05 Acura TL 3.2L misfire under load.

Vehicle came in knocking, replaced engine with reputable salvage yard. Ran great, even drove it down the road and back. I usually drive them about 20 mins after an engine or trans swap but this customer hadn't registered if yet so I didn't want to drive far. Anyway customer picks it up and calls back an hour later ckmplaing of a misfire so they brought back. Found a random misfire code along with misfire on 1,2,4,5. At idle it runs flawlessly, misfire monitor shows a few misfire when revved to 2k. Driving down the road how ever is a different story, under light to med load its a hard misfire with engine light flashing, wot it clears out alot but still not all the way.

My question is in regards to the fuel data, this car has AF Lamada sensors (wideband?) And seems to read at about a volt. I gave it a vac leak at idle and watched on scan tool, and then added brake cleaner. It seems that a rich condition is lower than 1 volt and lean is above one volt. Can anyone confirm this for me? Also short term fuel trim and long term fuel trim seems to be 1.00 being zero? Above 1.00 adding fuel, below 1.00 taking away?
I have watched the AF lamada data pid and at wot its around 0.93V. If im thinking correctly I'm not getting enough fuel?

Things I've already done, swapped plugs with new plugs another shop had just put in the old motor. I don't hear any vac leaks and map is about 20.5 inhg, that should be good? I'm not sure I understand these fuel trims enough to use for diag purposes. We did verify timing belt on used motor before install and timing was spot on. I have ran a relative compression test and all 6 cyl were nice and even. I tried scoping the primary ignition waveform but I'm honestly just not that experienced in doing that yet. I'm sort of lost in where to go from here and could use an idea where to go and look from here. Other things to note I noticed that cylinders 3 and 6 never misfire, no codes for them nor misfire counts. These 2 are also straight across from one another like such:

123
456

The intake is also a big cast tunnel ram with a plenum that kinda acts like a vaccum pocket I think.... also has a controllable valve between the sides of the intake in the top plenum portion. I will try and post a pic in the comments, it also has the throttle body on the back of the motor with cylinders 3 and 6. Not sure if this helps just thought it was kinda coincidental. Any advice please help lol

Anyway to verify fuel pressure?
I'm no Acura guy, but I would expect WOT throttle to have a lambda to be much richer than that (based on my GM experience wanting around .85 WOT with Power enrichment).
What does the lambda show when it is hard misfiring?

Kinda pain to check it but can be done, just trying to get in a direction before running needless tests. Can anyone verify .93 being a rich condition under wot? I will have to monitor that and report back on the lambda when misfiring.

Other things to note. I noticed this issue is only happening when the vehicle is cold. Once warm it runs like a champ, I monitored fuel trims today idle, 1500, 2500, and 3500
If im reading them correctly its adding about 15% total per bank no matter the rpm. I smoke tested the intake system today without any visible vac leaks. I did have alot of smoke in the crankcase tho.....
I then plugged off the valve cover to air intake hose. I blocked it off at the intake and the valve cover and then installed a vacuum gauge in the dip stick hole. I had about 6 inhg vac and lost vaccum when I uncapped the valve cover nipple for the pcv hose.

With this information I decided to change the plenum gaskets. There's more of a back storey as to why just the plenum and not the lower intake as well. Ya see in the course of diagnosing the knock in the last motor we attempted to adjust the valve in hope that was the noise. So the old motor had brand new valve cover and plenum gaskets. I chose to reuse them and figured that could be my issue.

Once back together the trims looked so much better just at idle that I decided to take it for a drive. At this time the vehicle had cooled off a good bit. It Once again was misfiring from light load and once warm went away.

I attempted Once again to do a primary ignition waveform on this engine with no luck. Not sure what I'm doing wrong. I have a snap on Zeus with an aes wave primary ignition 10:1 probe for snap on scanners, im setting my sweep to 100 ms and my scale to different settings from 50-400v. I've tried inverting the waveform as well setting a trigger. I will try and post up a pic of the waveform.

Someone please help I have an extra 3 days in this car and I have no direction in which way to go....

can you post the OBD freeze frame data for the misfire DTCs?

