The same day 2 loggers different afr readings???

[Greg1987]

Hello, Might be a stupid question but I was driving my fully tuned evo 8 to the store and I did logger and saved it. On the way back from the store I did same gear pull and logged it. To my surprise I noticed on my afr gauge I have different afr (leaner) than I had 15 min earlier. Same fuel, same temperature, same iat, same ect, very similar Injector actual Pw but afr different by around 0.4-0.5 point. And it is different in the whole pull not only in some specific areas. Whenever I was 11.9 at peak I am 12.4 and whenever I was 11.4 at full boost I am 11.9. First I thought failing sensor, but I have 2 afr gauges in car and they both showing same readings so this is out of the question. Can anyone help please. I noticed this before but I was always blaming it for maybe colder weather etc, but those 2 loggers are basically maybe 20 min apart so nothing changed but afr. Also I would like to add that I parked the car over night and next day in the morning I was driving to my parents house afr was again on richer side 11.4-11.5 on wot (that way was tuned) but when I was coming back from parents ( I was there 30 min) afr was 11.9-12.0 on wot. Again same fuel in tank I did not stopped to put fuel so same e85 content in tank, same temperature etc. Please help!!!

[Greg1987]

[Greg1987]

Nobody can help? Look like it is consistent problem. My tuner spend 2 days trying to figure it out and we were comparing both loggers and there is nothing in tune that can do this. Same iat same ect same fuel same outside temperature etc. What my tuner told me he has a feeling that this is something with injector timing and ecu sometimes firing on the second half of the engine for unknown reason. we did couple test and we changed injector timing but only at 500 btdc car runs smooth, any other numbers runs rough. Can someone help??

[Laminar]

Sharing log files and the tune that was active at the time would be helpful.

[Confused]

What wideband controllers?

 

I experienced this kind of weird massive swing from one day to the next, with no reasonable explanation. 

 

Turns out it was the analogue output of the AEM wideband controller that just couldn't decide what to do.

 

Switched it out for a Link CAN Lambda and the values in the ECU were spot on and totally repeatable.

 

The non X-Series AEM wideband controller is awful for it's analogue output. You'll find plenty of other posts here with people having similar issues.

[Adamw]

A screenshot is not going to give you a diagnosis.  Attach the tune, and logs, and some information about what we are looking at - what engine, what car, what ecu, what injectors? 

[Greg1987]

 Here is tune and 2 logs. Car was tuned not long time ago with link g4x ecu. It is 2005 lancer evo 8 (4g63). FIC 2150cc injectors, bigger turbo bigger cams. I notice afr change but never during driving. I can drive for hours and always is the same on full boost pull. The problem: I start it in the morning I let it warm up and while driving on wot (full boost pull) afr is 11.3-11.5 (this afr was tuned at). Then I go to the store for 15-20 min and on my way back when car is also fully warm afr on wot is 11.9-12.0. But when I let it seat overnight the next day is again good around 11.3-11.5 on Wot. My tuner was screeching his head for last 2 days  and he told me it must be something with injector timing because the only values the car runs good is around 500 btdc. He told me that on default 320 btdc it doesn't want to run smooth.  Tuner checked tune and compared loggs multiple times and nothing change but Air fuel readings, so he told me to reach for help here. Whole engine is brand new so every single sensor is brand new and ecu was installed at the same time. Everything has 1k miles. Afr gauge is aem X series with brand new sensor and readings on gauge are the same as on the log. Please help! Thanks!!!

new eng e74 28psi.pclx PC Datalog - 2024-06-3 8;19;25 pm3rdgoodafr.llgx PC Datalog - 2024-06-3 9;08;52 pmleaner.llgx

2 hours ago, Confused said:

What wideband controllers?

 

I experienced this kind of weird massive swing from one day to the next, with no reasonable explanation. 

 

Turns out it was the analogue output of the AEM wideband controller that just couldn't decide what to do.

 

Switched it out for a Link CAN Lambda and the values in the ECU were spot on and totally repeatable.

 

The non X-Series AEM wideband controller is awful for it's analogue output. You'll find plenty of other posts here with people having similar issues.

 

The wideband is Aem x- series and brand new sensor. The numbers on the gauge and numbers on logs are the same. Just sometimes car has different afr at full boost. Sometime is dangerous lean around 12.1-12.2 but was tuned to be 11.3-11.5. I notice the difference only at full boost pulls not on cruse.

[Greg1987]

Anyone have any idea what to do so the afr is steady?? It is happening every time I shut the car off for 30-40 min and drive again. Is it possible that this is any glitch in ecu like my tuner is thinking it is?

