18:1 AFR at Idle!?
Thread Starter
Evolving Member
iTrader: (8)
Joined: Jul 2003
Posts: 449
Likes: 0
From: Clarkston, MI
18:1 AFR at Idle!?
Okay, I just installed my WB this weekend. It seems to work just fine, and logging in EvoScan works perfectly too. I'm getting pretty good numbers, except at idle. I sometimes see 17:1, but mostly it's closer to 18:1. I know there was a thread on this recently, but I can't seem to find it.
Is the only really likely problem an exhaust leak? Or could an intake leak also cause this? I am planning on doing a boost leak test later in the week to confirm the intake side, but I'm not sure if that could even be the problem.
On the exhaust side, I know for certain everything is all bolted up tight, however, ever since I installed my Megan DP, my exhaust note during boost has been very different. You know that slight 'chirp chirp chirp' of the exhaust on an old VW Bus or VW Beetle? It's kinda like that, but a bit more subtle. I'm wondering if exhaust is leaking out the flex joint mesh material. If so, should I be concerned about it? Could this cause my 18:1 idle AFR's?
It just seems to me that even at idle, the exhaust pipe should be slightly higher pressure than atmosphere, so any leakage would go OUT the pipe (not sucking air IN). Now on the intake side, at idle your getting vaccuum, which makes me think there may be some unmetered air getting into the system. Which is more likely?
Rob
Is the only really likely problem an exhaust leak? Or could an intake leak also cause this? I am planning on doing a boost leak test later in the week to confirm the intake side, but I'm not sure if that could even be the problem.
On the exhaust side, I know for certain everything is all bolted up tight, however, ever since I installed my Megan DP, my exhaust note during boost has been very different. You know that slight 'chirp chirp chirp' of the exhaust on an old VW Bus or VW Beetle? It's kinda like that, but a bit more subtle. I'm wondering if exhaust is leaking out the flex joint mesh material. If so, should I be concerned about it? Could this cause my 18:1 idle AFR's?
It just seems to me that even at idle, the exhaust pipe should be slightly higher pressure than atmosphere, so any leakage would go OUT the pipe (not sucking air IN). Now on the intake side, at idle your getting vaccuum, which makes me think there may be some unmetered air getting into the system. Which is more likely?
Rob
where is your WB sensor? if there a leak between the exhaust manifold and where you WBo2 is mounted that will read lean.
Also if oyu have a vaccume after the MAF that could cause high A/F @ idle also but that usually translates into a rich condition while in boost... best bet is to do a boost leak test.
Also if oyu have a vaccume after the MAF that could cause high A/F @ idle also but that usually translates into a rich condition while in boost... best bet is to do a boost leak test.
Thread Starter
Evolving Member
iTrader: (8)
Joined: Jul 2003
Posts: 449
Likes: 0
From: Clarkston, MI
It's an AEM UEGO. I followed the AEM instructions to build a serial cable which runs through a Serial-USB converter into my laptop. The numbers in HyperTerminal and EvoScan exactly match the digital display on the gauge.
Ok I see what you mean
That is strange to read 18:1 at idle while at a dead stop.
I use a Zeitronix ZT2 and it will sometimes go as lean as 21:1 while
cruising real slow. Once at idle, it will settle back down to perfect a 14.7:1 stoich.
In your first post, you mentioned doing a leak test. IMO I would do a complete
leak test on the intake and and hard piping circuit.
Here is one of the most informative threads of the century regarding a very
illusive boost leak that becomes very obvious once some attention is given to it.
https://www.evolutionm.net/forums/sh...tle+body+seals
I'm not saying that's your problem for sure, it just seems that small boost leaks cause big problems.
In any event a leak test would be a good place to start.
That is strange to read 18:1 at idle while at a dead stop.
I use a Zeitronix ZT2 and it will sometimes go as lean as 21:1 while
cruising real slow. Once at idle, it will settle back down to perfect a 14.7:1 stoich.
In your first post, you mentioned doing a leak test. IMO I would do a complete
leak test on the intake and and hard piping circuit.
