AFR Target and Spark Standards...
 

Just wondering if there is a standard table for these for like anyone using MS on a 99. I understand everybody uses different settings, cause everyone's car is different. For my current application (see signature) what values should I be using for these tables? I am thinking, I shouldn't touch the Spark table, but perhaps my AFR targets look like shit. I VE analzed it yesterday after a few datalogs and it runs pretty good, I just want a second opinion of the AFR target Attachment 10942

Attachment 10943table.

post up a screen shot of the afr table and i'll look at it.

i don't use the 8x8 table in MT, I prefer the 12x12 in MLV VEA but here is what I would use in MT
http://www.miatamx5.com/ms_files/afr.gif

The above link "afrtargets.vex" is my AFR target table, I clicked on export. For some dumbassed reason my computer is being difficult doing a screen shot. But yeah. my table was pretty different than the one you posted. I'll make a duplicate of the posted one, and then I'll give it a try.
Thanks.

Yeah I know what a vex file is. I didn't feel like downloading it, starting up megatune, and importing your vex file just to view it.

Hi paul! My table doesn't look mucch like yours either.. I based mine off someone else on this forum (can't recall who, but i think they had a good reputation!)

Quick questions -- i've seen a lot of tables where atmospheric pressure (101 or so) is high 14s, 15 vs. your 13.0. In general, your table looks quite rich even in mild boost..

Am i asking for trouble with my targets?

I don't know, but my car would hate that table^^^

Here's mine, recommended by Braineack as a good starting point, I think I changed a few cells.

Attachment 207181

gospeed81, what's your max boost in PSI? I rescaled my table to better reflect my limited boost.. right now the highest kPa i hit is around 145..

-Tomaj

Right now? 14.7psi absolute.

I will max at 10psi, and am going to rescale to about 180kPa, or like yours above. Haven't rescaled as I was more concerned with putting in bigger injectors today.

From everything I've seen, it's good to have about 20kPa headroom above your expected max to accomodate spikes, and linear interpolation for the noise on the MAP sensor.

Right now I'd say your table is good for almost 6psi.



EDIT:

Here is what my table looks like now:

Attachment 207180

It is a blend of Paul's table scaled to my expected boost level, with a few things I liked from my old table.

I ran a few datalogs this morning, and ended up with a very smooth VE that I like. Chirps them in second, with no hiccups or rich overrun from the RX7 injectors.

I'd say so. What are you trying to accomplish by going leaner at atmosphere? Better fuel economy? I'm quite happy with my AFR targets and if I must say so myself my dyno graphs are pretty impressive. Truth be told though I have no idea how much power I make at 101kpa so if your goal is to beat me there by running leaner, I'll probably lose. Both my cars still average 27mpg.

FWIW, I did some testing back when I first got the parallel MS and WBO2. Hooked it up for logging purposes only, the stock ecu still controlled spark and fuel. AFR's dipped down into the high 10s at WOT over ~4100rpm. This is on a naturally aspirated car. Now that I call rich. Mine I call smart.

I like gospeeds the best...but I run 11.5:1 above 8psi.

Why do you guys have so many cells in boost in your AFR tables? Dont you run open loop when you are boosting?

You bring up a good point.

It's really a good idea to use the contour plot tool. Mine looks like this:

Attachment 207175

And it shows I could still move some MAP values around.

Basically this should have more lines where you have more drastic, non-linear changes in target AFRs. Notice mine has a lot of concentration in normal vacuum operation range.

Just like a countour map of a mountain, there shouldn't me a lot of lines in areas that will have a linear rate of change in AFRs. In reality you could have two lines in boost, from 12.5 at say 110kPa, to 12.0 at your max boost, since the program does linear interpolation anyhow.

In other words, highly tune the areas where you have transitions between vac and boost, and throttle roll-on points. This should help throttle response, boost onset etc.

I'm still a newb, and may be wrong here, but I think you're right about wasting resolution in the boosted cells. I'm going to experiment with this slowly though.

EDIT:

If hustler posts his, you will see that he does the same on the ordinate axis, with rpms, ramping up to his boost cells. Fine tuning both should be beneficial.

i guess i was going for emissions, my inspection is past due becuase i've been waiting on the MS working before i try to get inspected :)

I made some changes to my table, richened it up in boost, added more resolution near the on/off boost transition.

-Tomaj

we use more resolution for tuning, then I played with mine a lot to get the most out of EGO correction in cruise.

No, do you?

Even if you do go open-loop in boost, there's value in having valid data up there so that VEA has something to chew on.