If it's the flat top intake with the plate, it's easy to get that gasket upside down and block some of the EGR passages, (or they carbon up on their own) directing all the EGR to the unblocked cylinders causing cylinder specific misfires.
Try unplugging the EGR valve and taking it for another drive.
If the misfire is gone then you've got some playing around to do in the intake.
If not, then... Something else I guess

I'll do my best to answer your questions! :silly:

It seems that a rich condition is lower than 1 volt and lean is above one volt. Can anyone confirm this for me?

I believe you're looking at the AF Lambda PIDs in the OEM data. It's not a voltage you're looking at, just a lambda value. 1.00 represents stoichiometric, lower is indeed rich, higher is lean.

Depending on the year and the scanner you're using, you may also see PIDs like AF B1/B2 Lambda CMD. That's the lambda value the ECM is trying to achieve.

Also short term fuel trim and long term fuel trim seems to be 1.00 being zero? Above 1.00 adding fuel, below 1.00 taking away?

Correct. Again, the trim PIDs may look different depending on the scanner in use. AF FB and AF FB AVE are Honda terms for short and long term fuel trim, but may not always get displayed that way.

I have watched the AF lamada data pid and at wot its around 0.93V. If im thinking correctly I'm not getting enough fuel?

I don't necessarily see a problem with that value. I had some difficulty finding known good data captures for the 3.2L engine. But all the other Honda/Acura known goods I could find showed WOT enrichment anywhere from .87 to .91. Yours isn't quite there? But it's still getting power enrichment. I wouldn't sweat it right now.

I attempted Once again to do a primary ignition waveform on this engine with no luck. Not sure what I'm doing wrong. I have a snap on Zeus with an aes wave primary ignition 10:1 probe for snap on scanners, im setting my sweep to 100 ms and my scale to different settings from 50-400v. I've tried inverting the waveform as well setting a trigger. I will try and post up a pic of the waveform.

You won't be able to get a primary ignition waveform out of these coils. These are a three wire design, which means there's no primary circuit outside of the coil for you to connect to.

Your only options for scoping these coils are coil current ramps and secondary ignition. Current ramping the coils might be fun, but because you don't have a dead misfire, I'm guessing it won't show you anything wrong. Secondary ignition testing Honda/Acura coils can be tough due to the coil design and the shielding built into the coil itself. You might have better luck with some alternative secondary testing methods:

www.scannerdanner.com/forum/lab-scope-di...buy-one.html?start=0

Noah's suggestion of unplugging the EGR valve is excellent. Would you be able to redo your relative compression test after sitting overnight? Before it's driven.

Thanks for the replies. So I'm ok with that being the power enrichment if you say that's normal considering the car moves great under wot. Once I noticed this being an issue only when the car was cold I immediately went for a vaccum leak. I have great vaccum on a gauge BTW. Should the af lambda sensor still switch from rich to lean like a normal o2 sensor?

I've never done a current ramp tests but did watch one of scanner Danner videos where he does it. Il do some more research on it tho. How do you know what a good waveform is for the current ramp. Like is it vehicle specific or pretty universal waveform to go by? I can also use this to check base ignition timing? Just use the falling edge? How can you do a secondary test on COP? I thought you could do primary tests on 2,3, and 4 wire coils?

While I had the plenum off I cleaned all the ports with brake cleaner. There seemed to only be one egr port at the back of the motor. I cleaned it the best I could. I'm not very informed on egr operation and the issues that come with it. I noticed that snap on doesn't have a guided component test for the cam sensor on this engine. I've also not been able to find a cam/crank relation waveform. I'd like to be able to verify valve timing without pulling the covers. We did check before install but ya never know.

Any input on my internal vaccum leak test? I did it just like in the book. I plan to redo it now since changing the plenum gaskets. If it comes back the same is it safe to assume the lower intake gaskets are leaking? I will also do another relative comp test while cold. Thanks guys for the help

I left out possibly a important part of all this, I didn't think is was relevant but it maybe. When trying to do my relative comp test the car wouldn't do a flood crank no start. So I unplugged all the coils. Yal might not believe me believe this but the car started and ran without any coil hooked up. Mind you it ran like absolute crap but did stay running until key was turned off and fuel pump quit running. We also ran a can of seafoam thru the induction system in hopes carbon build up was our issue. The car stayed running with the key off and out of the ignition switch, again like absolute crap but did run. It was Very odd, possible spark knocking? Could timing be off causing this? I would figure if so timjng would have to be way out but the car runs to good at idle with everything plugged in. Let me know what yall thing.