Here is one of the most informative threads of the century regarding a very
illusive boost leak that becomes very obvious once some attention is given to it.
https://www.evolutionm.net/forums/sh...tle+body+seals
I'm not saying that's your problem for sure, it just seems that small boost leaks cause big problems.
In any event a leak test would be a good place to start.
Trending Topics
Been there, done that!
Are you using a simulated front O2 off your wideband? I tried to do it and I found the signal to be too lazy. This would cause wild swings at idle but good readings at cruise.
I replaced the narrow band signal back to the ECU and my problem was solved.
If you are not simulating the front O2, it may be going bad.
my 2 cents
HiVoltEVO8
Are you using a simulated front O2 off your wideband? I tried to do it and I found the signal to be too lazy. This would cause wild swings at idle but good readings at cruise.
I replaced the narrow band signal back to the ECU and my problem was solved.
If you are not simulating the front O2, it may be going bad.
my 2 cents
HiVoltEVO8
Rob,
It is most likely a leak at your downpipe. That will cause a lean reading at idle, but not at WOT or even cruise, depending on the leak size.
I had a leak on my old car before at the downpipe and at idle it would read around 17-18:1. WOT wasn't affected. The problem ended up being my downpipe bolts started to back out. After I tightened them back up, everything was back to normal.
So, look around for leaks, especially since the last thing you did was replace the DP.
A leak on the intake side shouldn't change your idle reading because the ECU will just use the fuel trims to adjust back to a stoich idle, unless the trim is already maxed out.
Eric
It is most likely a leak at your downpipe. That will cause a lean reading at idle, but not at WOT or even cruise, depending on the leak size.
I had a leak on my old car before at the downpipe and at idle it would read around 17-18:1. WOT wasn't affected. The problem ended up being my downpipe bolts started to back out. After I tightened them back up, everything was back to normal.
So, look around for leaks, especially since the last thing you did was replace the DP.
A leak on the intake side shouldn't change your idle reading because the ECU will just use the fuel trims to adjust back to a stoich idle, unless the trim is already maxed out.
Eric
Thread Starter
Evolving Member
iTrader: (8)
Joined: Jul 2003
Posts: 449
Likes: 0
From: Clarkston, MI
Well, I finally solved my idle problem. Like everyone suggested, it was indeed an exhaust leak. My Megan downpipe is custom welded to my Magnaflow hi-flow cat (long story.. the shop didn't have any 3" flanges in stock, so I sacrificed the cheap DP flange instead of the pricey CBE). Anyway, the guy left a small pin hole at the very top, just before the WBO2 bung.
I temporarily put some metal foil tape over it and my AFR at idle dropped down to 14.8-15.
Thanks to everyone here who made suggestions. I still suspect flex pipes leak under WOT/boost, but I kind of like the exhaust note anyway!
I temporarily put some metal foil tape over it and my AFR at idle dropped down to 14.8-15.
Thanks to everyone here who made suggestions. I still suspect flex pipes leak under WOT/boost, but I kind of like the exhaust note anyway!
Congrats on the fix.
Good thread, i might need to check for an exhaust leak i'm not sure. My AEM WB is never at 14.7, it always changed fast between 14.7 and 15.2. At WOT i'm good, a little on the rich side (yes im tuned).
Is 14.7 to 15.2 at idle normal...it's never steady.
Good thread, i might need to check for an exhaust leak i'm not sure. My AEM WB is never at 14.7, it always changed fast between 14.7 and 15.2. At WOT i'm good, a little on the rich side (yes im tuned).
Is 14.7 to 15.2 at idle normal...it's never steady.
Evolving Member
Joined: Dec 2006
Posts: 276
Likes: 0
In addition to an exhaust leak, the other reason you will often see a lot of widebands read lean near idle is miss fires (rich idle, or temps outside of optimum plug range because of intake mods).
A miss fire is a puff of gas where the lambda is essentially infinite (O2, no HCs). So you have a spike going full scale. But most instruments use a PID loop to drive the sensor, so they cannot respond fast enough to read the spike as a spike. Instead, they give a long hump (typically 1-2 seconds) of artificially lean readings. If you are periodically getting additional missfires, the humps smooth out into a seriously lean shifted reading.
