester). At the recommendation from a friend I picked up some carb cleaner and started up my car. Before it started to surge on its own, I sprayed carb cleaner at every junction I could see. Intake pipe to throttle body, the entire throttle body, Throttle body to intake manifold, intake manifold to head, oil catch cans, along all connections of intake piping. Nothing.

Just a thought. Did you check for boost/vacuum leaks on the afpr? They tend to leak on the adjustable stud/nut.

 

[Ludejim]

Your AirflowPerRev is kinda low at around 19-20....shoot for something between 25-30.
ISCPosition is also a touch low at mid/low 20's but it holds pretty constant.
TPSVolts should be at .63...your at .65....minor....but just pointing it out.

I agree with you assesment, but it is terribly difficult to dial in AirflowPerRev and ISC when the idle is all over the place.

 

[Ludejim]

I am changing my tactic as I suspect something is wrong. In the attached log, I have the ECU locked in open loop. RPMs are fluctuating a lot, and it is not coming from STFTs or LTFTs. Why would the idle be walking up and down like this? In my timing table, I can see the ECU bouncing between two cells 8 and 13.

From what I have read, a Speed Density setup should not be effected by a boost leak. I am positive I do not have a boost leak before my throttle body, and even if I did, I don't think it would matter because the air density is measured after the throttle body via the MAP sensor in the intake manifold. If there is a leak within the intake manifold, the MAP would be measuring that leak as well, hence compensating (that's partially a statement and partially a quesion ). This is covered in this thread, and the post I link to specifically has some interesting information about post MAP air leaks. I really feel like I'm grasping at straws here...

What would cause my idle to surge in open loop?

You must be logged in to view this image or video.

 

[Vegas Smith]

I don't think it’s tune-related. You probably have a leak at the fiav plate, isc oring or some other hard to find place. That log behavior simply represents a vacuum leak in my amateur opinion.

 

[Lucky13]

I agree that a speed density setup should be less prone to boost/vacuum leaks. Have you tried throwing everything else out the window and adjusting the Speed Density table to get AirflowPerRev to 25 - 30? As we modify these setups farther and farther from stock the base maps are just a starting point. I've looked at some logs from other cars on the ECMLink forums and try and work backward how they got to to where they are based on their mods. Spend time searching around on those forums you can find cars " close " to your setup and where they needed to be to get started.

 

[jed344]

Honestly the revent log does not lot terrible. Airflow per rev needs to come up a hair, otherwise i see mostly 50rpm movements in idle changing? To raise air flow per rev you will increase VE at idle in the idle area then likely have to pull some dead time to keep fuel trim within 5%.

 

[Ludejim]

I am going to attempt to respond to a few of the questions / statements

You probably have a leak at the fiav plate, isc oring or some other hard to find place. That log behavior simply represents a vacuum leak in my amateur opinion.

@Vegas Smith - I have an FIAV bypass plate installed in between the lower and upper half of my throttle body. This plate leaves the ISC port open for idle adjustments. I'll take a video of me boost leak testing the car if I come that far around again.

otherwise i see mostly 50rpm movements in idle changing?

@jed344 - My idle is surging from 600 to 800 in some logs, that's pretty big to me. I wish I could get it to lock in to +/- 50 RPMs...

Have you tried throwing everything else out the window and adjusting the Speed Density table to get AirflowPerRev to 25 - 30? As we modify these setups farther and farther from stock the base maps are just a starting point. I've looked at some logs from other cars on the ECMLink forums

@Lucky13 - In the log attached and the video linked here, I have done as you say and just ignored the wobbling idle. I got my AirflowPerRev to ~.25, ISCPosition is right around 30, fuel trims are hanging out around 0-ish. The RPM at idle still varies by 200 RPM (looking at the log, peak to peak is ~1100-900 RPM). It is audibly annoying.

I tried looking for some other tunes on the ECMLink forums as you suggested. I can find a lot of logs where people are having problems, but few if any logs where the poster says, "my car runs great, check out my tune!". I did still browse the logs of others just to see how my tune compares to other "bad" tunes. It seems pretty common for bad tunes to idle surge like mine. Something I have noticed is timing. My timing sure does seem to be all over the place compared to what my timing table says it should be...

 

[Lucky13]

I don't see it listed in your profile....are you running aftermarket cams?

 

[AWD-Tony]

Is it once in a while, possibly when your fan turns in and off?

EDIT: I don't see idle surge in the last log you posted. BUT, I see your target idle is 750 and in the log is between 950-1000. Does the car still have FIAV? If so, how is the cold start?

 

[Vegas Smith]

In the last log your car is idling at 1000rpms but your target idle is set to 750. I would redo your entire idle system because I think you're chasing your tail. I would completely reset the isc. Complete this checklist to a T : https://www.sixsigmatuning.com/dsm-idle-control

My guess is you are adjusting your idle when the "learnedidleadj" is still adjusting. You're not logging this value so go ahead and do that. If you read through this link^^ it explains how to reset the isc. I always adjust my idle with a freshly reset isc and a warm car. I reset the isc then immediately start the car and adjust to target idle with isc around 30 and learnedidleadj around 142/4 (?).

 

[Ludejim]

I was having difficulty last night setting the ISCPosition while at 750 RPM. That is why the target idle RPM doesn't line up. I will start from scratch later today and follow the guide that @Vegas Smith posted to get my throttle body into a known state.

@Lucky13 - @JackM (aka Jacks Transmissions) rebuilt the motor for me. I gave them my stock head, so I can only assume they put stock cams back in it.