assuming the test doesn't care there's visually a turbo on there...
-wire the wastegate wide open
-swap the injectors back to stock
-swap the MSPNP back to the stock ECU
-make sure the stock MAF is connected
?

or am i missing some stuff?
i'd love to boost an NB but i live in massachusetts...

 

yep, make sure all the monitors are "ready" and you're good.
the biggest problem for us is visual. if you don't have it, its pretty simple

 

Depends on your laws. If it's just an OBD2 check, yes that would be good. If it's a sniffer, then you should still be good, that will run close enough to a stock Miata, as long as you retained a cat. If it's visual, then obviously it depends on how in depth they get.

 

had a bear of a time finding if MA has a visual inspection aspect for emissions testing but it seems they only look to see if you have a cat. No Cat no deal

 

The term "visual" in this context mostly applies to California.

While nearly all states (and the Fed), have laws that prohibit tampering with emissions equipment, California is really the only state that vigorously enforces this with a mandatory visual inspection which goes beyond checking for the presence of a catalytic converter. In nearly all other states, they don't really care what you have under the hood so long as the cat is there and you haven't obviously done any massive tampering (eg: replacing the stock engine with one that burns coal and emits scorpions as a combustion byproduct.)

By contrast, in CA you can have the cleanest-burning engine in the world, which is 100% stock except for an aftermarket air filter, and you'll fail unless that filter has a CARB EO sticker on it.

 

yep. good 'ol nazifornia

 

One other thing to watch out for is the NBO2 sensors. If you kept them in the pipe, but disconnected, there's a fair chance that one, or both of them might be toast because the heaters weren't on while you were boosting. If you've got a Scangage then keep an eye on the voltage levels on the NBO2's when they are reconnected. The rear one should slightly lag the front one in time and the voltage on the rear one should be slightly lower (peak-to-peak) than the front one - this is what the ECU is looking for to determine that the cat is working. If you throw a P0420 don't panic - it might just be that one of the NBO2 sensors has crapped out.

If you don't have the EGR valve on the IM (mine is blocked off with a plate) you'll have to hook the valve up temporarily so you don't get a P0401 code (or some other EGR-related code). Also, if you have done an exhintake cam swap, you'll probably throw a P0300 (random misfire) code. You'll either have to put the stock cam back in, or use a variable cam gear on the intake to make the exhintake work with the stock ECU.

As was also mentioned, make sure that the Readiness Monitors are green. In GA, for my vintage of car, you can have up to two monitors not-green and still pass. Just know that the EGR monitor is usually the last one to get to go green.

One thing's for sure...you'll learn more than you ever wanted to know about OBD MIL's and how to fix them by doing this!

 

If its a second car consider looking up mileage exemptions. I bought my car stock, got it inspected, and then started to add things like megasquirt and turbo. But because its a weekend/fair weather car I put less than 5k miles on it a year which makes it emissions exempt in my area. I just keep all of my crap on it and it gets a sticker year after year. I even have a cat but all of the places I have taken it to didnt even lift the car to look at it. I guess it also depends where you take it. Worth considering if its available to you and you can meet the restrictions.

 

This will be the biggest, hairiest obstacle. If you retain the turbo manifold for the test, adding a bung on the runner is in your future (a real bitch if its a cast manifold, very limited options), as well as bending a new EGR tube. Tube isn't a big deal if you have material and a flare tool. And then you have to hope a "custom" set up satisfies the EGR valve and its operation, and if it doesn't, well there goes a bunch of hours down the drain as you will still have a CEL.

Swapping the manifold for a stock header/EGR set-up and plugging your oil/coolant lines starts looking very attractive.

 

AFAIK the EGR valve doesn't really "know" if any exhaust gases are actually flowing between the exhaust manifold and the intake manifold. The only thing the ECU "knows" (again, AFAIK) is whether the valve opens and closes based on the signals sent. Based on this assumption, one could have the EGR valve in free air (as I do) but connected up to the connector and, as long as the valve actuates appropriately, the ECU will be satisfied.

Which in my case works just fine. As a matter of fact, in preparation for my last OBD check (3 weeks ago), the EGR readiness monitor went green. This has always been the one monitor that rarely, if ever, has gone green for me. Empirical evidence that you don't have to have all of the plumbing in place, just having the EGR connected to the ECU (and functioning properly) is "good enough".

BTW, I have the cast FM manifold, and it does have the (plugged) port for an EGR pipe, but I've never used it.

 

The EGR valve doesn't but the EGR Boost Sensor does. I'm surprised you're not dealing with a P1195 code.

 

This. And that boost sensor's return signal to the ECU, determines when the EGR valve will open/close. From what I have read this is how this system works. Yes, it is completely possible with the sensor in free-air, to return the same values and "satisfy" the ECU, as if it were installed in OEM fashion. Albeit a rare occurrence, as RyWatt explains. I toyed with the idea of a simulator to send return values from the boost sensor to the ECU, but it wasn't worth the time. Especially since I discovered you can get two OBD2 emissions tests within a few days of each other and have the other one for the next calendar year. Swapping stuff once every 2 years instead of 1 year finalized my decision.

 

Ah, but the stock boost sensor is still in place, so no P1195!

 

That's interesting.

Based upon the description of the system in the service manual, I'd have expected the fault to still occur if the sensor was in place but the valve wasn't. That, after all, is the whole point of having the sensor in the first place.

If this is not the case, then it suggests an interesting vulnerability in the software,