Yikez: Closed loop, the ECU will typically 'dither' around 14.7. Slightly leaner is to be expected. That is when the Cat (if it is present) picks up O2 to do its work. However .5 AFR is a bit large. It could be a leak, or it could be your instrument. The chemistry of the UEGO sensor is different on each side of stoichiometric. Normally the calibration resistor in a Bosch sensor is trimmed on the lean side, so instruments that use it tend to get more accurate as you move closer to that point. However, the curve of the sensor also changes over time, so .5 AFR error on one side of stoich or the other is not at all unusual for a current based design.
-jjf
A miss fire is a puff of gas where the lambda is essentially infinite (O2, no HCs). So you have a spike going full scale. But most instruments use a PID loop to drive the sensor, so they cannot respond fast enough to read the spike as a spike. Instead, they give a long hump (typically 1-2 seconds) of artificially lean readings. If you are periodically getting additional missfires, the humps smooth out into a seriously lean shifted reading.
Yikez: Closed loop, the ECU will typically 'dither' around 14.7. Slightly leaner is to be expected. That is when the Cat (if it is present) picks up O2 to do its work. However .5 AFR is a bit large. It could be a leak, or it could be your instrument. The chemistry of the UEGO sensor is different on each side of stoichiometric. Normally the calibration resistor in a Bosch sensor is trimmed on the lean side, so instruments that use it tend to get more accurate as you move closer to that point. However, the curve of the sensor also changes over time, so .5 AFR error on one side of stoich or the other is not at all unusual for a current based design.
-jjf
But most instruments use a PID loop to drive the sensor, so they cannot respond fast enough to read the spike as a spike. Instead, they give a long hump (typically 1-2 seconds) of artificially lean readings. If you are periodically getting additional missfires, the humps smooth out into a seriously lean shifted reading.
-jjf
-jjf
http://www.bridgeanalyzers.com/900345_auto_desc.php4
to get a base reading would be worth while.
Imagine having something like that for real time logging
In addition to an exhaust leak, the other reason you will often see a lot of widebands read lean near idle is miss fires (rich idle, or temps outside of optimum plug range because of intake mods).
A miss fire is a puff of gas where the lambda is essentially infinite (O2, no HCs). So you have a spike going full scale. But most instruments use a PID loop to drive the sensor, so they cannot respond fast enough to read the spike as a spike. Instead, they give a long hump (typically 1-2 seconds) of artificially lean readings. If you are periodically getting additional missfires, the humps smooth out into a seriously lean shifted reading.
Yikez: Closed loop, the ECU will typically 'dither' around 14.7. Slightly leaner is to be expected. That is when the Cat (if it is present) picks up O2 to do its work. However .5 AFR is a bit large. It could be a leak, or it could be your instrument. The chemistry of the UEGO sensor is different on each side of stoichiometric. Normally the calibration resistor in a Bosch sensor is trimmed on the lean side, so instruments that use it tend to get more accurate as you move closer to that point. However, the curve of the sensor also changes over time, so .5 AFR error on one side of stoich or the other is not at all unusual for a current based design.
-jjf
A miss fire is a puff of gas where the lambda is essentially infinite (O2, no HCs). So you have a spike going full scale. But most instruments use a PID loop to drive the sensor, so they cannot respond fast enough to read the spike as a spike. Instead, they give a long hump (typically 1-2 seconds) of artificially lean readings. If you are periodically getting additional missfires, the humps smooth out into a seriously lean shifted reading.
Yikez: Closed loop, the ECU will typically 'dither' around 14.7. Slightly leaner is to be expected. That is when the Cat (if it is present) picks up O2 to do its work. However .5 AFR is a bit large. It could be a leak, or it could be your instrument. The chemistry of the UEGO sensor is different on each side of stoichiometric. Normally the calibration resistor in a Bosch sensor is trimmed on the lean side, so instruments that use it tend to get more accurate as you move closer to that point. However, the curve of the sensor also changes over time, so .5 AFR error on one side of stoich or the other is not at all unusual for a current based design.
-jjf