Raw MAF frequency on EE def

[Skinny Pedal]

I am wanting to log raw MAF frequency with the EE definition. I just don't know where the package offset would be to do so. Anyone done this? The way I figure I can build a history table to calibrate the MAF if I can data log the frequency. I build VE multiplier tables for speed density and they work great. I would love to be able and calibrate the MAF though.

[Skinny Pedal]

Oops, I should have put this in Tunerpro section. Feel free to move.

[EagleMark]

MAF Frequency is in XDF/bin along side airflow grams per second. But data ADX only has airflow grams per second. So you'd have to build the history table in airflow to match up with MAF table in XDF/bin.

[Skinny Pedal]

Thanks I see it now, I must have over looked it. Matching up shouldn't be a problem. Next time I tune one I'll let you know how it goes.

[Skinny Pedal]

So I am guessing the MAF frequency is coming in then converted to 256 bit, right? If that's right I should be able to figure out the formula to convert back to hertz. I would just keep verifying the formula with the calculated air flow. The 4 MAF tables would show me the frequency range the PCM can take. Once the formula is right I could then tune the MAF much more quickly. Or am I way off?

[EagleMark]

I think your making this to hard? I'm not sure how things are handled at PCM level.. but the frequency is hard coded, the airflow is what is in calibration to change/adjust.

You already have the airflow in data stream, so make a history table for the airflow and use it to adjust airflow in bin/MAF tables. Unless you have a monster engine you'll only use 2 1/2 of the 4 tables. Remember these are interpolated just like VE tables, so changing one cell does not cut it, it's like a blend of 3 cells.

Someone had built a XDF with all four tables combined for smother transitions. This is one table that has to be smooth, completely smooth. I them built the MAF history table in ADX and they worked well. But I can't find them... must have been some of the files I lost last year in a computer crash...

The BLM tables are already built in the ADX, if you build the MAF post it up for the $EE info thread. The BLMs also work for doing the VE and then MAF without the airflow step... but your really doing it correctly.

Make sure you do the VE tables first in SD/Speed Density. There's lots of info that says VE is not used with MAF... but their wrong and I have proved it in tuning them. We've never had complete disassembley of the LT1 to prove/disprove in code or know where it goes to MAF only? But later models we do and after say 4000 RPM it is MAF only. Under 4000 RPM the MAF looks at VE as a sanity check.. short explanation. It's easy to prove in tuning, just screw up the VE table and watch fuel trims in that area, you'll have to have a huge jump in MAF tables to compensate.

What I do is... after VE is dialed in you could adjust the MAF airflow with BLM data by using the MAF history table you built for airflow. Just change the update with for a new value. Build a new value in ADX, It's a Base Efficiency Numerator and will give you the decimal to adjust from. Goes like this Right BLM + Left BLM / 2 / 128
So 120 + 118 = 238 / 2 / 128 = 0.9296875 and there's your number in each cell of history table to adjust the MAF airflow by.

For PE/WOT and no wide band use the BLM locker program in $EE info thread. This locks INT and BLM to 128 when in PE. Data log WOT and watch the O2 sensor reading, get it dialed into .850-.900 using the PE AFR table.

[Skinny Pedal]

I have reasons for wanting the frequency and not just the airflow. Basically what I have been doing is making my own auto VE and MAF tuning multiplication tables similar to what you can do with EFI Live. VE is a piece of cake but without raw frequency you can't build a history table that doesn't have to have the x and y updated every time you make a correction. I am using a wideband for both not a narrow band. I know I could use the airflow but would prefer to dial in the airflow not based on calculation of airflow. I tuned a 383 lt1 with an lt4 hot cam the other week and it ran perfect in speed density. Every time I would enable the MAF it would go very very lean. We ran out of time on the dyno to fix the MAF so I left it SD for the time being. I am just wanting to speed up my tuning and offer faster more accurate correction.

[EagleMark]

I was editing when you posted... read above for the MAF multiplication tables off BLM, same could be done with WB Lambda readings.

There's just no frequency in data and we don't have hacks to put it there. Maybe your way to calibrate MAF frequency from airflow would work?

[Skinny Pedal]

Thanks for the help. I have another EE pcm to do in the next week or two. I'll give it a shot and see if I can get the formula right to figure frequency. If not I'll just have to take the long road. Thanks again.

[RobertISaar]

you can estimate VE without MAF Hz.... i've actually done it inside of tunerpro before. the math is kind of crazy, but it seems believable.

also, putting MAF Hz in place of MAF G/S, for example, can be done. i've got enough of the 94-95 LT1 BIN decoded to do that.

MAF Hz is on the E-side, 16-bit, word 185.

the problem here is that you normally don't communicate with the E-side... it's all done through the T-side. E and T pass some information to each other through the SPI link on a regular basis, but MAF Hz isn't one of them. MAF G/S gets passed, but the T-side uses it for a couple of functions. so at this point, there are a few options:

mode 3 request to the E-side to grab ONLY MAF Hz immediately after grabbing a T-side frame(really simple to setup, little impact on logging)

grab mode 1 packets from the E-side, edit in MAF Hz if necessary(obviously, very little to no t-side parameters will be available)

graft MAF Hz into the SPI link and read it over T-side mode 1(would need to find a 16 bit value to replace that the T-side doesn't also use for some function)



those are the options how i see them.

[Skinny Pedal]

WOW!!! Thanks. I'll see what I can come up with and will share my findings.

[EagleMark]

Your right, I was playing with that today and remember I could not get data tracing to work either on MAF because no X and Y axis.

Probably try and get the MAF frequency in place of Prom ID data stream? Then make new ADX value.

We do have some dissasembly of LT1 $EE:
http://www.gearhead-efi.com/Fuel-Inj...ull=1#post2423

Robert is only one I know that can pull off the Mode commands to do so. When you say E said T side is that Engine Side Trans side? Cause the Trans data is in the ADX

[RobertISaar]

E and T are more or less engine and trans side. officially, they stand for "event" and "timing", but after going through the code, i really don't see what would limit either to mostly timing or mostly events... both sides are pretty competent in what they can do.

T side has a device ID of F4
E side has a device ID of E4