ECMlink Weird idle when operating temp.

[SynthRocker]

So car runs smooth, until I drive it for a little, usually after a drive, turn off, than a hot start, I get a rough idle thats random. I caught a log of it, and the idle smoothed out and went normal again when the cooling fan kicked on, and it was nearly instant normal idle after running rough. Any ideas?

95 TSi AWD 7 bolt , Charcoal, EGR, fuel pressure solenoid deleted (rough idle was before they were deleted also). Has intake, DSMlink.
When the rough idle starts it almost sounds like the car is cammed and has a choppy cam. Drivability is fine besides the idle, smooth acceleration. Was gonna possibly do the FIAV bypass, I got the ISC to sit at 30 when the idle is normal.

 

[Stapl3]

It's not a cooling fan thing, rather it stops when it gets out of closedloop. It is odd to me that your afratioest is a ridiculous value before this switch happens. I also wonder what the trigger is for the ECU to switch to openloop and stay there.
The quick hack would be to just lock the car in openloop. You could even play around with the openloopthresholds to have it be only in closedloop during cruise.

As for how it would be properly fixed by say, a dealership, without ecmlink, I am not sure.

 

[SynthRocker]

It's not a cooling fan thing, rather it stops when it gets out of closedloop. It is odd to me that your afratioest is a ridiculous value before this switch happens. I also wonder what the trigger is for the ECU to switch to openloop and stay there.
The quick hack would be to just lock the car in openloop. You could even play around with the openloopthresholds to have it be only in closedloop during cruise.

As for how it would be properly fixed by say, a dealership, without ecmlink, I am not sure.

So, strange turn of events, I thought I had reset the used ECMLink to stock settings but after looking diving into it again the other night I noticed it seemed to have reverted back to the settings the prior owner had, anti lag, fuel injector specs, etc. so I’ll be again trying to set it back to stock

I downloaded the stock 2g file from ECMLink but can’t implement it because it’s based on older firmware, the only difference I see that I can’t match is under the fuel tab there’s extra tabs after global dead time that the old firmware doesn’t have so I’m not sure what the specs should be there. I zeroed the global size and dead time out.

 

[b00zt3d spyd3r]

try enabling hot start enrichment? Had a similar issue during hot starts, would idle super lean and stumble and sound cammed until it kicked in to closed loop. hot start enrichment fixed it.

 

[SynthRocker]

try enabling hot start enrichment? Had a similar issue during hot starts, would idle super lean and stumble and sound cammed until it kicked in to closed loop. hot start enrichment fixed it.

Yeah that was unchecked I’ll be doing that when I get home to try it again tonight

 

[SynthRocker]

So I believe I set ECMlink back to factory settings, the reset procedure didnt work so I copied the stock 2G ECMlink file they have on their site (its an old firmware version so I cant copy it to ECM as theres differences in the software now.)

I reset fuel trims and started the car and o boy it barely wanted to run on factory settings. It evened out as I added global deadtime to STOCK injectors when it should be zero..right?... but I got the idle to even out by putting deadtime up to 800... thats very incorrect is it not? Theres also more options under the fuel tab that are not on the old firmware file so I can not tell if theyre factory settings? attached are pictures of what I believe "factory" settings are (besides the fuel tab no idea if thats right because its not on the old file) as thats what the old firmware "stock" file has. Anyone confirm? Also attached a log but no idea what itll show, though i noticed my front o2 is not switching and staying at .2 so either something is wrong with it or the settings are incorrect and its not setting the parameter to meter (staying in open loop)? I guess ill install my wideband earlier than expected (AEM x series) I had a bung welded in so I could run both front o2 and wb but looks like ill just put it in the front o2 spot and simulate. (I wanted to get most issues ironed out before I added stuff, but a WB would only really help fine tuning so meh)

Are there other datapoints I should be datalogging, any idea what else to look at?

After adding deadtime to the stock injectors I got a smooth idle and no drivability concerns, it pulls fine, boosts fine, no stumbles etc. Even after warm starts, the idle issues went away

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

Thanks!

