This is essentially the third time I'm making this thread. But now I actually have enough experience/hardware to actually do something about it properly this time.

Trying to get my 1995 Miata to pass Arizona emissions. They stick an exhaust gas sniffer up the cars tailpipe and then simulate some city driving on a dyno. They are checking 3 gasses HC (hydrocarbons), CO, (carbon monoxide), and NOX (nitrous oxide).

I've recently gotten a wideband and finally fixed the connection issues between my laptop and ECU so I can start tuning since I've made some significant changes to my car since it was first dyno tuned. After a several runs with autotune combined with some manual edits and fuel VE table smoothing, this is the tune I've ended up at, and a data log of some driving on it. This tune wasn't optimized for emissions, this is the tune I plan on running on the street. It still needs some work, have only tuned moderate street and highway driving, haven't touched the driving like a jackass part of the fuel map yet.

This time around I failed emission at:
HC 1.49/0.80 PPM
CO 20.62/12.00 PPM
NOX 2.15/2.00 PPM

Usually I always pass NOX by a good margin (usually around 1.6/2.00ppm) and then fail HC and CO right under 2x the limit. but this time I failed all of them by a decent amount.

Mod List (the stuff that will impact emissions)
Rebuilt motor (forged internals with 9.5:1 compression)
cold air intake (actual Cai, not a hot air intake)
raceland Headers
stock cat and exhaust
MS3 ecu
NB idle valve
Egr delete
91 octane gas (cant get 93 in A)

Car doesn't have anything to dramatic done to it and still has a cat, so some basic tune changes should be all that's needed. I got a data log of the actual test, but for some reason the AFR reading is all fucked up, but it at least shows the RPM and load areas that they will be testing around. Tune and a data log of normal driving and the actual test is attached. What should I change in the tune, in order to finally pass emissions?

 

NOx is primarily dealt with via EGR. You have deleted the EGR. Raising the compression over stock (8.8:1 stock in 1995?) increases likelihood of combustion temps being high enough to generate NOx.

CO/HC being elevated indicates rich mixtures. The NOx might squeak by- the tune needs fixed so it's not super rich. The catalytic converter is designed to removed remaining byproducts of efficient combustion, not be a sponge that infinitely swallows HC/CO. Age alone with a stock car may require replacement of the converter, my 94 passed tail-pipe tests with a factory cat with ~80k on it. You also need to confirm the vehicle has the timing set correctly. If the AFR is right in the load cells being measured, and car cannot meet CO/HC, the converter has degraded. I ran 14.7:1 in all the cruise/low load cells on my 94 chasis with a megasquirt, VVT engine swapped. 14.7 will mean minimal HC/CO and peak CO2, while not raising NOx. There's a reason all the stock stuff targets this.

I would be curious what a stock ECU and functional EGR would do with non-factory compression ratio. I don't have tunerstudio on this laptop - just general information.

 

Turn on EGO at idle too. Turn the authority table on too. You might put a hotter thermostat in it, its only running in the low 180s, high 190s or even low 200s would be better. Idle ignition is pretty aggressive, pulling/adding up to 20 degrees. Your idle seems pretty stable, maybe turn that down to +/-5 until after emissions.

Your AFR input was not working for the emissions log, so I'm not sure if it was running the same as "example drive", but that log is fairly rich. Minorly, but you're at 14.2 at idle instead of 14.7, so the idle EGO will help. That's a common misconception, VEAL shouldn't be used at idle, but EGO can be once idle is tuned.

Otherwise it looks like you've optimized the tune a lot for emissions, but physical things like the thermostat and what 2ndgear said may need to be addressed.

 

The egr will reduce the NOx and CO. Without the egr you need to have an exceptional cat to pass and the cat has to be hot to function efficiently. Maybe two cat’s inline. Lower timing will increase the egts which will help keep the cats at temp. I often will tune for the tests with a little methanol mix because it reduces the NOx and CO. There is an off the shelf methanol product called Heet in the yellow bottle. I usually add 2 bottles to 5gal.

You should have a test only set of plugs. I usually use the NGK BKR5EIX-11, 5464 gapped at the as delivered 1.1mm gap. The afr will need to be around 15:1 at idle with 8-10 degs of advance. If a dyno test, under load you can’t be too lean or you will have high NOx and CO. The cat has to take care of the HC. Timing under load can be a challenge. Lower is typically better. Maybe a max of 23-24. This will keep the NOx lower and the cat temp up. Ideally the best way to tune for the test is with a 3-4 gas analyzer beforehand.









