Auto-tune & Emission testing

[DonMaximo]

I recently failed a Texas Emissions test and I 'suspect' a contributing factor may be that my injectors were too large for my engine and I should retard timing for the Idle test (hotter EGT reduces emissions). Was running 42# on N/A 363" which produced 460HP on engine Dyno. All online Injector Calculators I've used recommend 36#. So I have a set coming in. (how I missed this earlier is beyond me, getting old sux)

I wanted to make use of the Auto-Tune feature while Emission testing, where I set ALL AFR cells = 14.7 for the RPM range the Emission test covers. Yes, I incorporate AFR target. The technician conducts 2 tests, 1 at idle, the other at High Speed (HS) running 25MPH on a dyno. Each test runs for 90 seconds with only a few seconds in between tests.

I've driven around town on my Emissions tune in order to have tables 'close to the mark' and have the car operate in a civilized manner.
.
I wanted to make use of the autotune feature for the testing as the dyno test is not easily simulated on the road and "I thought" (right/wrong/indifferent), that if I can get an AFR of 14.7 for both types of tests, that I should be in a good position. My concern is related to any latencies associated with the Auto-Tuner. In other words, I don't know how many seconds must pass before the tuning alters VE values in order to meet AFR and is this latency magnified based upon how far the VE values are from where they should be.

While I can influence Exhaust Gas Temps (EGT) by retarding timing, I was wondering what, if any, other recommendations or suggestions you may have.

As always, your assistance is greatly appreciated.

-D

[LT401Vette]

I would expect the 42 lb injectors to be able to pull it off, it is only a 16% bigger than the 36 lb/hr.

On your test what failed? Was it NOx or HC's that caused the miss. I would be careful with the assumptions of what will fix it, lowering NOx's is pretty much dead backwards from lowering HC's.

Anyway on AutoTune, it will sends the updates to the to the MegaSquirt every 15 seconds while Update Controller is checked.

[elaw]

I'll second what Phil said - it would help tremendously to know what tests it failed and what the numbers were. A log showing conditions as close as possible to those in the test would probably help too.

There's a chart about halfway down this page: http://www.teamswift.net/viewtopic.php?f=31&t=57408 that shows how CO, NOx, and HC are affected by rich and lean mixtures. Simply put, a rich mixture increases CO a lot and HC a little, and decreases NOx. As the mixture goes leaner from stoichiometric, CO remains steady at a low level, HC slowly drops until you reach the point where the engine misfires then it increases like crazy, and NOx increases then decreases.

Timing can also affect emissions and from what I understand NOx is affected much more than the other two. In general, retarding timing decreases NOx and advancing does the opposite. So if your NOx numbers are high and the other two are okay, pulling some timing might help.

One other thing: if you have an aggressive cam, passing emissions could be difficult or impossible. Most "hot" cams make the engine idle poorly, and tuners compensate by enriching the mixture. The rich mixture will cause the CO test to fail, and if you try to lean the mixture to meet the CO goal, the engine will misfire and fail on HC.

[DonMaximo]

The Saga continues as I failed Emissions again yesterday

FYI: This car/engine has passed emissions in the past using the same cam so I don't believe the cam is the issue.

Since I Incorporate AFR targets, I Increased the AFR target table entries. I drove around a bit, attempting to simulate the test conditions by putting car in neutral and run RPMs between 2200 and 2800. I get idle AFRs into the 15's.

Despite this, I still fail on high CO(%) (Carbon Monoxide) on the low speed test. When you look at the Datalog, I'm clearly running lean, so I left the Inspection station scratching my head.

As I understand it, CO results from unburned fuel.

Regarding the datalog:
The Emission test starts with a High Speed test (RPM=2443) for 90 seconds and is immediately followed by the engine idling for 90 seconds. Is this 2nd test, while idling, that they call the "Low Speed" test. The datalog shows the very end of the High Speed test and transitions into the Low Speed/Idle test.

Having recently replaced plugs and wires, this leads me to suggest that this MSD 6AL box, which caused me significant headaches setting inital timing on my MSPNP2 Mustang 5.0 unit, may be suspect. I've already disabled use of the MSD6AL and the stock ignition is now using a standard coil. I am also consider replacement of the MSD dizzy, with the Stock Ford TFI distributor. I am not too concerned with these changes as I have aspirations to go with wasted spark once this Emission testing is behind me

The question in my mind, short of going back to the shop for another emissions test is: "How do I determine if any of these changes help ?"

I will next check the plugs to see if I can identify a plug which is not burning. I would be happy to find a plug wire was not seated properly and contributed to unburned fuel appearing in the exhaust.

Any feedback would be greatly appreciated.

[LT401Vette]

It is tough to cure without the instrumentation to test with, your sort of flying blind.


It looks like you did great on the high speed test and it looks like they aren't testing NOx's as that can't 0.

Looking at your log, it looks like you are trying to run it quite lean. and your HC's made it through.
However as the CO still failed... Often with a good sided cam as it seems you have idling at 70+ kPa, the WB sensors don't read so accurately at idle.
Normally unburnt fuel leads to high HC's more than CO, high CO is impacted more by a rich burn.
Does it seem to be hitting al the cyl well? I see some oscillation in the idle, that usually leads to slightly higher HC/CO numbers as well.

But overall, I would turn to the ignition timing, that appears to be down at 11 degrees, the lower spark timing will cause additional CO. Get the idle timing up to the 20 degree range and I bet that will help CO levels a good bit.

[DonMaximo]

Phil,

Thank you for the quick reply.

As for the timing, I intentionally retarded with the intent of raising gas temperatures which can help the catalytic converters burn off unspent fuel, but in hindsight I also understand how this may contribute to an incomplete burn.

I was not entirely comfortable with the engine hitting 220+F on the high speed test. You can observe this from the beginning of the log file. So, adding more advance will have me be a slight more comfortable.

I'll take your advice regarding the timing. I'll add a little more advance for the high-speed test, but will increase idle timing considerably (from:11 to: 16,18,or 20).