OK so i went to the MVC (think DMV) inspection station today... For you non New Jersians, we have state run facilities as well as private. It was free so i figured i could get a clue as to where i stand.

I failed on two counts, one was the aftermarket horn, no big deal.. THe other was the emissions.

I went for two rounds :giggle:

Test, Allowed Value, Round 1, Round 2
NOx - 1304, 750, 1806
HC - 166, 178, 185
CO - 0.93, 2.01, 1.83

Between the two rounds i did a little tuning which ended up leaning out the cells where the test was run. (Around 2000 rpm, 50kpa)

Do you guys know what would cause the excess HC and CO?? I figured too rich, so i made it leaner.. that helped with the CO a LITTLE but everything else got worse :vash:

-Tomaj

See, now that's a good graph.

You have too much CO and too much NOx. So you need to make the engine leaner and richer.


Seriously though, what's your ignition timing look like? Too much advance = high NOx, too little advance = high CO. You may have to finesse both mixture and timing to find the sweet spot. Put a little bit of fuel back in, and advance the spark a bit.

Well WTF then? If i read the chart correctly, either i can do the 1304 NOx, OR i can do the 0.93% CO...?

I'm assuming that this chart doesn't take the catalytic converter into consideration -- in which case my question is -- what's a cat more effective at? Converting NOx, or CO?

-Tomaj

Joe, thanks for the reply, but like you said, i'm too rich AND too lean simultaneously :)

Would regapping the spark plugs help at all? Maybe the gap's too small resulting in incomplete combustion? but wouldn't that lead to a very rich mixture which my wideband would let me see??

If anyone can, please take a look at my datalog.. The 'test' starts after the 2nd shutoff event, around record 4752 thru 6077.

-Tomaj

Wikipedia... yea...

Three way catalysts:

These three reactions occur most efficiently when the catalytic converter receives exhaust from an engine running slightly above the stoichiometric point. This is between 14.6 and 14.8 parts air to 1 part fuel, by weight, for gasoline. The ratio for LPG, natural gas and ethanol fuels is slightly different, requiring modified fuel system settings when using those fuels. Generally, engines fitted with 3-way catalytic converters are equipped with a computerised closed-loop feedback fuel injection system employing one or more oxygen sensors, though early in the deployment of 3-way converters, carburetors equipped for feedback mixture control were used. When a 3-way catalyst can be used in an open-loop system, NOx reduction efficiency is low. Within a narrow fuel/air ratio band surrounding stoichiometry, conversion of all three pollutants is very complete, sometimes approaching 100%. However, outside of that band, conversion efficiency falls off very rapidly. When there is more oxygen than required, then the system is said to be running lean, and the system is in oxidizing condition. In that case, the converter's two oxidizing reactions (oxidation of CO and hydrocarbons) are favoured, at the expense of the reducing reaction. When there is excessive fuel, then the engine is running rich. The reduction of NOx is favoured, at the expense of CO and HC oxidation.

No, like I said- so far you're only tuning one variable, and you seem to have found the threshold point. It's time to move the fuel slightly back in the direction it was (put it where you found the low NOx point, for starters) and then try advancing the timing a few degrees to see if that does anything positive for CO.

Joe, thanks. I must've misunderstood, I'll give that a shot and see what happens.

I wonder how many times i can go back before they get tired of seeing me :)

I don't boost full stop. :crx: But I see where you're coming from with the VEA, hadn't thought of it like that.

This is kinda old news by now, but i wanted to post in case anyone comes by this thread searching about emissions inspection.

I was able to pass with pretty good margin by dialing in the tune with MLV, and in the process discovering what i think is a bug (maybe). My AFR targets in the MS were always coming up 0.883 richer than what was in the table.. somewhere something was screwy. My solution -- just add 0.883 while NOT tuning in MLV, then make the changes while tuning, then go back, etc. So in the end, enabling EGO correction did the trick and i sailed through.

Do you have MLV set correctly for your wideband?? I had the same problem until I realized that I had it set wrong...

Yeah that's all fine. I followed Joe's instructons to get AFR instead of volts in megatune, and then proceeded to enter all my values in megatune. When i open MLV i see the same values so that's A-OK.

The issue comes into play when you go ahead and try to use gEGO. It was always trying to correct way too rich. I noticed there's a target AFR gauge, so i loaded that and noticed that it was always off by .883. It seems that some formula somewhere isn't carried through to everything, hence gEGO's target is off.

I hope that explains a little more.
-Toma