My G started running rough, throwing check engine light due to running too rich? lean? on Bank 2 (passenger side). Power down 10%-15%, exhaust droning more than normal (I'm almost positive that this is not caused by any exhaust system issue.) I'm trying to sequence the trouble shooting from easiest to hardest, while also considering probability.

The exhaust smells rich, but data logging indicates otherwise.

I'm thinking possibly:
1. Bad Coil -
-I plan to swap all from Bank 1 to Bank 2 to see if problem moves. This would rule in/out coils.

2. Faulty Bank 2 #1 02 sensor (narrowband on my early '04)
-I plan to swap from bank 1 to bank 2 to rule in/out o2 sensor.

3. Compression problem
-I plan to run compression test

4. Faulty injector - Sticking open/rich? clogged/lean?
- Any troubleshooting on this requires puling intake manifold for access so I'm reluctant to go here.

I have Uprev Cipher and can read ECU parameters and log engine performance. I've attached a short log.

This problem occurred about 6 months ago. I thought it was a problem with my exhaust. I replaced my rear muffler thinking there may have been a problem with it. It did nothing. I used Cipher to selectively turn on/off each cylinder (shuts off injector and spark) to try to identify which one was not performing correctly. (The problem was on Bank 2, I don't remember which cylinder. During the cylinder shutdown testing, the problem suddenly cleared and the engine has run perfectly since then until now.

Using the cylinder shut down test I am currently unable to discern which cylinder is causing the problem. The problem IS clearly Bank 2 specific making me think that I should perhaps look at the o2 sensor first. (I think a coil, injector or compression problem would be cylinder specific) If the 02 sensor is faulty, perhaps reading leaner than actual, then it may be causing the ECU to flood that bank. That could explain logs that say lean while the exhaust actually smells rich. Having said that, if that were the case, why would the injector bandwidth be lower on average for Bank 2???

I have attached a pdf of an operating log from cipher. I observe in it that:

The A/F correction generally reads lower on Bank 2
The difference between B1 and B2 AFR readings oscillate but average slightly higher on bank 2 (leaner)
The injector pulse width is generally reading a little higher on Bank 1

Bank 2 o2 Sensor 2 readings are wild because my high flow cat is below efficiency and I have an anti-fouler installed on B2 (for off road testing purposes, yah...)

Please respond: Anyone who has experienced anything like this or who can share useful advice on my troubleshooting process.

My brain hurts trying to figure this out and I have little experience with tuning or this sort of diagnostic work.

 

See the AFR differential charted in the attached Word doc

 

I'm having similar problems: rough idling, black smoke from the exhaust when accelerating (running rich), CEL is on, possible drone. Had it checked and there were 3 error codes, apparently all linked back to faulty o2 sensor.

It took me two weeks to get the o2 sensors from Japan (I am in Australia), but they arrived yesterday and will be fitted on the weekend. I'll let you know how I go.

 

How many miles do you have on your car? Change anything recently? My first guess would be that you have a bad or multiple bad O2 sensors.

 

70,000 miles on car. Nothing changed lately. I guess I'll make my next troubleshooting step switching the #1 o2 sensor from bank to bank and see what happens. At least it's an easy (I think) job. Easier than switching over all coils or messing with injectors and a reasonably likely source of the problem.

 

If you have a Cipher, go read the error code number and see what sensor is bad (or is reading bad). An exhaust leak upstream of the main 02s might cause a lean read on the o2 and tell the ecu to throw more fuel into that bank to compensate.

 

The only code I've been getting is L0420 (if I remember correctly) which is a cat efficiency below threshold (Bank 2), which is a o2 sensor 2 reading of too rich or too volatile.

There are no codes being thrown indicating a faulty o2 sensor. There's no exhaust leak ahead of the sensors. (The first ones are in the manifold!)

The injector bandwidth reading is telling us that the ecu IS trying to lean it out, ie there's a shorter injector pulse being delivered on the problem bank 2, but the AFR is reading a little leaner on bank 2 as well. I wonder, could it have to do with the bad cat and be leaning due to sensor 2 input in an effort to reduce emissions?

At this point the more I think about it the less sure I am about any possible cause. I've just got to get to switching parts from side to side to rule things out. Watch this space!

 

the 0420 (cat efficiency) can be caused by a bad o2 sensor.

 

Yeah, that's true, but I already swapped the #2 sensor from left to right and the error code still comes up on the same side so that's not the problem. Also, measuring the temperature front to back the converter gets cooler rather than hotter towards the rear. A properly functioning converter is hotter at the rear as a result of the burning of excess fuel inside, that's what they do.

In any event, I'm a lot less worried about that code or the fact the converters aren't up to par than I am about running like crap.

 

Aren't there two 02s for each cat? In front of and in the rear of the cats? Could it be a rear 02 sensor reading bad?

 

Cipher will tell you if you have bad O2 sensor... also will allow you to run compression test...

Have you tried to diagnose with Cipher yet?

 

cipher lets you do a compression test?

 

Trying but lacking extensive experience. How do you use cipher to test compression? What function? I've been going through the manual and various functions but still not finding it. I've done Cylinder Power Tests. I've heard compression tests can be done electronically but I can't see that functionality in Cipher.

I'm getting no message on a bad o2 sensor. Are you talking about reading a code or is there some test function in cipher you are referring to?

 

my bad on terminology... meant "cylinder power test"... as for O2 sensor, yes I meant reading the code with Cipher - if you have no SES light and no codes, the sensor shouldn’t be bad... apparently you have done these things - so nvrmnd on my suggestion

 

Have you checked the following:

PCV valve - gummed up or sticking

Engine vacuum - it should be a rock steady 19 to 21 hg at a warm idle

O2 sensors - have they ever been dropped, handled roughly

MAF - cleaned it recently?

Intake and tract - inspect and verify all connections are secure and tight.