Honda Pilot - Honda Pilot Forums banner
21 - 39 of 39 Posts

· Registered
Joined
·
60 Posts
Discussion Starter · #21 ·
Nope, I believe the op has found the issue, the a/f sensor is bad. I was only saying that for a cat to have a very rich input and very lean output, it would have to have a leak drawing in fresh air. That's the only way for that situation to happen if both sensors were correctly reading. Unplugging or swapping the a/f sensors front to back was the correct diag step. When the sensor was unplugged it no longer can affect the trims and the pcm reverts to the base map with corrections applied. When the op did this his issue was resolved, proving the a/f sensor most likely is reading incorrectly causing the issue.
I hope you're right but we shall see. Wouldn't surprise me at all if something else besides this is wrong here but I should know on saturday. I'll clear the codes and data and see how the trims look once the new sensor is installed. Thanks for your thoughts.
 

· Registered
Joined
·
60 Posts
Discussion Starter · #22 ·
Well I put the new upstream 02 sensor in, and I'm still getting misfires at cold startup. It pretty much acts the same. So much for that diagnosis! Bank 2 sensor still reading rich, and seems to be really leaning towards a leaky injector or some kind of problem similar. So next step will be pulling off intake manifold etc, again, and measuring resistance across the fuel injectors. Already ordered in new o rings for the injectors and 2 injectors as well. Only way to run rich is either a bad sensor, not enough air or too much fuel. Again once the engine warms up, the flashing check engine misfire goes away. I will probably put a new injector on cylinder 4 and see what happens.

I did note a few things once the engine was up to normal operating temperature:
1) Bank 1 short term fuel trim runs about 1.08 (lean). When I give it some gas maybe up to 2500 rpms, the trim will dip down to 1.00. That means it goes from a little lean to perfect (slightly richer)
Bank 2 however does the opposite. It will run from .85 (rich) and goes UP to 1.00. So if a leaky fuel injector, dirty, whatever is dribbling fuel in there, it would look rich. But as the fuel and air increase with the throttle, it makes less of a difference than at idle. That's the theory anyway.
2) Bank 2 sensor runs rich regardless from start of cold engine when misfires happen, to 5 minutes later on a warm engine and no flashing CEL or even steady CEL. Also, downstream sensor will still read very lean, but can fluctuate. It seems to work a little better once warm, but never steady like bank 1. So I probably also have some type of exhaust leak on bank 2, or a bad downstream connector, maybe cat problem not sure yet.

I will continue to plod onward and post updates. Other thing I will probably do is pull all the timing covers off and look over my timing marks. I had been just viewing through the little peek holes but I want to be certain on this.
 

· Registered
Joined
·
60 Posts
Discussion Starter · #24 ·

· Registered
Joined
·
77 Posts
How do you like autoenginuity? What are you using it with? Laptop?
it’s more difficult to evaluate AFs than O2’s. How do you read them?
 
  • Like
Reactions: royalbiggster

· Registered
Joined
·
60 Posts
Discussion Starter · #28 ·
I like it, I've never used any of the really expensive stuff but it seems ok. With Honda module I can talk to honda sensors etc way better than your generic code reader. Like ABS sensors etc. Hard part is knowing what I'm looking at and what it means. Like accelerator position sensor A volts, wtf is that? Would be nice if I had a honda manual describing what all this means, and what are normal values. Would also be nice if a range was provided with the reading like some of the better tools provide. But better than generic data like I said. And yes I'm using a laptop with a usb cable/dongle that connects to odbd2 port.

I don't think AF sensors are more difficult, just different. A standard 02 sensor has voltage between 0 and 1, call .5 normal. If voltage approaches zero you have a lean condition. Approaching 1 is a rich condition.