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

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

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

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

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

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

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

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

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

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

 

[SynthRocker]

Sorry to bump anyone able to confirm those settings for me and any idea why I had to add deadtime to stock injectors ?

 

[b00zt3d spyd3r]

first thing i notice in the log is that your front o2 isnt cycling, just stuck at .20v. If that isnt cycling properly you wont enter closed loop operation for idle. I would start there.

 

[SixBolt_16G]

Only had time to skim because im at work so i might have misread or not read enough.. But noticed that on the same screen as the hot start enrichment, you also have the box above checked for disable for airflow smoothing SD....If youre trying to go back to st9ck and dont have Speed Density, you should uncheck that box

 

[SynthRocker]

first thing i notice in the log is that your front o2 isnt cycling, just stuck at .20v. If that isnt cycling properly you wont enter closed loop operation for idle. I would start there.

Yeah I noticed that too, I have a wideband I guess I’ll try and put into simulation to see if it’s the sensor. I had a extra bung welded in so I could run both but o well haha

 

[SynthRocker]

Only had time to skim because im at work so i might have misread or not read enough.. But noticed that on the same screen as the hot start enrichment, you also have the box above checked for disable for airflow smoothing SD....If youre trying to go back to st9ck and dont have Speed Density, you should uncheck that box

Ok will do, it’s funny because the “stock 2g” file from ecmlink has that checked off never had speed density and have factory maf

 

[SixBolt_16G]

Looking at your log. Your TPS is off... Its at .67 .... it should be at .63

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

 

[SynthRocker]

Looking at your log. Your TPS is off... Its at .67 .... it should be at .63

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

Ah word, I’ll look into adjusting that tomorrow, I’ve read a few threads for it, and seems there’s 2 different ways to set it (the FSM way and ECMlink way, and you can simulate the IPS. The FSM way is to adjust the IPS so it may mess with TPS volts?)
Could this be a reason my o2 isn’t going into closed loop and staying at .2v? I guess I gotta also test the o2.

To much to do so little time

 

[SixBolt_16G]

Ah word, I’ll look into adjusting that tomorrow, I’ve read a few threads for it, and seems there’s 2 different ways to set it (the FSM way and ECMlink way, and you can simulate the IPS. The FSM way is to adjust the IPS so it may mess with TPS volts?)
Could this be a reason my o2 isn’t going into closed loop and staying at .2v? I guess I gotta also test the o2.

To much to do so little time

For me, i just loosened the 2 bolts on the tps sensor that sits on the front side of the throttle body with a socket, started a log in ECMlink and twisted the tps until it read .63 and then tighten the bolt carefully because it wanted to fluxuate between .63 and .65. There is a way to compensate for it in ECMLink, but since i didnt do it that way Im not going to pretend that i know how to do it. I have a 6-bolt and a lot less stuff in my engine bay so it was just easy for me to get to it. I dont remember the accessibility with a fully loaded 7-bolt bay.

I dont have enough knowledge to know if its associated with your o2 problem, but it needs to be adjusted regardless.

 

[SynthRocker]

For me, i just loosened the 2 bolts on the tps sensor that sits on the front side of the throttle body with a socket, started a log in Ecmlink and twisted the tps until it read .63 and then tighten the bolt carefully because it wanted to fluxuate between .63 and .65. There is a way to compensate for it in ECMLink, but since i didnt do it that way Im not going to pretend to that i know how to do it. I have a 6bolt and a lot less stuff in my engine bay so it was just easy for me to get to it. I dont remember the accessibility with a fully loaded 7-bolt bay.

I dont have enough knowledge ro know if its associated with ur o2 problem, but it needs to be adjusted regardless.

