[Sankar]

I disconnected the down stream O2 sensor coupler this evening and went for a short 2km drive, the CEL didn't come on. Why?

[vikram_d]

Use it longer with the sensor disconnected. The CEL does not always come on immediately. At first the ecu tries to see if the problem gets rectified. If does not after repeated attempts, the CEL will come on and a DTC will be recorded.

[Sankar]

Went for a longish 50km test drive and no CEL popped up yippeee

But when i plugged in my BT OBD scanner the CEL popped up in the meter console and i searched for error log with the Android app. The error shown was sensor 2, bank 1 heating circuit incomplete... it didn't say the sensor was faulty or the cat was faulty, it just said heating circuit incomplete. I cleared the error and the light disappeared. There wasn't any power loss with the light on. The second lambda couple is still disconnected.

So there goes the reason why lots of people on various forums reported no CEL with 2nd probe decoupled. The light comes on the moment the scanner is plugged in. Makes sense since the second sensor isn't critical so the CEL isn't lit during regular usage, but when a scanner is connected to the port the ecu lights up the CEL so that the technician can look into it.

[Sankar]

Scorpian, i'm now almost certain that the CEL your car (its the K12 isn't it?) had and the power loss wasn't due to the second sensor. Where was the first O2 sensor placed in the header?

Disconnecting 2nd probe made no difference to the performance of my car or its FE all remained same.

[scorpian]

Sankar.
In my headers the o2 sensor was located right on top on the second cylinder, i guess thats what has caused the cel.

[vikram_d]

Originally Posted by scorpian

Sankar.
In my headers the o2 sensor was located right on top on the second cylinder, i guess thats what has caused the cel.

You mean it was getting input from only one cylinder?

[Sankar]

Originally Posted by scorpian

Sankar.
In my headers the o2 sensor was located right on top on the second cylinder, i guess thats what has caused the cel.

Hmm.... the K series seems to be sensitive then. A friend recently installed an Automech header in his Esteem and like most Indian headers the lambda reads only one cylinder and there are no issues. Its a 4 wire lambda in the Esteem like in the K.

[Sankar]

Installed a header in the Swift today, there is no issues with the 2nd O2 sensor and wasn't expecting any either after my experiment of running with 2nd sensor removed.

http://www.team-bhp.com/forum/modifi...ml#post3019522

[Sankar]

Connected the OBD scan tool and there it was, the error code P0420 Catalyst System Efficiency Below Threshold (Bank 1). Will disconnect the sensor and run it like i did before, there wasn't any loss n performance when i tried it before.

And also going to do an experiment with the sensors, the 1st sensor reads from only 1 cylinder, the 2nd sensor reads from 2 cylinders. So i'm gonna connect the 2nd sensor to the first and leave the 1st disconnected and see how its going to run. Lemme see if the sesnors are the same and if the connectors are alike.

And here's some data from the O2 sensors, the first set of data is of the system with CatCon and the 2nd data set is with the headers and reading from only 1 cylinder.

CatCon


Headers


As long as the 2nd sensor doesn't contribute towards the AFR its fine, and from the error codes that came i guess it only does a post catcon checkup. Hoping i don't get a P2096 Post Catalytic Fuel Trim Bank 1 System Too Lean.... fingers crossed

[Sankar]

Resolved the 2nd O2 sensor issue with a simple fix which has been floating on the internet since forever. I used an O2 sensor extender (you can also use two spark plug anti-foulers) to keep the sensor away from the exhaust gases so that it reads lower amount of oxygen in the the exhaust stream.

I removed the sensor fixed this spacer and then installed the sensor. Its only possible to do this on a cold car so parked the car for the day and did it the next morning. Two days later I had to put a washer between the spacer and the bung and the sensor and the spacer and the best washer to put there is the cast iron Bullet's copper decompressor washer (use the bigger one).

There has been no CEL since the last 136 kms. If a 2nd O2 related CEL does return then i might have to look into a resistor and capacitor hack to solve it. I've bought the components but hoping that i won't have to use it and it stays fixed.

The pic is of the first install, the one without the washer.

[Sankar]

Back to square one, the CEL popped up after 269Kms since the mod with the same error P0420. So this isn't fixed. Now i need to see if the cap and resistor mod works.