,

 

My cat is newish (about 4000 miles on it) so it shouldn't be degraded too heavily. I have a 185 degree thermostat, I've just done a bunch of cooling upgrades to the car (reroute, crossflow radiator, hood vents, distilled water+ water wetter), so its probably over cooling a bit, switching to regular coolant should probably get me to the 190s. I used the same tune as in the example drive log, it just seems that the SD card datalogging didn't want to record AFR's correctly unlike the other log that taken by my laptop. Kinda sucks that I cant see exactly what AFR it was at when it was tested but at least I know the cells that the test will take place in, so I know where to focus most of my efforts. The idle settings (including the overly rich AFR) are a hold over bandaid solution from when my car had some other issues that I complete forgot to update.

I knew that there was 0% chance to pass with the tune, just wanted to get a baseline for how it would do emissions wise with minimal optimization.

This is the tune I've started with your guys's advice, only did a quick drive with VEAL. Just to make sure im heading in the right direction with my AFR targets and Timing changes, before I start messing with it any further.

 

Just tried again, somehow did even worse.

This attempt
HC: 2.30/80 (50% worse as last time)
CO: 18.03/12.00 (a smidge better than last time)
NOX: 4.27/2.00 (2x worse as last time)

Main changes I made was pulling some fuel and setting my afr targets to be at 14.7 in my low load/cruise cells and then took a bit of timing out. Tomorrow I'm going to undo my timing changes and see how that changes things, unless there is something obviously wrong with my tune that I'm not noticing. I used a different tune than the one I posted before so ill attach the one I used below.

 

I never had an issue passing emissions with my 91 on MS1, MS2, or MS3 with my basetune i provide on trubokitty.com.


you numbers suggest it still way too rich and way too advanced. i would lean it out, retard timing, and use way colder spark plugs.

 

that’s what I did in the second test, leaned it out to 14.7 in the low load/cruise cells and pulled timing and it did way worse.

tbh now I’m starting think that something isn’t mechanically right with the car, maybe I have an exhaust leak or my cat is dying. I’m gonna confirm that everything is working properly before I start trying again.

 

Emissions under load are very difficult to tune for. Manufacturers have enormous budgets and still resort to cheating. There is probably no way to pass without an egr and an aftermarket cat. An egr and factory cat would ideal and second would be an egr with an aftermarket cat. No egr, a clean high functioning factory cat and some time on a dyno with a 3 gas analyzer might be doable. The nox and co will be lowered with cooler combustion, but the hc will be high which is what the cat needs to take care of. Since you are NA you want a factory heat range plug with a relatively large gap.

 

wouldn't he want a colder plug to help dump more heat into the coolant? Nox is already high, and to cool it you need to dump in fuel -- but HC/CO is already too high. So the only real other option here is to retard timing aggressively and I agree open a the gap on the plugs to promote a better burn.


Are we sure the cat is even working? I had no issues passing sniffer tests with a 3" 300 cell metal core cat.


Make sure your wbo2 is reading correctly AND your ignition timing in MS is sync'd properly and not still locked at 10°.

 

I verified the wideband works correctly and I only locked it to 10° 2 years ago because I have no idea what I was doing back then, I'm using an actual timing table now. I did however discover that my headers are cracked (4 year old ebay headers... what a shock), so I'm gonna replace the headers before I start trying again.

 

Plug heat range will have no effect on the combustion temp. The plug needs to operate at a temp that will keep it clean to prevent carbon tracking. The typical non-enhanced tail pipe sniffer test is many times less difficult to pass than a load test on a dyno. When the afr is lean, numerically high the distance between the fuel molecules is increased which requires a longer spark plasma to ignite the mixture. What might work if the inspection does not confirm the cat’s service number could be any oem cat from a late model vehicle.

 

The equipment “inspection” they do here is not very thorough. They don’t even bother to check and see if I even have a catalytic converter. Both egr and evap are deleted, and they don’t even check the egr, just evap. The test they do for evap is very easy to fool, they check to see if the gas cap and fuel tank will hold pressure, and see if the pressure in the evap line doesn’t change pressure. So I put a vacuum plug on my fuel tanks evap port (only for the test, I’m not dumb to drive around like that) and then point to a dummy line as say that it’s my evap line. They once they “test” the evap systems they never check it again on future tests (within a month of the first test.

I still have an uphill battle trying to pass emissions, but atleast I only have to focus on the gases and can cheat however much I want.

 

Maybe saying into the coolant was the wrong choice of words, just the heat of the plug itself into the head and away from the electrode to prevent pre-ignition. Colder plugs have a smaller porcelain section that's exposed to heat, making them literally heat up slower and stay cooler.

Since there's no EGR to cool the combustion chamber, running a step colder plug could help prevent any pre-ignition that's happening and get a more efficient burn.

 

NA engines don’t usually have plug overheating issues, especially this vintage of engine with its relatively low output. Also the dyno test is not checking for max output. Typically they’re testing idle, lean cruise at 45mph and 55/65mph. 1 heat range colder plug should be used for normal use and maybe gapped slightly less at 035-.040”.