Haha yeah, the 2g tps has that IPS for idle the 1Gs don’t, so the FSM adjustment seems like a pita but I’ll give that a shot first, if they doesn’t work I’ll just do it with ECMlink and figure out how to simulate

 

[SynthRocker]

So I tried to adjust the TPS the factory way, with the proper feeler gauge in the throttle set stop screw and it simply wouldn’t get hit a spot where there was no continuity. Pissing me off

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

Also after not being able to adjust TPS the FSM way I did it with ECM link and just enabled ips simulation. Idle is still rough without deadtime on stock injectors (which according to the calculator is wrong) also seems my o2 is bad, no continuity between pin 3 and 4 (if I’m looking at the correct pins as there’s no good pin readout for connector side o2 to see if I’m ohming 1-2 or 3-4

Trying to post logs but apprantly I can’t on mobile will post before after adjusting deadtime when I get home

 

[SynthRocker]

Smashing my head over here.

My o2 is fine, it cycles through volts when manually tested (jumping the wires, revving engine) I adjusted the tps to .63 and simulated idle. Everything I do this and start the car the tps stays at .63 until I rev than it goes to .65 and stays, and ecmlink shows my o2 not cycling just staying at .2v. Losing my mind here.

Idles rough until I had 800usecs of global deadline on stock injectors. Wtf

log was adjusted tps to .63 after a rev it stayed at .65 wont go down, o2 wont cycle, even though its working, and no global deadtime on injector running super rough.

why is adding deadtime of 800usecs to injectors smoothing out the idle

 

[b00zt3d spyd3r]

If i remember correctly for a car on a maf you need to select mdp on the pin assignments for ecmlink functions (ecu side) manifold pressure (map) drop down. Might not fix your issues but may help.currently looks like its not measuring maifold pressure.

 

[SynthRocker]

If i remember correctly for a car on a maf you need to select mdp on the pin assignments for ecmlink functions (ecu side) manifold pressure (map) drop down. Might not fix your issues but may help.currently looks like its not measuring maifold pressure.

Did that, first picture is correcting that. Other picture is ecmlinks “stock” settings file from their site which make no sense.. I don’t have those GM sensors.

I really wish there was an easy way to set settings back to stock so I know for sure I’m starting at square one

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

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

 

[SynthRocker]

new log got the o2 to finally start switching, mdp is selecting, still a rough idle with no deadtime in injectors, looks to be going intoi close loop though. revved a few times at throttle body and it stayed running rough at idle

 

[b00zt3d spyd3r]

so you are completely stock? stock o2 sensor in front bung? i like to run the stock narrow band sensor up front and my wideband to the rear o2 input so theres no need to simulate narrow band. i still dont see any movement in your o2 until you revved a bit.

 

[SynthRocker]

so you are completely stock? stock o2 sensor in front bung? i like to run the stock narrow band sensor up front and my wideband to the rear o2 input so theres no need to simulate narrow band. i still dont see any movement in your o2 until you revved a bit.

Besides ECMlink, only mods are intake, deleted EGR, deleted charcoal canister, FPR solenoid, 2.5 in exhaust after o2 housing. I have a bung welding in a little after the 02 housing ill be putting my wideband in so i can still run the stock o2 sensor.

I got it all running fairly smooth without any deadtime in the injectors, still a bit rough but it smoothed out. only issue im seeing is it seems to running a bit rich at idle, AFratioest is hovering about 11.5 at idle. This log is the best ive got it to run so far withoout deadtime, and its very drivable, few revs in this log but this shows the o2 now switches, and tps is at .63 (well, can be, looks like I need to rebuild the TB, seems to be sticking a bit, I can use my fingers to push the spring to the butterfly to .63 and itll stay so for sure sticking because sometimes itll go back to .63 after taking foot off throttle )

 

[b00zt3d spyd3r]

make that switch, Make sure wideband is set up correctly and wired into rear o2. Your fuel trims are waaaaay off. Im guessing because the o2 sensor isnt correct. Start with the o2 sensor and make another log.

 

[SynthRocker]

make that switch, Make sure wideband is set up correctly and wired into rear o2. Your fuel trims are waaaaay off. Im guessing because the o2 sensor isnt correct. Start with the o2 sensor and make another log.

Hmmm the o2 seemed to be cycling fine, really think there could be an issue with it? I’ll get the wideband in and try simulation before I get a new one?

 

[b00zt3d spyd3r]

Oh my bad I thought you said you still had the wideband in the front o2 spot.