I guess i know why the extender didn't work, its due to the idiotic placement of the sensor bungs in the header. The second O2 should read lower amounts of oxygen in the stream compared to the first sensor, and with an extender with a narrow opening it should. But because in this header the first O2 sensor is mounted on a single runner and because the 2nd post cat O2 sensor is mounted after the merging of two runners it is still reading more than the first O2 because the 2nd O2 gets more gases (and heat) where its mounted.

I've emailed N1 many weeks ago and asked them if they have any remedy for this since they've claimed that several K series 1.2 engines with twin O2 sensors cars are running without any error. I'm yet to get a reply.

[vikram_d]

Sankar, ideally both the sensors should be reading output from all the cylinders. I just cannot understand how reputed tuning houses can make such elementary mistakes by taking feedback from only one or two cylinders.

Also in your case, I think even if the sensors are getting the correct feedback the CEL may happen again as you have removed the catcon that was built into the OEM headers. So I guess this is one error that you may have to live with. Did your car have 2 catcons originally?

[Sankar]

Originally Posted by vikram_d

Sankar, ideally both the sensors should be reading output from all the cylinders. I just cannot understand how reputed tuning houses can make such elementary mistakes by taking feedback from only one or two cylinders.

Exactly! The pics of headers from abroad i've seen on the net have the O2 bung on the collector and i've never seen one with the bung on a single runner.

Here the scene is like this, for mail ordered headers i didnt have many choices, the builder builds them as a plug n play part with no lengthening of wiring reqd and thus mounting the sensor on the single runner.

My colleagues Esteem is running a header (bought from N1) and after some miles it started pinging. We opened the plug and found that the cylinder which was being read by the O2 was running alright and 2 other cylinders were running kind of hot. Since the Esteem doesn't have post cat sensors and lesser complications i've asked him to get an O2 mounting done at the first collector at least if not the second one.

Also in your case, I think even if the sensors are getting the correct feedback the CEL may happen again as you have removed the catcon that was built into the OEM headers. So I guess this is one error that you may have to live with. Did your car have 2 catcons originally?

There was a cat con on the manifold which is now removed when the header was installed. There is a muffler like thingy on the underside which i'm not sure whether if its just a central muffler or a secondary cat.

If i can reliably delete the 2nd O2 from the system with some electronic hack i would seal the first bung and install the first O2 where the 2nd is being mounted now. That should be a little more reliable for metering than the current setup.

[Raag Sandhu]

Two reasons why I would want to step in here:-

I own a Ritz ABS in India with no CEL issues
I own a 2005 Cadillac SRX AWD V8 notorious for CEL and other small niggling issues

Common point here being me the owner, I decided to give my two cents:-

CEL on with a P0420 code? No problem - The Caddy has had it since 2 years now and has passed emission tests in a jiffy. CEL light with a P0300 is a BIG problem since it indicates random midfire and is a pain the neck to solve especially with no particular cylinder misfiring at no particular time interval. I was asked to change all 4 O2 sensors for a couple of thousands which I opted out since I upgraded the brakes on the car instead.

Only issue I see is a slight increase in fuel consumption. So from 8.0kmpl to 6.7kmpl it doesnt really matter for the Caddy, considering fuel is cheap here. I drive 280 kms a day with a bit of a heavy foot so probably I am the culprit!!

So if your CEL is on with a p0420 code, just forget about it and drive!

[vikram_d]

Originally Posted by Sankar

There was a cat con on the manifold which is now removed when the header was installed. There is a muffler like thingy on the underside which i'm not sure whether if its just a central muffler or a secondary cat.

Sankar, if there is a secondary catalytic converter in your car then you will be able to see it easily. It will not be a part of the muffler. Let me explain how the setup would be if you had a secondary catalytic converter.


Single Catalytic Converter System Layout.
Exhaust manifold > Catcon > Primary muffler > secondary muffler

On dual catcon systems it is like this.
Exhaust manifold > Primary catcon > Secondary catcon> Primary muffler > secondary muffler.

The secondary catcon is never a part of the muffler as it is a replaceable part and hence is always kept separate.

Oh and normally in twin O2 sensor setups, the O2 sensors are placed before and after the primary catcon as this is the one that does most of the CO to CO2 conversion.

Edit: @Raag Sandhu - It's not the question of passing emissions control. Cars without catalytic converters can easily pass emission tests. But some people like me don't like having any extra lights glowing on the Instrument cluster, especially the CEL.

@Sankar - If possible can you measure the pipe diameter of the stock exhaust manifold and the after market one. I have a feeling the after market one is going to be slightly restrictive at higher RPM's.