What is the key to getting WB AFR to match the map AFR?
 

Every time I do a pull, my WB is leaner than the actual fuel table itself. I tried LS on and off with simular results. LS on is about .75 leaner than without it.

At cruise it is perfect 14.7 just like the fuel table.

My injectors are scaled to only have -1.3 low and -2.0 mid. They are closer to 0 on colder days.

I am running pump 93, altitude of 550 ft, revolver cams, AMS 780cc injectors, 255 pump, fuel rail, FP Green and all the other supporting mods.

Please let me know what I need to do to get my WB to match the fuel table on a WOT pull.

MAF scaling table.

Log MAF Hz, AFRMAP and WB AFR.

Do an Excel scatter x-y plot of MAF Hz vs AFRMAP/WBAFR.

Then multiply the MAF scaling table for each MAF Hz by the AFRMAP/WBAFR. Smooth it out and pick a representative blob on your scatter plot.

Note that the MAF scaling table has an offset added to it, as previously discussed in Eric's threads. Off the top of my head I can't remember what it is, but you'll need the table scaled correctly to make your adjustments also correct.

On the other hand, does it really matter?

I guess it really does not matter all that much. I am just a bit OCD and want everything to match.

I also believe that it will allow for quicker tuning. I will be tracking my car quite a bit soon. I want to be able to make quick, easy adjustments to compensatew for weather, track conditions (boost), and barametric pressure.

offset is 140

I know how to scale the MAF with the 140 adder. The thing is though, with the green and revolvers, I am out of the scale pretty quickly. At least from what my logs can show. I guess i need to use the airflow patch to log it.

I will keep you all posted.

Thanks for your help.

Yes log 2 byte, and change the 1600 Hz in the MAF scaling table to something higher.

I used the "MAF Smoothing" table as it is a multipler that does not involve an adder. Displayed in Percent128.

A 10% change in that table makes a 10% change in AFR.

At WOT above ~800Hz, my actual AFR matched the mapped AFR very well (within 2%) when lean spool was disabled using the factor values. If this isn't the case, I would first change your injector scaling to get a good match at WOT.

Will logging 2byte airflow allow a reading higher then 1600Hz?

EDIT:
You need to be in open loop to do this. I think you'll find your cruise stuff is very rich. The only reason it's at 14.7:1 is because of closed loop feedback. You can disable closed loop through the periphery bits.

The 1600Hz in my MAF Scaling Map?

I already know that 2 byte AirFlow will read higher than 1600Hz.

I have no idea how to get my MAF Scaling to read correctly if i were to change the 1600Hz to "something higher".

Please advise.

How would I go about changing the MAF Smoothing table? The numbers there are are not a smooth curve. Increase to add fuel or decrease to add fuel?

Which periphery bit to disable close loop?

I was always under the assumption that the AFR tables were just a "representation" of the injector pulses in more of a user friendly interface than in actual milliseconds. 99% of all other tuning software I've used just listed ms pulse time instead of AFR's, this has been the first one I've used that is set up this way, so I never really look at it as an actual AFR value.

Up until now I have been just like you and used the table as more of a reference than the actual value. I just believe that if we are going to have AFR tables, why not make them accurate to the actual AFR of the car?

Mine are not off by much. The table is usually a bit richer than my WOT pull actual AFR. I am talking maybe 1 point or so. (yes lean spool was off).

Some standalones use a target AFR table instead of an injector pulsewidth. Autronic and I believe Motec both use this method. They also use a VE table to go with it, allowing you to separately tune AFR from engine efficiency. Very nice setup actually.

The table is a multiplier either way, so it's pretty realistic to display it as AFR instead of a pulsewidth. As a multiplier, it would be like looking at lambda values.

If you change the scaling to percent128 on the table it makes more sense while you are changing values. You are right, it's not a smooth curve. Also, if you look at some of the ralliart ECUs, they use the same MAF Scaling tables but different MAF Smoothing tables, which has lead me to believe the MAF Smoothing table is actually a MAF Scaling trim table were the MAF Scaling sets the characteristic scaling that is expected from the MAF and the MAF smoothing is a table that is actually tuned at the factory for different MAF/intake systems.

Which scaling do I changet percent 128? Load or g/s?

Never mind....I think I got it. I changed the scaling for the Load to AirFlowHz (gives me the same values as Maf Scaling) and then I changed the g/s scaling to percent 128. The graph is exactly the same but the values on the smoothing table reflect a percentage.

So realistically shouldn't this be 100 across the board to get an actual reflection of the fuel table? I ask becasue my AFRs are almost exactly the percentage in the smoothing table off from the fuel table....if that makes sense.

<table name="MAF Smoothing" address="XXXX" ... scaling="Percent128">
<table name="MAF Frequency" address="YYYY" scaling="Airflow"/>
</table>

Or something roughly like that. The table should be based on MAF frequencies ranging from 19Hz to 1600Hz when properly scaled.

got it...please see above.