Failed SMOG in CA, got new Cat, Still failed for a DIFFERENT reason

[sargentpilcher]

Hello all, I'd like to thank you all in advance for any help or advice you're willing to give.

I drive a '97 Honda Civic DX with 180,000 miles on it, and I've owned it for 6 years now. I have a long history of smog failure with this car (Even using up my once per vehicle waiver from the state). 2 years ago I got passed TO THE NUMBER on my NOx measurement. This time, I couldn't get it down low enough despite getting my engine hot for the cat. So I broke down and got a new catalytic converter, the legit one straight from honda for 800$ plus tax. I replaced it myself, and to my crushing dissapointment STILL failed but for an entirely different reason. Now my CO was too high, where before I had a brand new cat, it measured half of the maximum.

I took it to a mechanic friend of mine who told me that I should get a new downstream 02 sensor, so I did that, and it got the number down a little bit, but not enough. He recommended I get a new Idle Air Control Valve for it (As my car doesn't have an EGR), but I don't know if I can trust him because I gave him money for the part and he took off with it (Haven't heard from him for 2 weeks). Before I go and spend more money on this car I wanted to ask for some advice on what could be causing it, and if my "friend" is full of crap?

I do regular oil changes. Air filters, oil filters have all been replaced within the last 10,000 miles.

- Catalytic Converter replaced a month ago
- downstream 02 sensor replaced 2 weeks ago
- upstream 02 sensor replaced 2 years ago
- Fuel injectors replaced 4 years ago
- Battery and spark plugs replaced 2 years ago
- ECM replaced 4 years ago


Any advice or ideas or opinions would be incredibly helpful!

 

[sargentpilcher]

There was a "hidden" code of a P0420. I say hidden because it didn't have a check engine light, but if you plug it into a OBD-II reader it would show. That code is the reason he recommended I get the new downstream 02 sensor.

 

[sargentpilcher]

Sorry, I should have been more clear. The "Pre Cat" smog picture is my smog failure before I replaced my catalytic converter where my CO was fine but my NOx was outrageous (Measured 544, MAX allowed is 487).
Post Cat picture is my smog fail with a brand new catalytic converter only now my NOx is fantastic, but my CO is off the charts (Measured 1.67, MAX allowed is 0.55).

So the 2 readings switched with the only difference being a brand new catalytic converter. What could be wrong with the car that would cause that to happen?

1st test was done January 4th, and the 2nd test was done early february. Different locations. On a Dyno with a sniffer up my cars butt.

 

[sargentpilcher]

A bad secondary O2 sensor would not cause P0420. Dump that mechanic posthaste.

Oh he's dumped. Air filter was replaced about 10,000 miles ago. Upper limit on those is 25,000 right? I didn't think it would make that big of a jump on the CO, but I will definitely replace it if you think it will help though.

 

[sargentpilcher]

Start there because an air filter is cheap. Also check the entire air intake and throttle body for anything that might restrict air flow.

Bone stock Civic?

Ok I will definitely replace the air filter, and I'll check out the air if that could be a problem. I just don't see why those things wouldn't have made the CO higher when my cat was shitty, or maybe I just don't understand the way it works properly.

Sorry, I would have answered you before but I thought you were making a joke or something. I had to google what a bone stock car meant. Yes, it's bone stock. No modifications at all.

 

[XpL0d3r]

Here's a good, detailed link about that code, which may also help you out.

http://www.easterncatalytic.com/education/tech-tips/the-misleading-nature-of-the-po42o-code/

 

[MotorMo]

The "CO" is a measurement of the air to fuel ratio before the combustion. Since your NOx is now low, and your CO is now high, I would assume that your car is running a bit rich rather than having an unmetered air issue (vacuum leak).
The four main pieces that tell the ECU how much fuel to give are the MAP sensor, ECT, IAT, and the primary O2 sensor. If your IAT, or ECT are showing a colder than actual reading, your car will run rich. Very rare for a MAP sensor to fail, but I have seen aftermarket O2 sensors cause more harm than good. If you didn't install an NTK, Denso, or a Honda original part, I would recommend swapping it out. I have even seen a Bosch O2 sensor cause stalling issues in a Honda. Good luck.
One last note- make sure your car is up to operating temp before you smog it. As that can also change the readings.
- Also make sure that after clearing all codes (even hidden ones) that you run the OBD2 "monitors". Smog shops can give you an automatic fail if any monitors haven't run, or if there is a code in the system. I think 97 year models are allowed to have all monitors ran except for one. But double check that with your local smog shop. Gotta love California regulations.

