Elevated fuel trims and rear oxygen sensor

Hello all. Hoping to get some help with the issue. The car is 2015 Mazda 3 with 2.0 Skyactiv gas engine (120 HP). About 45,000 miles.

I bought this car a couple of months ago and been driving and it felt like the fuel consumption was less than it should have been. So I checked with the scan tool and both long and short term fuel trims are indeed increased in the range of about 10%-15% total for both of them. They do not get back at 2,000 RPM so doesn't look like a vacuum leak. I did poke around the live data and the MAF and other PIDs seem to be okay as far as I can tell, but what caught my eye is the rear (downstream) O2 sensor. It kind of drops down periodically while driving or while at idle without any apparent reason and goes lean (0.1-0.3V) and then looks like comes back alive. The front O2 doesn't seem to change at all at the same moment while the rear O2 does its thing. I am attaching a screenshot (in fact a couple of them because I could not fit them on one picture. This is the same time moment, just split into 3 images). This is at idle, and you can see at around 03:30 the rear O2 goes down. Everything else seem to stay the same - no change to RPM, fuel trims, front O2. So my question is - does it look to you guys as the faulty rear O2? I also want to make a screenshot while cruising and will post it later.

Thanks for any input!

From your photos, your total trim is under 9% at idle. I'd be more concerned about a performance issue if the LTFT was above 10%.

Your MAF reading might be a little low, so something easy to do is clean the MAF and see if the trims improve.

The AFR sensor is for fuel control while in closed-loop engine operation B1S1 ( Bank 1 Sensor 1). Create a vacuum leak and see if the current goes up, then add propane to the intake stream and see if it goes down. This will tell you if the sensor is telling the truth and is capable of responding to Rich/Lean conditions. I wouldn't worry about B1S2 for now.

Post the same info you have in the photos, parked, with the vehicle in drive, foot on brakes, IDLE and 2500 RPM.

Another easy test to do in a SAFE place ie., empty parking lot, is Wide Open Throttle test with the vehicle in low gear. B1S1 should respond immediately Rich, if it doesn't your fuel pump could be on the weak side or some sort of fuel restriction.

This is just some easy stuff to do, chasing LTFT under 10% can turn out to be a waste of time.

Hi Weycraze - Thanks for your suggestions. Here is what I have done so far.

I am attaching two pictures with fuel trims and other data - one is at idle and another one at 2,000 rpm. The car is manual, so no load, both were done in neutral gear.

I have noticed the MAF reading before too, so I already tried cleaning it. There was no hot element visible outside on this style of sensor, so not sure how effective was the cleaning. But there was no change in the fuel trims before and after cleaning as far as I can tell. Of course it might still be not working correctly even after cleaning.

I did the WOT test too and the AFR was rich at WOT. Also when coasting or doing snap throttle, it does go lean too, so appears to be okay. But from my research it seems that the downstream is also part of the fuel trim strategy on these cars.

The EVAP purge solenoid is at about 15-20% duty cycle at idle - but from the service manual they say it should be at 0%??

Let me know what you think!

My two guesses are - a MAF that reading lower than it should or the downstream O2 that periodically drops to lean for no apparent reason...

Disconnect ( and Plug) the Purge Valve line from the purge to the intake and see if the trims improve.

MAF Still seems low, what's the highest you can get out of the MAF? A quick 'Ball Park' goes like this:

VE (g/s) = Engine Displacement (liters) X RPM Achieved X .008

So you do a bunch of snap throttles ( try to attain the highest MAF reading possible) and watch the RPM and the MAF data Pid, Record them both at the highest peak, then plug in your rpm to the formula.

EG, Recorded MAF 50g/s @ 5500 RPM.

enter 2.0(l) X 5500 X .008 = 88 g/s ( this is a ballpark VE formula set for %80 ). If your numbers are way off you either have an airflow restriction from
the intake to exhaust system or the MAF is reading too low.

I attached this photo for reference but notice the trends on the snap throttle, the CALC Load, MAF, Rpm, TPS should all rise together; and the CALC load should peak out with the RPM.

Here are the numbers from the WOT pull test. The maximum RPM were 5653, and the MAF peaked at 98.37.

So by the formula from your post - 2.0*5653*0.008 = 90.45. So looks like close to what the MAF was showing... Does this take the MAF off suspicion? Or can it be under-reporting, say, in the bottom to mid-range and be okay towards the top readings?

Haven't get a chance to check the trims with purge line closed off. Will post when I do that.

Restricted cat.?!