Thread: How to: Pass smog (Nismotronic+Wideband)

My NX is due for inspection, and I've changed so many things on my car since last year, so I need to plan my smog test accordingly.

The three biggest things that have changed since last years inspection:
-SR20VE Swap
-Nismotronic install
-Narrowband 02 sensor deleted, Wideband installed (Innovate MTX-L)

There are two things that are going to affect my emissions in the Nismotronic program, when using wideband closed loop.
-WBO2 Target AFR Table
-WB02 PID Gain

I'm due for my test in 30mins. I've got a local garage that lets my test it there, so I can fine tune my tables whilst measuring. When Ireturn, I'll upload my settings so someone in the same boat can hopefully find help from this.

Watch this space

And: I Failed.

Not because of the config, but because of my 23-year old catalytic converter. The flanges have rusted away so much, that air is allowed to enter the system. This in turn makes it impossible to smog test, since it's throwing the lambda value way off.

Hmm, reweld the flanges or get a new cat? Decisions, decisions... Here's a pic from last year, bodged it up with exhaustpaste last time:


Keep watching this space! Might take me a few days to get the exhaust fixed

Protip, semirelated: If you have a wideband with a narrowband output, take advantage of it! It's much quicker for the ECU to make changes with a narrowband input, than it is with a wideband. But I'm all for going the more advanced route, so I'll make it work with the WB02 tables & gain

new cat

Yessir, new cat ordered! Update within 5 days then..

Cat arrived today, 169€, 2.5" Nissan Universal fit (1.3L->3.0L according to specs)

Time to rip the old one out:


This is why I can't get any good readings


Old vs New. Even supplied with new metal reusable gaskets and bolts.


Smog retest in 18h!

Retest #2: Failed

Idle;
Lambda (0.97-1.03): Pass
NOx: Pass
CO: Pass

Fast Idle;
All failed to some degree. I could have the lambda spot on, but then the CO was too high.

I tried to disable closed loop, and just tune the fuel table. Worked to some degree, but CO was still too high.

Hmm. Any tips except getting a narrowband? Maybe try the Innovate simulated narrowband output?
My WB02 is mounted 10cm from the head, and measuring on two cylinders (3&4). Maybe I should move it downstream?

I tried wiring the simulated narrowband output from the Innovate unit to ECU pin 19. It seems to be more stable at 2000rpm, and make adjustments quicker and more accurately.

Thrid retest in 18h!

What causes the high CO2?

Is the cat getting hot enough before testing? Was it installed in the correct direction of flow? I only ask as I've done simple things like that, no insults implied

Not quick enough regulation from o2-sensor, causing it to run rich for too long before compensating. See below



Cat is 100% the right way. Drove 30min in 4th gear 60mph before test. It was hot

So the results are in for todays #3 smog test,

Result: Fail

Idle;
Lambda (0.97-1.03): Pass
NOx: Pass
CO: Pass
O2: 0.6%

Fast Idle;
Lambda (0.97-1.03): Fail (1.05)
NOx: Pass
CO: Pass
O2: 1.6%

So, to summarize; switching to narrowband was a great choice. The ECU now makes corrections more quickly, and CO emissions is no longer a problem. The test still failed because of too much oxygen in the exhaust gas, which in turn affected the Lambda value and made it too lean.

Guess I have a super hidden microscopic exhaustleak that is allowing air to enter it at high rpm. Lets start tracking it down...