[Greg1987]

Ok guys here is update. I spend with my tuner over 80 hours to try to find the issue. We ruled out the car by installing stock ecu and afr is always the same when car is restarted etc. We put the link g4x back in evo and my tuner found that ECU is syncing in the wrong half of the engine cycle when is restarted couple of times or if its shut off for 15 min and restarted, and injector timing is wrong that way. We tried every possible way to make it work but my tuner said it is syncing problem from ecu. Another member here had exactly same problem with exactly same car here on forum and link told him that it is syncing issue and link will release firmware update but from what I was reading firmware update did not helped and member moved to different ecu. Would you guys be able too look into this and possible fix it with firmware update???? It is crazy frustrating to the point that we almost hurt the engine because it was set up on the leaner side and when afr changed it dangerous lean and started misfiring and we almost lifted the head. Please help!!!!! Thanks!

[Rossobianconero]

2 hours ago, Greg1987 said:

Ok guys here is update. I spend with my tuner over 80 hours to try to find the issue. We ruled out the car by installing stock ecu and afr is always the same when car is restarted etc. We put the link g4x back in evo and my tuner found that ECU is syncing in the wrong half of the engine cycle when is restarted couple of times or if its shut off for 15 min and restarted, and injector timing is wrong that way. We tried every possible way to make it work but my tuner said it is syncing problem from ecu. Another member here had exactly same problem with exactly same car here on forum and link told him that it is syncing issue and link will release firmware update but from what I was reading firmware update did not helped and member moved to different ecu. Would you guys be able too look into this and possible fix it with firmware update???? It is crazy frustrating to the point that we almost hurt the engine because it was set up on the leaner side and when afr changed it dangerous lean and started misfiring and we almost lifted the head. Please help!!!!! Thanks!

I know that is not really the best way but, there isn't any trigger wheels that can be use to fix the problem? of course the real and proper fix would be the ecu not having this problem.

[Vaughan]

maybe take several trigger scopes and attach them here to start with so we can see if your triggers are different to other EVO 8s, there has also been another 3 major firmware updates since what is in your ECU so not a bad idea to update that too.

[Adamw]

5 hours ago, Greg1987 said:

We put the link g4x back in evo and my tuner found that ECU is syncing in the wrong half of the engine cycle when is restarted couple of times or if its shut off for 15 min and restarted, and injector timing is wrong that way.

How is he determining this?  Does he have a scope capture of cam sensor Vs injector pulse?  My daily driver is an Evo7 with direct spark ignition so I can quite confidently say there is no sync issue with this trigger mode.    

 

[Greg1987]

1 hour ago, Adamw said:

How is he determining this?  Does he have a scope capture of cam sensor Vs injector pulse?  My daily driver is an Evo7 with direct spark ignition so I can quite confidently say there is no sync issue with this trigger mode.    

 

Yes he did scope capture of cam sensor vs injector pulse, at least that's what he told me. He spent 80+ hours on my car trying to find this problem. I think your direct spark ignition might make a big role of this, I have wasted spark maybe this is the difference? Today my tuner changed Injection mode to Multi point group and Rate 2 and he tuned fuel map so the car is driveable. We took it out, I was restarting it multiple times, then we shut it off for 30 min and always afr was the same where on Sequential mode it would change afr already. He told me that is because on Multi point group Injector timing doesn't work and that's where the issue is. Any idea what to do to have steady afr every time I restart the car on sequential mode?

4 hours ago, Vaughan said:

maybe take several trigger scopes and attach them here to start with so we can see if your triggers are different to other EVO 8s, there has also been another 3 major firmware updates since what is in your ECU so not a bad idea to update that too.

Here are couple triggers. I have some news. Today my tuner changed injection mode to multi point group and rate 2, he tuned fuel map so we could drive the car. We restarted the car multiple times and we did couple 20 min breaks between driving and afr was always the same where on sequential mode it would change afr already. By the end of the day we notice your message about new firmwares, so he put the car back to sequential mode and tuned fuel map to be 11.5 at wot. I will test it tomorrow and let you know if its still changing afr. But why on multi point never changed afr it was the same after every restart. He told me because injector timing does not work with this mode and that's where the problem is. I don't know please help!!!

Trigger Scope - 2024-06-9 9;55;07 pm trigger1.llgx Trigger Scope - 2024-06-9 9;55;30 pmtrigger2.llgx Trigger Scope - 2024-06-9 9;55;46 pmtrigger3.llgx

[Vaughan]

13 minutes ago, Greg1987 said:

I think your direct spark ignition might make a big role of this, I have wasted spark maybe this is the difference?

Ignition mode doesn't affect the trigger mode but it does mean that if Adam's car was miss-syncing then it wouldn't actually start, just keep on cranking.

14 minutes ago, Greg1987 said:

Here are couple triggers.

You appear to have quite a lot of noise on Trigger 1 which could definitely cause issues so I would recommend looking into that.

[Greg1987]

11 minutes ago, Vaughan said:

Ignition mode doesn't affect the trigger mode but it does mean that if Adam's car was miss-syncing then it wouldn't actually start, just keep on cranking.

