So I failed Emissions

The 1991 Civic So that I just bought did not pass emissions. It failed on NOX. 1951 ppm at high speed and 2056 at low speed. Standards are 1123 and 1242. Everything else was good.
The emissions guys says "you need a new cat" and sent me on my way. What else could it be?
The engine has a slight head shave, long tube headers and an intake and a chip as well as a piggy back computer that mucks with the map signal. I know it gives a -4% correction at the engine speeds they were running. It's also 50 degrees out and the car was running cool.
Any ideas?

A bad cat on an old FE isn't uncommon, but I don't know about all the other changes your car has. How "cool" is "cool"?

run the tank as close to empty as you can and fill it with E85 or dry gas then run it through emissions again

1/4 of the gauge. The car came up to temp and I could see the thermostat open. No numbers on the gauge of course, just a little cooler than I had seen. Probably not relavent, I'm just looking at stuff. Oh, and I checked timing anf it's right on the mark. Aftermarket crank pulley so there is just the one mark.
Edit: 18 degrees.

No lie seafoam will get it to pass. Always seems to work for me on Honda's. I make more money taking cars that won't pass smog then I do on the exotics.

Also make sure the cats are good and hot, I always took a 30 min drive to make sure they were really warmed up if a car was borderline for passing emissions.

Methyl Hydrate did the trick for me, back when I was trying to get my thrashed f100 through.

This is the SCCA street-whatever class car right? Does it have some sort of high flow or gutted cat? I've had more than one car fail BADLY only to find that the cat had no core- add cheap universal cat, pass test.

Does the car have a shaved head or other high compression modification? If I recall my Air Pollution class correctly more NOx is formed under high compression. Add a urea injection system like a new diesel maybe?

What is this emissions test you talk about? Move back to Michigan and you wont have such frippery to contend with.

No, I have nothing constructive to say, I just have to poke you when I can to prove your life is not entirely charmed :)

In reply to Adrian_Thompson:

Not cool man. Not cool.

mazdeuce wrote: In reply to Adrian_Thompson: Not cool man. Not cool.

You're right! Not cool....COLD.

No emissions on cars before 1995 here in NC. No experience to help you.

But your Civic is awesome!

Ok, so looking at the emissions sheet, NOX is the only thing not in line. This SHOULD indicate that it's a combustion temp issue. Lean or advanced timing or both. I pulled the piggyback computer so it's not mucking with the MAP signal. I sucked 1/3 can of sea foam through it. I'm going to bump the base timing back to 16 degrees and try again.
Sound like a plan?

wearymicrobe wrote: No lie seafoam will get it to pass. Always seems to work for me on Honda's. I make more money taking cars that won't pass smog then I do on the exotics.

A whole can in the gastank? Or are you throwing it in the oil too?

Adrian_Thompson wrote: What is this emissions test you talk about? Move back to Michigan and you wont have such frippery to contend with. No, I have nothing constructive to say, I just have to poke you when I can to prove your life is not entirely charmed :)

Ditto,

you just need a registration fee and brake/turn lights and you can register a car here, its awesome! However the motor city/state has the WORST roads in the country, irony at its finest.

mazdeuce wrote: Ok, so looking at the emissions sheet, NOX is the only thing not in line. This SHOULD indicate that it's a combustion temp issue. Lean or advanced timing or both. I pulled the piggyback computer so it's not mucking with the MAP signal. I sucked 1/3 can of sea foam through it. I'm going to bump the base timing back to 16 degrees and try again. Sound like a plan?

What were HC and CO?

High NOx can be a dead cat, but it also can be that the engine is running lean- so HC and CO are very low. Which can be caused by a leak, or even a dead O2 senor as well as the catalyst. The combustion temp thing will up the NOx some, but not that far out.

If I recall, that car has a single wire O2 sensor, which should be easy to monitor. The odd thing- those usually fail at a low voltage, which would tell the computer to go rich. If it's switching +- around ~.5 volts, it's probably ok.

In reply to bgkast:

hence the dry gas :) Alcohol won't make NOx

Yes, single wire sensor. Car runs fantastic.
CO was .1 low and .11 high (fail point of .86 and .89) HC was 60 and 75 of 153 and 159 standard. O2 was 0.7 and 1.0.

In reply to mazdeuce:

With those numbers, I'd start looking for a small leak that's causing it to be barely lean. TP O2 should be close to 0, especially when HC and CO are reasonably low.

Also check that the O2 sensor is switching- you should be able to monitor the line while it's running with a good volt meter.

Still, you have a cat that has seen it's better days. It's not that alive- enough to pass, but not for too many more years.

Texas has a rolling 25 year emission exemption. I need to make it 9 more months. In reality, if it needs a new cat I can legally drive it on paper plates until summer, and I'll probably just tow to the 5-6 event in the fall until I get to January 1 of next year.
I know the Apexi controller was reducing the MAP signal by 4% at every RPM below 3500. I'm not entirely sure why the PO did this, but any thought on the possibility that this would be enough to lean it our enough?
I'll test the O2 sensor this afternoon. Single wire sensors are amazingly cheap.

So it's running 4% lean all the time? Stop it from doing that, and you probably will be ok. I thought it was at least running stoich off the O2 sensor.

From what I can tell (it's confusing) the Apexi controller grabs and manipulates the MAP signal before sending it back to the ECU. A negative correction says there is really less air there so add less fuel and a positive correction says more air and more fuel. If you believe the book, it compensates directly as a percentage, so a -4% correction is telling the computer there is 4% less air than the MAP is saying there is, so it should run lean.
The part that is problemstic is that the car also has a tuned chip of some sort, and who knows what that does. That's why I left everything alone for the first go round with testing. I know people use the Apexi controller to lean out cars at cruise for MPG purposes, so maybe that's what's happening.I do know that it fattened up the fuel above 3500 and above 40% throttle. A significant stack of dyno charts accompanied the car, so someone has spent time trying to extract maximum power.

Did the previous owner block off the EGR, or are the passages clogged? EGR lowers combustion tension to lower NOX.

mazdeuce wrote: A significant stack of dyno charts accompanied the car, so someone has spent time trying to extract maximum power. Poorly

FTFY

I'd probably try a new cat. They aren't too expensive for the 49-state "high flow" jobbies and you can eliminate one possibility right off the bat.

If that fails, then maybe adding fuel via the Apexi piece slightly to try and bring the NOx numbers down would also be a potentially quick solution.

Then you might look at removing the aftermarket chip and Apexi stuff and getting back to stock EFI solution (even temporarily) couldn't hurt either if the above fails.

That or figure out how hack the stock ECU or install a more modern EFI solution that you might actually stand a chance of tuning would be the paths I'd take.

I know on my 951 with the aftermarket EFI, I had to remove some fuel from the idle tables to get it to pass properly. It didn't idle very well, but I passed and while in the parking lot, I added the fuel back in. Now that I fixed the cam timing and put a OE thermostat in, it seems to not run anywhere near as rich at idle anymore.

What's emissions?