[FIXED]2008 Honda Odyssey 3.5L Lean Condition

I've got a 2008 Honda Odyssey Touring with the 3.5L V6 with 168K trouble-free miles that I have owned since new. I don't have any codes, nor any driveability problems yet, but recently I have noticed higher than normal fuel trims indicating the engine is running lean. I am seeing +10 on the short terms and +10 on the long terms for a total of +20. It shows the same fuel trim numbers both at idle and at 3000 RPM. Due to the number of miles I replaced both upstream A/F ratio sensors and I cleaned the MAF. At first it seemed like the new A/F sensors fixed the problem as shortly after replacement the fuel trims were down to +3 on long term and around +1 on short term on both banks. However, after a few hundred miles the fuel trims are back up high again.

I have checked the MAF sensor on my scan tool (Autel DS708) and it shows around 10 g/s at 2500RPM and the service manual calls for 9-12 g/s so the MAF appears to be good. I suspected a vacuum leak, however, since the fuel trims don't improve at high RPM I'm less suspicious of that but still tried pinching off the PCV, brake booster, and purge valve lines but no improvement. When pinching the purge valve there is slight improvement but just momentarily. It appears that the purge valve is duty cycle controlled and I noticed it commanded open to about 22% or so and I can feel it clicking while the engine is running.

I have checked the EGR commanded position and the actual position and both read 0. I did the EGR test in the Autel and the test says "Test Complete" but didn't give me any info as to whether it passed or failed so I assume it passed. The EGR was replaced at around 17K under warranty but hasn't been touched since.

What should be my next steps? Valves were adjusted at around 140K. I'm wondering if the rear O2's could be getting a bit lazy and causing this. They read around 640mV and are fairly steady at idle. I have read that Hondas with A/F ratios on the upstream do actually use the downstreams for fuel trims but don't know to what extent. With the number of miles on the vehicle I'm not opposed to replacing them if there is a reasonable possibility they might be causing this.

As a side note, I have a friend with a 2009 Pilot who I did a timing belt and valve adjustment for about 20k miles ago (currently at 167K) who brought the vehicle back to me yesterday with lean codes (P0171, P0174) and showing over 20% on the long terms and around 7-8% on the short terms. Symptoms are the same as on my Odyssey above. Is there anyone out there with some Honda experience that might have some tips for an elusive lean condition on these?

Thanks,
John

You did not mention fuel system, maybe cleaning injectors, checking fuel pump pressure, changing fuel filter, if not in tank. A smoke test on intake would be first thing I would do. hth

A smoke machine is a great idea and I'd do that...if I had one. Maybe it's time to get one. The fuel filter is in the tank and I believe it was changed during a recall just a year or two ago. I haven't actually tested the fuel pressure (Honda doesn't make that easy) but usually low pressure is accompanied by reduced power, or at least higher fuel trims at higher loads, is it not?

Fuel injectors are a possibility, of course, but I've got no data to point me to those yet. There are no misfires.

MAF g/s seems to be a bit on the low side. What are the g/s at idle and at 90% load?

cj1 wrote: MAF g/s seems to be a bit on the low side. What are the g/s at idle and at 90% load?

Thanks for that reply. I'm out of town for a bit so I don't have the vehicle in front of me for a few days. I want to say it was around 2-3 g/s at idle but I don't recall for sure. The factory service manual says the following:

While holding the engine speed at 2500rpm, check the MAF Sensor in the data list with the HDS. Is there about 9.1 - 12.9 gm/s?

Yes--Check the engine valve clearances and adjust them if needed. If the valve clearances are OK, replace the injectors.
No--Replace the MAF sensor.

The only other thing the service manual says to do for lean codes is check the fuel pressure. Unless there are other codes associated with the lean codes then Honda doesn't think there is anything else that will cause a lean code other than low fuel pressure, MAF sensor, or out of adjustment valves. That seems a little vague, especially considering these instructions are for the lean codes AND rich codes.

For my MAF reading the Autel showed 10.2 g/s at a steady 2500RPM. The middle of the range Honda gives above is 11 g/s so I guess I'm nearly 10% below the median range Honda gives. I've ordered the new rear O2's, to go along with the new front A/F's I put on a while back just because they've got a lot of miles on them and they've been on my list of regular maintenance items to change out anyway. I can also throw in a new MAF (I've already cleaned it trying to fix this and they're not that expensive) but I don't want to just throw parts at this.

Thanks for the help,
John

MAF at idle normally equals liter numbers. So should be at least 3.5 g/s.
Noticed that after you had cleaned MAF and same time changed A/F sensors fuel trims normalized for a few hundred miles.
Could be the MAF cleaning helped since new A/F sensors shouldn't have degraded.