You appear to have quite a lot of noise on Trigger 1 which could definitely cause issues so I would recommend looking into that.

All sensors are brand new. This engine has 1k miles with all new sensors brackets etc.. Is Trigger 1 camshaft or crankshaft sensor? I did trigger scope while car was running is that correct?? Why on multi point group the car was not doing this, it was not changing afr? What my tuner told me the problem is around injector timing. Please help to find this issue.

[Vaughan]

Crankshaft sensor, it isn't necessarily due to the sensor itself, could be due to the trigger wheel, wiring or other reasons.

Multi point injection will inject at the same time in both halves of the engine cycle (like waste spark does for ignition) and so it doesn't matter if you are miss-syncing when you are running multi-point group but it does if you are using sequential.

The injection timing being wrong would be because the ECU thinks the engine is in the other half of the engine cycle, this affects the timing of everything engine position related such as the injection timing and ignition timing. You don't see the effect of the ignition timing being 360 out because waste spark fires on both halves of the cycle but you do see the effect of the fuel injection being 360 out when using sequential injection. 

[Greg1987]

4 minutes ago, Vaughan said:

Crankshaft sensor, it isn't necessarily due to the sensor itself, could be due to the trigger wheel, wiring or other reasons.

Multi point injection will inject at the same time in both halves of the engine cycle (like waste spark does for ignition) and so it doesn't matter if you are miss-syncing when you are running multi-point group but it does if you are using sequential.

The injection timing being wrong would be because the ECU thinks the engine is in the other half of the engine cycle, this affects the timing of everything engine position related such as the injection timing and ignition timing. You don't see the effect of the ignition timing being 360 out because waste spark fires on both halves of the cycle but you do see the effect of the fuel injection being 360 out when using sequential injection. 

Thank you so much for explaining it, but what I can do to fix the afr changing on sequential mode? 2 days ago my tuner installed stock ecu in the car and problem was not there and afr was always the same after restarts etc. Is it something that link doesn't like or what is happening. My tuner is scratching his head and he told me that there is something wrong about injector timing but he can't figure it out.  I did the trigger scope while car was on idle running is that correct? Thanks a lot!

[Vaughan]

24 minutes ago, Greg1987 said:

Thank you so much for explaining it, but what I can do to fix the afr changing on sequential mode?

Fix the reason it is miss syncing.

24 minutes ago, Greg1987 said:

2 days ago my tuner installed stock ecu in the car and problem was not there and afr was always the same after restarts etc

Stock ECU may be less affected by noise on the triggers than the Link as the stock ECU only ever has to deal with that specific sensor type and trigger pattern.

26 minutes ago, Greg1987 said:

I did the trigger scope while car was on idle running is that correct? Thanks a lot!

yes although getting a scope while cranking before it fires would show the signals while the ECU is actually trying to sync.

[Greg1987]

15 minutes ago, Vaughan said:

Fix the reason it is miss syncing.

Stock ECU may be less affected by noise on the triggers than the Link as the stock ECU only ever has to deal with that specific sensor type and trigger pattern.

yes although getting a scope while cranking before it fires would show the signals while the ECU is actually trying to sync.

 

I will do the trigger scope tomorrow again before it fires. How do I fix the reason why it is miss syncing if I don't know what to fix? This engine has new trigger plate new crank sensor, every possible sensor is new, I build this engine  with new crankshaft/rods/pistons new cams, valves etc . Every possible part in this engine is brand new even bare block is brand new( I am engine builder) . I confirmed timing with timing lamp etc. Also if that would be something in the car wouldn't the problem be all the time with miss syncing not only at restart? Any idea what I can check to fix that?

[Adamw]

I dont see any good explanation for your apparent problem from the logs provided so far.  Im not sure if there is actually noise on the trigger signal as the built-in scope doesnt really have adequate resolution or sample rate to capture it.  

I think we need to first confirm that it is actually mis-syncing, this would require a standalone scope connected to an injector and the cam sensor so you can confirm if the injector is always lined up with the long cam tooth or short cam tooth or if it does in fact randomly change at start up as you suggest.  If it does randomly change then the only explanation that I could fit to the observation would be a noise related issue.  But since we dont see any trigger errors in your logs at high RPM/Boost when electrical noise would usually be the worst, I have some doubts that there is a big noise issue.  It would have to be something that only occurred at startup such as a voltage spike for the starter motor or similar. 

 

Actually, one other method I just thought of that you may be able to use to confirm injector timing/mis-sync.  Quite often a timing light will trigger off an injector wire, so if you have a timing light, try clipping it around the injector wires to see if you get a flash.  You can try both wires together and each wire separately.  If you can get a timing light flash from the injector wires, then the way to use this would be to remove the top cam belt cover and paint a mark on one of the cam pulleys.  When the car is running good, point the timing light at the cam wheel and take note of where the painted mark is.  Next time it starts with the lean AFR then connect the timing light again and check if the paint mark is in roughly the same place or not.  You dont need to be accurate as if it did mis-sync then the paint mark would be on the completely opposite side of the wheel, 180 out.  

