Poor MPG, pinging, exhaust smells rich on startup, no codes

[shinckley08]

2013 LR4 5.0 169k miles
Pinging on light acceleration especially under a load like going up a hill. I did a 5 hr highway drive and got about 14mpg, I’m used to 17+.

Recently replaced the evap valve because I was getting a fuel smell on hot days when at idle also, that seems to be gone but no change otherwise

Replaced plugs

About a month ago the cats were smelling for a day or two, but that went away. I assumed it was from idling too long and not driving enough to get them hot. Didn’t think much of it but now am wondering if running rich has ruined the cats.

No codes, no perceivable loss of performance.

I don’t have a gap tool or know what I’m doing but it seems to me like it’s running rich, and I’m wondering if with my cheap odb2 dongle if I can do my own diagnostics. Forums and and searching points to bad cats, bad 02 sensors, issue with cam phasers, cracked manifold, or a leaking injector, or a combination of these things.

Long short if anybody has any advice on diagnosis I’d happily take it. It’s my daily driver and I’m getting around 9mpg in town $$$

The only thing I looked at was #2 O2 sensor voltage.

 

[itsaguything]

In my mind, this is not a plugs issue.
CATS do clog, though the cheap fix is to confirm the sensors are working properly, first.
And if there is an air flow issue, I would expect codes. We really need the thrown codes.

 

[jlglr4]

See if you can find fuel trims on your cheap scanner. Look at long term and short term for each bank at idle and at one or two higher rpm levels (in park, just hold the throttle as steady as you can). Won’t give you a firm diagnosis but will tell you if rich or lean and might help point you in the right direction.

 

[shinckley08]

In my mind, this is not a plugs issue.
CATS do clog, though the cheap fix is to confirm the sensors are working properly, first.
And if there is an air flow issue, I would expect codes.

See if you can find fuel trims on your cheap scanner. Look at long term and short term for each bank at idle and at one or two higher rpm levels (in park, just hold the throttle as steady as you can). Won’t give you a firm diagnosis but will tell you if rich or lean and might help point you in the right direction.

Thanks for the reply’s. Still no codes.

I captured fuel trims at idle (first image). And at 2k rpms (second image)

I’m not sure what to look for but I expected bank 1 and 2 to more closely mapped on the chart and think this means one bank is running more rich than the other

 

[jlglr4]

One thing I forgot to mention - engine needs to be warmed up so the car goes into closed-loop feedback fuel control to evaluate fuel trims. If it wasn’t warmed up, you’ll need to do this again with the engine warm. Same for o2 sensors. Also, when looking at o2 sensors in the context of fuel control issues, you generally want to focus on the lambda sensor, which is usually designated as sensor 1 on our LR4s, although whatever sensor you tested seems to be giving a nice wave form.

Turning to your fuel trims graphs, add long term to short term for a given bank to get the overall trim value. Positive is lean (positive means that the engine is adding fuel to compensate for the o2 sensors not seeing enough in the exhaust). Negative is rich (engine subtracting fuel). It looks like Bank 1 starts out around +11 to+13 at idle then drops to about +3.5 to +8 or so at higher rpm. Bank 2 looks like it starts out at +5 to +8 at idle, then drops into a range of somewhere from +1 to -5 at higher rpm. So both trending from lean to less lean (or even a little rich in Bank 2). Although “normal” range is said to be +/- 10, so none of these values are really too far out of wack.

(One caveat: With the lines split and tracking each other on a graph like this, it’s tough to tell if you might actually have two y-axis scales and one might be hidden. Some scanners will default to overlapping scales so that it shifts the lines of the graphs making it easier to see the lines. Look for some button or option to show all the y values, or if you can pull up a screen showing live values (just numbers, not the graphs), it might tell you whether the scale is the same for all the graph lines.)

Assuming the number are correct, trending from more lean to less lean as RPMs increase most often points to a vacuum leak. At idle, the unmetered air being sucked in accounts for more of the total volume of air being consumed by the engine, so the lean condition is more significant. When the RPMs are raised, more total air is being consumed and the amount that is unmetered (vacuum leak) is a smaller portion of the total, so the lean condition is not as significant. However, the leak would be a fairly small one with these numbers, but might be enough for a little pining or loss of fuel efficiency. If the PCV valve hasn’t been changed in a while, you could check it - but that usually gives a much bigger lean condition when it goes. Otherwise, maybe buy a cheap smoke tester off amazon and smoke test the intake plenum.

You also could check or clean the MAFs - see if that makes a difference. Use a dedicated cleaner (spray on - leaves no residue and doesn’t damage the elements or the plastic housing.

Double check the o2 sensors - make sure they are giving a nice sine wave when the engine is warm.

You mentioned the evap purge valve. Do you ever have problems filling up the tank (clicks off prematurely)? If you have excessive fuel vapors (or even liquid fuel) in the charcoal canister, you can get temporary periods of rich conditions when that purge valve opens (supposed to open when the car is cruising).

 

[shinckley08]

This detail is so helpful.

The engine was warm for my readings. I cleaned the MAFs on my long trip when I noticed mileage was bad. Never had an early fuel stop issue while filling. The evap system definitely overflowed when the purge valve was bad, I’ve put about 1500 miles on the new valve, no remaining fuel smell issues so far. But I’m certain that system was overburdened at that time. I believe all the known conditions existed prior to the new purge valve, I replaced it due to the combination of the cabin fuel smell and pinging. The only thing that is new is the unspent fuel smell from the exhaust on startup. It’s distinct because it reminds me old cars as apposed to the fresh fuel vapor smell i was smelling before purge valve.

PCV valves are new as well, within the last few weeks.

The vacuum leak logic makes sense, not sure where to start, but it does make sense.

Thanks for all the detail, I’ll see if I can play with the y axis on the apps I have, one may offer more flexibility.

The attached is the O2 voltage at idle and warm. Originally I thought this might mean the bank 2 sensor is lazy as it doesn’t peak and drop, it plateaus.

 

[jlglr4]

The only thing that is new is the unspent fuel smell from the exhaust on startup.

This suggests injector leaking after shutoff.

The attached is the O2 voltage at idle and warm. Originally I thought this might mean the bank 2 sensor is lazy as it doesn’t peak and drop, it plateaus.

That’s a good catch. could be a lazy sensor, but also could be a symptom of something in the fuel control system. bank 2 is also where you are seeing the short term fuel trim fluctuate into rich territory (even though overall fuel trim isn’t maybe that rich). i suppose a leaking injector might cause this too. hmmm, easier to swap out o2 sensors than injectors if you wanted to try it.

 

[shinckley08]

That was my thought too, switching the sensors and see if the problem moves.

I noticed a light tapping sound today that doesn’t sound like the lifters ticking, its loudest from the bottom passenger side, which might be a high pressure fuel pump. I’ve replaced one about 100k miles ago, but the other is original. Not sure if some noise is ok, or if that means one is failing. It does not sound like a manifold tick, but that’s another possibility too I suppose.

And a leaking injector is one possibility I was hoping rule out, but it makes sense. Unplanned fuel in the cylinder would come out the exhaust on startup. I did the drivers valve cover gasket and I believe I replaced all the bank 2 injectors, they were a PITA.

I’m assuming there is no way to tell which cylinder has the leaker, that if it comes down to it they all have to come out for testing or replacement.