That's a good rule of thumb to know. Thanks! Actually, I cleaned the MAF before replacing the A/F sensors and it didn't appear to help. That's when I decided to go ahead and replace the A/F sensors just due to so many miles and to protect my cats.

I'll focus in a little more on the MAF and see what other data I can gather (maybe like a good working J35 MAF reading from someone.) The J35 engine isn't really known for vacuum leaks, and I couldn't find any, but anything is possible. The EGR port goes through the side of the front cylinder head, up through the lower intake runner, and then into the upper intake so it's not really prone to leaking either. This shouldn't be that difficult to track down but it has eluded me so far. And then to have two with similar symptoms, one worse than the other, I thought was a bit strange. I'm tempted to just put a MAF in it but if it doesn't fix it I know Danner would slap me for firing the parts cannon at it and all the hours watching his videos would be for not! I wanted new O2's all the way around regardless so I don't count that as firing the parts cannon.

I believe I have fixed this. I replaced the two downstream O2 sensors. Since the downstream sensors contribute to fuel trims, and the originals have nearly 170K on them, I went ahead and replaced them. After replacement, the fuel trims went down to around +8-10% total. I drove it for a bit and they seemed to stay around 5-6% on ST and LT. That was three days ago.

Just due to miles I ordered a new Hitachi (OE) MAF from RockAuto that arrived today. However, before installing it I decided to take a look at the fuel trims again first, and they are now running about -3-5% on ST and -0 - -1% on LT. So, they are fixed and have even moved slightly to the rich side now. I don't know if Honda takes a bit longer to "learn" on fuel trims after replacing the rear O2's or if the most recent tank of gas did it but things are now running normally. I went ahead and put in the new MAF just because the old one has so many miles on it, and as I suspected, it didn't change anything. However, with all the miles it seemed like the prudent thing to do.

While I'm happy to have fixed it, the worst part of all this is I have no data to condemn anything I have changed. The original upstream A/F's seemed to read normally. The downstream O2's seemed to work normally, as well. I could drive the downstreams rich and lean with no problem. However, after replacement my 10-20% lean condition (no codes) is gone and now is maybe a couple percent on the rich side and the PCM is subtracting a little fuel. I'm guessing the old downstream sensors were just old & dirty and not reading correctly even though they were still working and not setting any codes.

Are these things really just maintenance items? I have friend with an 09 Pilot with about as many miles who has lean codes. Should I really suggest replacing all four O2 sensors for $250 plus the labor to put them in for him?

" I'm wondering if the rear O2's could be getting a bit lazy and causing this. They read around 640mV and are fairly steady at idle."
Since you changed the rear sensors have the sensor readings changed?

Regarding the 09 Pilot with lean codes, would take similar steps that you took with the Odyssey.

cj1 wrote: " I'm wondering if the rear O2's could be getting a bit lazy and causing this. They read around 640mV and are fairly steady at idle."
Since you changed the rear sensors have the sensor readings changed?

I didn't look at the new downstream sensor readings but will do that. What should they read? 640mV is reading rich which I would think would make the PCM reduce fuel. It would sure be nice to know the strategy Honda uses for downstream O2 fuel trim adjustments.

Regarding the 09 Pilot with lean codes, would take similar steps that you took with the Odyssey.

Yeah, last night I had the vehicle again for a bit more time. His lean codes came back so I did some more looking. Seems the last place that changed his oil and filters left the air intake tube clamps loose. As Eric O would say, "Well, there's your problem, Lady!" I tightened them up and the fuel trims went to under 10% on both ST and LT so I think that's fixed but told him to drive it for a while and stop back by to check them. He ordered all new O2 sensors even though I told him I needed more time with it first to diagnose it further before I felt comfortable replacing them. With his 170K on the clock, based on my own findings, it may not be a bad idea anyway. I'll post back here just to follow up in case someone else is searching later on.

cj1 wrote: " I'm wondering if the rear O2's could be getting a bit lazy and causing this. They read around 640mV and are fairly steady at idle."
Since you changed the rear sensors have the sensor readings changed?

I just looked at the downstream sensor readings and don't really see anything different than with the olds. They are reading around 600mV at idle and will go rich then lean when throttle is snapped, etc. My fuel trims are pretty much at zero so I'm calling it a fix. These things apparently do wear and go lazy after many miles. You can wait to get a code at 25% or more but I think it would run for many months in a condition just under the threshold. I see lots of no-prior-warning P0420/P0430 codes on Hondas on the Pilot/Odyssey forums and I just wonder if that's to blame. Almost nobody looks at scan data when there's no code set but if you're running at, say, 22% more fuel than normal, with no MIL set, how long until that cooks a converter?