    

[Greg1987]

1 hour ago, Adamw said:

I dont see any good explanation for your apparent problem from the logs provided so far.  Im not sure if there is actually noise on the trigger signal as the built-in scope doesnt really have adequate resolution or sample rate to capture it.  

I think we need to first confirm that it is actually mis-syncing, this would require a standalone scope connected to an injector and the cam sensor so you can confirm if the injector is always lined up with the long cam tooth or short cam tooth or if it does in fact randomly change at start up as you suggest.  If it does randomly change then the only explanation that I could fit to the observation would be a noise related issue.  But since we dont see any trigger errors in your logs at high RPM/Boost when electrical noise would usually be the worst, I have some doubts that there is a big noise issue.  It would have to be something that only occurred at startup such as a voltage spike for the starter motor or similar. 

 

Actually, one other method I just thought of that you may be able to use to confirm injector timing/mis-sync.  Quite often a timing light will trigger off an injector wire, so if you have a timing light, try clipping it around the injector wires to see if you get a flash.  You can try both wires together and each wire separately.  If you can get a timing light flash from the injector wires, then the way to use this would be to remove the top cam belt cover and paint a mark on one of the cam pulleys.  When the car is running good, point the timing light at the cam wheel and take note of where the painted mark is.  Next time it starts with the lean AFR then connect the timing light again and check if the paint mark is in roughly the same place or not.  You dont need to be accurate as if it did mis-sync then the paint mark would be on the completely opposite side of the wheel, 180 out.  

    

Thank you Adam very much, this is actually a great idea to check that with timing light. I will definitely check that and post results most likely tomorrow. I would like to mention that I installed newest firmware and this did not helped.  Thanks a lot!!!

[Greg1987]

On 6/10/2024 at 7:13 PM, Adamw said:

I dont see any good explanation for your apparent problem from the logs provided so far.  Im not sure if there is actually noise on the trigger signal as the built-in scope doesnt really have adequate resolution or sample rate to capture it.  

I think we need to first confirm that it is actually mis-syncing, this would require a standalone scope connected to an injector and the cam sensor so you can confirm if the injector is always lined up with the long cam tooth or short cam tooth or if it does in fact randomly change at start up as you suggest.  If it does randomly change then the only explanation that I could fit to the observation would be a noise related issue.  But since we dont see any trigger errors in your logs at high RPM/Boost when electrical noise would usually be the worst, I have some doubts that there is a big noise issue.  It would have to be something that only occurred at startup such as a voltage spike for the starter motor or similar. 

 

Actually, one other method I just thought of that you may be able to use to confirm injector timing/mis-sync.  Quite often a timing light will trigger off an injector wire, so if you have a timing light, try clipping it around the injector wires to see if you get a flash.  You can try both wires together and each wire separately.  If you can get a timing light flash from the injector wires, then the way to use this would be to remove the top cam belt cover and paint a mark on one of the cam pulleys.  When the car is running good, point the timing light at the cam wheel and take note of where the painted mark is.  Next time it starts with the lean AFR then connect the timing light again and check if the paint mark is in roughly the same place or not.  You dont need to be accurate as if it did mis-sync then the paint mark would be on the completely opposite side of the wheel, 180 out.  

    

Hello, Adam. Ok I did the test with timing light. I set the timing light to the injector 1 and I started the car I took a note where the painted mark is on cam gear. Then after some restart the afr changed by 0.5 leaner so I tested with timing light and it was in the exactly same spot as before. I also restarted the car multiple times and every time I was checking with timing light and mark was in the same spot on cam gear so it is Not Mis-syncing but afr is changing. I also changed starter and alternator for another good one and same issue. It is also not failing afr sensor because I changed sensor for new one and to confirm that it is changing afr we set the wot afr to be 11.0-11.1 so it is easier to remember etc. and we were driving the car for over 1 hour and every single pull afr was 11.0-11.1 WOT. Then we shut the car off and restart it after 10 min we did multiple wot pulls over 30 wot pull and every single time afr was 11.5-11.7 and we were driving it for over 1 hour again and afr was always 11.5-11.7 Wot , and you can feel that the car gained power when afr was 11.5-11.7 compare to 11.0-11.1 what we tuned for. So this gave us confirmation that the afr is changing and it is not some gauge malfunction or something. Also we connected second afr gauge to confirm. Iat temp and ect temp was exactly the same. Same weather outside same fuel in tank etc. Any idea what to do now? It is crazy weird issue, that only happened when car shut off for 10 min and restarted. Please give me some idea. Thanks!