AF sensor (wideband) can measure a bigger variance of lean and rich vs standard narrow band o2 sensor. I forget what the voltage means but there are values. But you also measure current in mA, and negative means rich, positive means lean. Fuel trim is what the computer is doing, and is measured kind of like current, but with a base value of 1 = stroichiometric. Anything above 1 is a positive fuel trim, which means the AF sensor is reading too much oxygen (a lean condition), so the computer is adding fuel. If the fuel trim is below 1, that means the AF sensor is reading a rich condition, too much fuel, so you have a negative fuel trim which means the computer is removing fuel. Then there's the lambda value, 1 being stoichiometric, above 1 lean, below 1 rich. Still learning about lambda but I think it's important.

Truth be told I'm still learning. Next tool is going to be a picoscope so I can start getting data that way. But whatever you use, if you can graph your o2 or af sensors and see how they respond to testing I think helps tremendously.
 

· Registered
Joined
·
60 Posts
Discussion Starter · #29 ·
Well I'm pretty sure I figured out why my initial diagnosis of a bad upper o2 sensor is wrong. I did a smoke test this morning, looking for exhaust leaks, and found this:
Automotive tire Wood Gas Auto part Automotive wheel system


That's the downstream O2 sensor, and it sure looks to me like it's plugged into an o2 extender or defouler, whatever you call those things. At some point somebody had some cat codes being thrown and fixed it with this. I'm not positive what effect this will have on a scan tool, but it sure has to mean you can't trust whatever the reading is.

Smoke test passed, I did it twice and ran 2nd one for 10 minutes, just to be sure enough smoke went up the system. No real leaks discovered, but will be buying some type of laser pointer to better see the smoke I think.

My next step is removing intake plenum and cap, I will be testing resistance on fuel injectors. Maybe even pulling them out and pressurizing system to see if they leak.
 

· Registered
Joined
·
1,148 Posts
If it's an injector issue, most likely it's a bad spray pattern instead of an electrically bad injector. If it were me I would run techron injector cleaner through the system. Buy 2 big bottles of it, first one put in at 1/4 tank or so, let idle for 30-45 min. Then go run it hard, multiple 20-70+ wot runs, run the rest of the tank close to empty. Then refill and pour the other bottle in, that has resolved 99% percent of spray pattern issues for me. In 20+ years with Honda I think I have only replaced 3 or 4 injectors (excluding direct injection injectors, which happens to also be a spray pattern issue).
 

· Registered
Joined
·
60 Posts
Discussion Starter · #31 ·
If it's an injector issue, most likely it's a bad spray pattern instead of an electrically bad injector. If it were me I would run techron injector cleaner through the system. Buy 2 big bottles of it, first one put in at 1/4 tank or so, let idle for 30-45 min. Then go run it hard, multiple 20-70+ wot runs, run the test of the tank close to empty. Then refill and pour the other bottle in, that has resolved 99% percent of spray pattern issues for me. In 20+ years with Honda I think I have only replaced 3 or 4 injectors (excluding direct injection injectors, which happens to also be a spray pattern issue).
I've used chevron techron cleaner before, it's good stuff. And you have great advice here, but I think I've got a bad injector on bank 2 up front, number 5. It's ohming out around 3, and the other five are 9.8s, 9.9s and 10. Manual I have says between 10-13 is valid. Sure wish I took the 5 minutes doing the valve job to ohm out my injectors :-(

I get this thing back together I will be running the techron through it as well :cool:
 

· Registered
Joined
·
77 Posts
What’s your freeze frame data for the misfire look like?
Does the downstream O2 have any bearing on AF control?
 
  • Like
Reactions: royalbiggster

· Registered
Joined
·
60 Posts
Discussion Starter · #33 ·
What’s your freeze frame data for the misfire look like?
Does the downstream O2 have any bearing on AF control?
Freeze frame data is on my first post. Downstream O2 does not have any bearing on upstream AF control (I believe). However, I was using the downstream o2 reading of very lean all the time, while the upstream AF sensor was reporting rich, to call upstream sensor a liar and bad, so I replaced it, incorrectly.
 

