Wideband A/F sensor issue

[ramsayh]

I would like some advice on how to tackle this issue.
After installing aftermarket exhaust manifolds on an 06 Nissan Altima 3.5L Automatic, A check engine light came on.

All stock 2005+ model altimas have 2 wideband (5 wire bosch) primary AF sensors attached to the exhaust manifolds, which are then connected to a y-pipe with 2 precats. There are 2 heated oxygen sensors attached to the precats as well.

When performance headers are installed, the new y-pipe elimantes the precats since there is no space for those anymore. The secondary sensors are then moved down farther (these are not the problem).

The issue lies with the 2 primary A/F sensors. These are the readings that I got after header install:

bank 1 reads 1.2v
bank 2 reads 1.6v

The problem is that one bank is reading too lean while the other is reading too rich. The A/F after the Y-pipe and before the main catalytic converter is ~14:1. According to the A/F sensors, Bank 2 is at 16:1 and bank 1 is at 12:1.

Would it be a mistake to attempt to reprogram the cars ECM to read those values as normal?

Maybe tap a device into the pin outs that serve the ECM info from sensors and some how monitor and adjust the voltages?

All help is greatly appreciated. Thanks in advance.

Here is an illustration of the stock setup:

 

[mattsooty]

ramsayh

As you may already know the purpose of the wideband sensors is for closed loop fuelling control and the purpose of the two step sensors in the precat is for O2 storage control of the catalyst and also for catalyst monitoring.

If you remove the cat but keep the sensors then the ECU is going to think that the catalyst has zero O2 storage (since the cat is no longer there) and this is most likely the reason why you have the MIL activated.

I am not sure why the AFR is different from bank to bank but maybe this is a limp home condition when the catalysts are found to be deficient. The fact that you have, on average, stoich would tend to agree with this.

My advice would be to get somebody with a scantool to take a look at the vehicle and retrive the DTCs, if post them on here and I will take a look for you as soon as I can.

MS

 

[ramsayh]

I will get those DTC values and post when I get a chance.

The thing is that a pre-catless Y-pipe with stock exhaust manifold does not throw MIL. The downstream o2 sensors are in place with o2 simulators so it fixes that issue.

But the same setup with aftermarket performance headers will throw MIL.

Anyway, I will get some more numbers and let you know. I have a diagram of the exhaust system for my specific car. I'll put that on here as well.

Thanks for your help!

 

[ramsayh]

Okay, here are the 4 DTC's right after the header install:

P1148 = closed loop bank 1
P1148 = closed loop bank 1
P1283 = A/F ratio lean
P1031 = A/F Ratio Sensor Meter

A lot of modifications have been done in attempt to fix this issue. Different codes kept coming up so no luck there. I posted original codes since I feel that we need to start from scratch.

 

[mattsooty]

ramsayh,

Just to double check, is the second line a misprint and is infact supposed to be P1168??

It sounds obvious but have you checked for air leaks in the new headers?

I think that your ECU is failing to start closed loop AFR control and you are getting some big discrepancies with the open loop fuelling (or measurement thereof at the sensor).

Posting that diagram might be helpful in understanding the problem further.

MS

 

[ramsayh]

I linked the diagram in the first post at the bottom.
The second line is not a misprint by the way and no leaks in the headers.

There is a big discussion regarding this on a nissan forum. Here is the link:

http://www.nissanclub.com/forums/en.../212103-hs-headers-ses-cel-help-please-4.html

 

[Thecardoc]

quote [The downstream o2 sensors are in place with o2 simulators so it fixes that issue.]quote

Actually that's not correct. OBDII is an emissions control system where the primary function is to make the catalytic convertors work correctly as much of the time as possible. To do this virtually every manufacturer employ's a rear O2 sensor fuel trim strategy. Simulators might trick the computer into thinking everything is OK with the circuit, but ultimately will cause a fuel trim shift that would be very difficult to diagnose without the ability to read not only short term and long term fuel trims, but rear trim in scan data as well.

 

[brettsb16a2]

What type of downstream O2 sensors does it use, are they narrow band???

I am wondering if you have some type of exhaust leak causing the lean signal.

I didn't think you can use simulators on CAN. I used a simulator on obd2 and it worked great on my turboed civic.

ITS CAN RIGHT???