Please help interpret O2 sensor readings.

[Dan]

Hello there and thanks for reading. I’ve been having some major performance issues with my 2006 Mazda 3 2.3L for a while now and I’m still trying to figure out whats causing the problem. The car has a lean condition and a decent lack of power. The Short term fuel trim looks ok, bounces back and fourth without really going higher than 7+ or – but the long term is constantly at or above 10% usually maxing out around 17%.

Symptoms:
Heavy clutch and labored shifting (the clutch/shifting isn’t always heavy but the worse the symptoms get the worse the feeling gets. Sometimes its really not that bad but it still isn’t correct as to the way it drove when everything was ok.)
Power steering is stiffer. (Again not always the case)
Lack of overall power/ struggles under WOT.
Can hear the fuel pump running. (It should be noted that I installed a new fuel pump to try to fix this performance problem but it didn’t help and the new fuel pump makes noise right from the first time it turned on. I performed voltage/voltage drop/ground tests on it and everything looks fine afaik. The voltage drop was 0.37v and its my understanding that anything less tan 0.5v is ok…..unless I’m mistaken. I also found that the hose that leads from the fuel pump to the fuel rail was a little kinked in one area so i replaced it with a good one but this didn’t help anything.)

I’ve been all over the car with water/propane/brake cleaner trying to find vacuum leaks and I’m pretty confident there are none. There was a little soot as well near the exhaust manifold too so I changed the manifold gasket but I’m unsure if there are any other exhaust leaks, it doesn’t sound like it though when I’m driving but I’m going to try to do a smoke test at some point to be sure.

So onto the O2 sensors. I just got a decent scan tool recently so I can finally look at some numbers although I don’t quite know how the downstream sensor should look. The upstream sensor is an A/F sensor and its readings look good. It osculates its current usually between -.08 and +.4 at idle and it doesn’t really get to crazy but the downstream O2 sensor just seems funny to me. It will read very rich almost all the time in around the 0.845v range give or take a little but then if you just let it sit it will drop down to 0.1v or below and sit there. Am I correct to assume that it should not go below 0.1v? It doesn’t hit zero but it will fluctuate between 0.07v and 0.1v but its usually below 0.1v. On the flip side if the numbers are sitting up in the rich range and I snap the throttle a bit, it will drop right back down to around or below 0.1v then it will sit there for about 10-30 seconds then climb all the way back up to the 0.8 or 0.9v range.

I find these readings strange and I’d like to say the sensor is pooched but I would like to hear your guys input on this. Do you think its the sensor or does this actually sound like unmetered air/exhaust leak/catalytic converter territory?

Thanks again

Topic

[Wyr Twister]

If you can unplug the rear O2 sensor , try that & see if you are getting power to the heater ? Also see if you are getting voltage from the computer to the sensor wires ?

Use an ohm meter and test the resistance on the heater in the O2 sensor .

Post the results , for some one here that knows more about this than I .

Just for grins , See how the engine runs with the O2 sensor unplugged . This will probably set another code .

God bless
Wyr

[Travis]

Try cleaning MAF. What u indicate is a rich condition. Could be vacuum leak but u said u got that covered. Quick easy.

[Mike]

Since you have a scan tool, you can do a quick test.
With the car warmed up,record the at idle STFT and LTFT readings.
Next, hold the rpms at 2500 for 30 seconds while watching the fuel trims.
Record the readings at 2500rpm.
Does the LTFT drop or does it stay the same or increase?
Answer these questions for the STFT.
Post all the answers and actual readings when you’re done.
This should help others determine where the trouble lies.

[Rich]

The Short term fuel trim looks ok, bounces back and fourth without really going higher than 7+ or – but the long term is constantly at or above 10% usually maxing out around 17%.

Bonnieman makes a good suggestion, and I’d like to see the results as well. It’s important to record fuel trim data against some parameter. In the early stages of diagnosis the most useful secondary parameter is engine rpm.

A lean condition that subsides at high rpm is consistent with a vacuum leak. Given that we have no fuel trim data at high rpm, we can’t rule out a vacuum leak.

Other factors can cause lean conditions – for example a fuel delivery issue. The lean condition should get worse under high rpm in this case.

Beware: your long term fuel trim may be pegged lean. If this happens you can’t make the engine go any more lean and may be misled.

Whatever you do, resist the urge to start cleaning or replacing parts until you can confirm one of them is bad.

[Andrew Phillips]

The purpose of the LTFT is to keep the STFT near 0 to allow for the computer to oscillate lean/rich to feed the catalytic converter. The catalytic has two clearing beds in it. One bed needs CO for its chemical reaction to work, the other needs O2 for its chemical reaction to work. The computer oscillates the mixture between slightly lean and slightly rich so that the cat has the gasses it needs to do its job. A narrow band (usually Zirconium) O2 sensor is used for this job. These sensors should oscillate between 100 and 900 millivolts ideally, and between 250 and 750 millivolts is commonly observed. The point is they should constantly be moving back and forth in closed loop. On older vehicles, a second narrow band O2 sensor was used to measure the output gasses to monitor the efficiency of the cat and let the computer know that the engine management strategy was working. Newer vehicles employ a wide band sensor (also called an air-fuel ratio sensor) for this purpose. The purpose of the STFT is to add or subtract fuel to the engine to attempt to maintain its stoichiometric ratio of 14.7 parts air to 1 part fuel. When the STFT goes positive, this indicates a lean signal from the O2 sensor, which causes the computer to add fuel to the fuel mixture (make it richer). When the STFT goes negative, this indicates a rich signal from the O2 sensor, which causes the computer to reduce fuel (make the mixture leaner). A LTFT of +/- 10% is generally not a worry, and is perfectly normal for cars more than a couple of years old due to wear and dirt. When the LTFT remains higher than 10% or lower than -10%, this can indicate a malfunction in progress, even if a code hasn’t been thrown. As long as a narrow band sensor is oscillating and a wide band sensor is maintaining a steady reading, then the sensors can generally said to be good and the problem is elsewhere, such as a vacuum leak, pirate air entering the intake, a clog in the fuel filter, a clogged air filter, faulty fuel pressure regulator, or a weakening fuel pump. Those aren’t all the reasons, just some of the most common ones. My recommendation is to check out the air flow sensor(s) and make sure there aren’t any leaks or splits in the fresh air intake tubing, and that all the joints are sealed and tight, between the filter box and the throttle body.

This is what a graph of properly operating O2 sensors should look like:

Attachments:

• o2_sensors.png

[Author]

Replies