· Registered
Joined
·
60 Posts
Discussion Starter · #34 ·
I ohmed out the new injector as well, it curiously it was 9.9 so looks like all my injectors are good now. (I replaced the bad one) It's all buttoned up and ready to start, but I am waiting to let all the gas evaporate on top of the engine. At least that's what I'm telling myself to stall this a bit :)

I'm expecting misfires the first time, there's probably residual fuel left over from the leaky injector.
 

· Registered
Joined
·
60 Posts
Discussion Starter · #36 ·
No more misfires! Success! She ran good first time even. Let it idle up to warmish, and then did a idle relearn everything ran fine. I'll test drive around the block and then it gets dropped off for an alignment job. (lots of suspension work) But I'm confident it's fixed, this thing was misfiring cold every single time, badly.

It's amazing, multiple misfires, random misfire code all from 1 bad injector. Not only that, but initially I would often work up to 3 bad cylinder misfires, and it was never the bad one number 5. Eventually all of them would report as misfiring. Sweet baby jesus I stared at this engine a long time working things out, fixing and doing things I've never done before. Got some new tools along the way which is always super cool.

Thanks to all that helped along the way.
 

· Registered
Joined
·
77 Posts
Congratulations, Jerry! Yes. One cylinder that’s misfiring can affect other cylinders- especially with OBDII. if you dive in and learn how the PCM even figures this out, it’ll be easy to understand. I just did a cylinder balance lab last week on a Ford 4.6 Explorer. The attached picture shows the dip from each cylinder kill via unplugging injectors. You can see how certain cylinders will affect others in this graph. Surely this is what was happening to you.
min a week or so, and after you’ve completed your drive cycle, go back and look at your long and short fuel trims and let us know what you have.
Output device Gadget Tablet computer Communication Device Computer
 

· Registered
Joined
·
13,640 Posts
No more misfires! Success!
In which case I'm happy to award you the Piloteers Dancing Banana of Success.

Toy Electric blue Event Beard Fictional character





Of course, get a printout of your alignment. I remember one place didn't want to and tried to tell me it's useless and customers don't care.

I insisted and found out why. He wasn't too happy but he had to ask his senior guy to redo his flunky's out of spec work. That's one thing I like about specs: they're harder to argue against.
Font Art Symmetry Circle Electric blue


Yeah, you're not the only one to have an alignment tech fail. It's not rocket science, especially when that's what they do for a living, and when you point it out they could at least listen and consider your ignorant peasant suggestions instead of denying and giving you guff. Not to mention the hassle of having to go back.

Rear Camber on 2006 Pilot 4WD


That said, to err is human, to forgive is divine.
But to forgive after an apology and a full refund is even better.* View attachment 141767




*...for us lowly and imperfect humans, anyway.

 

· Registered
Joined
·
60 Posts
Discussion Starter · #39 ·
Congratulations, Jerry! Yes. One cylinder that’s misfiring can affect other cylinders- especially with OBDII. if you dive in and learn how the PCM even figures this out, it’ll be easy to understand. I just did a cylinder balance lab last week on a Ford 4.6 Explorer. The attached picture shows the dip from each cylinder kill via unplugging injectors. You can see how certain cylinders will affect others in this graph. Surely this is what was happening to you.
min a week or so, and after you’ve completed your drive cycle, go back and look at your long and short fuel trims and let us know what you have. View attachment 166804
I cleared out the trim data and noticed an immediate result. both banks now running similar on short trim, somewhere in the close to 10% positive range. Probably a slight vacuum leak, dirty sensor, or just a 17 year old vehicle, but no engine codes and no misfires and I'll just leave it alone for now haha. Before bank 1 was that value and bank 2 started at 13% negative trim and I watched it go down to almost 30% negative. I've heard of those power balance tests but don't think I can do that with my setup.
 
21 - 39 of 39 Posts
Top