 

[sargentpilcher]

The "CO" is a measurement of the air to fuel ratio before the combustion. Since your NOx is now low, and your CO is now high, I would assume that your car is running a bit rich rather than having an unmetered air issue (vacuum leak).
The four main pieces that tell the ECU how much fuel to give are the MAP sensor, ECT, IAT, and the primary O2 sensor. If your IAT, or ECT are showing a colder than actual reading, your car will run rich. Very rare for a MAP sensor to fail, but I have seen aftermarket O2 sensors cause more harm than good. If you didn't install an NTK, Denso, or a Honda original part, I would recommend swapping it out. I have even seen a Bosch O2 sensor cause stalling issues in a Honda. Good luck.
One last note- make sure your car is up to operating temp before you smog it. As that can also change the readings.
- Also make sure that after clearing all codes (even hidden ones) that you run the OBD2 "monitors". Smog shops can give you an automatic fail if any monitors haven't run, or if there is a code in the system. I think 97 year models are allowed to have all monitors ran except for one. But double check that with your local smog shop. Gotta love California regulations.

Is there a way to test the MAP sensor before i replace it? With a multimeter or anything like that? I have an OBD-II to USB cable and some software to get readings from my car computer. Anything in particular I should look for to isolate it? It could easily cost 1,000$ replacing all those parts with genuine honda parts.

Is it a safe assumption that, since replacing the cat and 2nd O2 sensor, no real or pending CEL codes have been thrown?

Yes definitely safe to say that.

FYI - The emissions sheet clearly shows that the engine is not running rich.

How can you tell?

 

[sargentpilcher]

The HC numbers are fine, indicating that the engine is NOT running rich.

Focus on the entire air intake system, including the IAT and TPS sensors.

Any engine performance problems?

I see, that makes sense. I just took my car out for a drive with my OBDII connected to my laptop to get some readings, and a P0133 was triggered which is a malfunctioning upstream 02 sensor so I'm definitely going to replace that.

But I also want to be sure it's not something else too. That is, I don't want to buy an 02 sensor, then pay another 60$ for it to fail again.

I took a picture of my readings here
http://imgur.com/a/CD85M

and uploaded a video of the readings live

Do you think I should get one from Honda? Or do you think it would be safe to get an aftermarket like a BOSCH?


Also, rarely do I have performance problems, but a couple days ago I had a really hard time going up a hill. Like I had to push on the gas pedal way harder than I normally would to get it to move. It felt like I was dumping fuel but not getting much performance for it. But then today on my drive getting these readings it felt fine the whole time.

 

[MotorMo]

If you don't want to buy a 3rd O2 senor next month, don't buy Bosch. My 2 cents.

 

[sargentpilcher]

Ok, I got the new 02 sensor in the mail and installed it (I got NGK as I did some research and that's who Honda uses). Some more strange behaviour resulted however that I'm trying to wrap my head around.

I've been reading up on fuel trims (Short term and long term), and how and why they do what they do (I found an amazing youtube explanation). So I plugged my car into my laptop and got ready to take some readings with this new knowledge. I read that it's best to take some measurements when you're at idle, and when you're at 2500 RPM, which I did and everything was looking good, but then afterwards I took it for a drive, and my brand new 02 sensor was not functioning properly at all (Staying at about .9v the whole time), even though it was fine during the idle/2500RPM in neutral test I did.

I decided to take it for a longer drive for 10 miles last night to get some readings and it was not looking good for me. 02 sensor error code P0133 popped up again despite it being a brand new 02 sensor, and that it was working hours earlier. So I decided to try something different. I reset the code and my whole computer. A warning popped up "Are you sure you want to erase all these settings in the computer?" (It was a lot of settings), and I drove 10 miles back, and everything looked good again! The 02 sensor started functioning properly, my short term fuel trim was jumping around everywhere but the "mean" measurement was 5% which from what I read is pretty damn good. The long term fuel trim was perfect as it landed right on 0% so there was nothing weird going on. I recorded all of the drives so I still have all the data to provide and back this up.

I'm just trying to figure out why resetting the computer like that would "fix" my problem. To be honest I'm not sure how fixed it is, and I'm scared that it's going to go right back to being broken. Only driving it more and taking more tests will tell but it's unsettling when I don't understand what is causing it. 02 sensor functions, then stops functioning, then after a computer reset starts functioning again. Any possible explanations?

 

[XpL0d3r]

I'm just trying to figure out why resetting the computer like that would "fix" my problem. To be honest I'm not sure how fixed it is, and I'm scared that it's going to go right back to being broken. Only driving it more and taking more tests will tell but it's unsettling when I don't understand what is causing it. 02 sensor functions, then stops functioning, then after a computer reset starts functioning again. Any possible explanations?

The ECU will hold onto stored codes related to sensors, and won't clear automatically. Using the scan tool to clear the codes will set the sensors back to a not-ready state. After driving around for 20-30 minutes, the o2 sensors give enough feedback to the ECU and switch to a ready state.

EDIT: @boofoo beat me to it lol