Thrashnasty13 wrote: Things I've already done, swapped plugs with new plugs another shop had just put in the old motor. I don't hear any vac leaks and map is about 20.5 inhg, that should be good?

If that's the MAP PID at idle no it ain't good. Manifold Absolute Pressure is the reverse of the gauge pressure you read with an old school. With absolute pressure zero is a perfect vacuum and atmospheric pressure is 30ish inHg depending on elevation and weather. So 20.5 in Hg MAP is only about 9.5 inHg of gauge vacuum.

Is it accurate to to subtract the the map inhg pid from the baro inhg pid to get engine vacuum?

Thrashnasty13 wrote: Should the af lambda sensor still switch from rich to lean like a normal o2 sensor?

Not like a normal O2, no. To be clear, wideband sensors do still switch across stoich, they just don't use a binary rich/lean signal.

I've never done a current ramp tests but did watch one of scanner Danner videos where he does it. Il do some more research on it tho. How do you know what a good waveform is for the current ramp.

Looking at enough of them, mostly. :silly: Paul has some fantastic videos about current ramping coils on YouTube. As an example, I took these ramps off a 2008 Nissan Altima 2.5L:



Coil on the left is lovely. Coil on the right is missing the first part of its ramp. Straight up line. Classic indicator of a shorted secondary winding in the coil itself. This also had a dead misfire at all times.

I can also use this to check base ignition timing? Just use the falling edge?

Together with a relative compression test, absolutely.

How can you do a secondary test on COP? I thought you could do primary tests on 2,3, and 4 wire coils?

Check out that thread I linked earlier. It's got a ton of different secondary probes, as well as some easier/cheaper methods of getting secondary waveforms. Keep in mind that reading secondary waveforms can be challenging. I'd suggest learning secondary testing on several known good cars first before using it as a diagnostic tool.

Any input on my internal vaccum leak test? I did it just like in the book. I plan to redo it now since changing the plenum gaskets. If it comes back the same is it safe to assume the lower intake gaskets are leaking?

Yeah, that test kinda confused me. :silly: I don't believe the crankcase and the intake share any gasket surfaces on this engine. It's not like domestic OHV engines where the lower intake caps off the valley and the crankcase. I'm not sure where you were getting that vacuum, unless there was something physically wrong with the valve cover itself (as the PCV valve is in the front valve cover).

Thrashnasty13 wrote: Is it accurate to to subtract the the map inhg pid from the baro inhg pid to get engine vacuum?

Yeah that'll give you the gauge intake vacuum. Just make sure the BARO is believable.

Matt T wrote:

Thrashnasty13 wrote: Is it accurate to to subtract the the map inhg pid from the baro inhg pid to get engine vacuum?

Yeah that'll give you the gauge intake vacuum. Just make sure the BARO is believable.

Agreed. I've seen the BARO PID lie due to a failed MAP sensor.

The BARO PID is updated every time the key is turned on, based on whatever the MAP sensor happens to be reading at the time. 2.8V from the MAP sensor represents atmospheric near sea level.

awesome info guys I'm going to try and attack this monster again tomorrow. I totally agree with you Tyler on the vaccum testing, I never knew the reason that causes crank case vaccum. Makes since for it to only be valid with the intake covering the valley. This engine doesn't share gasket surface with crankcase at all. Weird I have a vaccum tho.... should have pressure if anything correct?

Something else I thought about. The car sat for over a year until they had me fix it. And the gas is really low, what's the chances the fuel is just junk? Wouldn't make since of the symptoms I'm having. If bad fuel was the case I'd have misfire at wot? Not just partial throttle. It would explain my spark knocking issue

I feel like a moron! Lol it was a bad coil. It was only an issue under a load but not wot. Today while driving it I noticed number 2 had alot more misfires than the others, it made me think this may be a single cylinder misfire and since I had extra coils and still don't really know what I'm doing with ignition waveforms. I caved and swapped parts. Fixed it right up I ever verified the fix by reinstalling the bad coil in a different cylinder, misfire followed. I then did a current ramp test on a good and the bad coil and they were Identical, I assume because it was idling. Since this was only an under load misfire how would I could about troubleshooting this type of issue next time? Like what kind of test could I have done to see this?

Have test lead on problem coil while driving under load or when fault occurs

An adjustable spark tester is your friend in a case like this.
It's my preferred method for flushing out ignition coils that can't hang under load.