MAF Calibration - Diff AFR at same MAF HZ at different speeds/loads

[LS6FD]

This has me beating my head against the wall. I have dialed in my VE table before starting calibrating my MAF BTW too.

I am now trying to dial in my MAF table and have run into a really frustrating issue. I have different AFR error at the same MAF hz reading at different speeds (different load I would guess too).

Quick synopsis--at low speed when gently accelerating (1st, 2nd, and 3rd gear) staying in the same range of cells (approx 4250-4750 hz) and trying to keep throttle pretty constant they are rich (~11.5 - 12.8 AFR). When cruising down the road in 4th, 5th, and 6th gear, I'm a tad bit on the lean side in those same cells (15-15.2 AFR). It actually seems to get a little leaner each gear higher. Needless to say this really throws off my average WB AFR % error.

This brings me back to an issue I posted in another thread, entitled "OLMAF idle tuning". I found out why I was running WAY rich (10-11 AFR) while warming up idling--it was going into higher HZ MAF cells when warming up (around the 3750hz cell upon first cranking going down to 3125 when warm. I assume this is due to settings in the RAF table). Naturally I leaned those cells out and it now runs as it should warming up. The problem is though, if it's on for idle and warming up, when I hit those cells while driving, it's WAY lean (like 17-19 AFR). So I can either barely run when warming up, or barely run when driving around in those cells. WTF!?

I'm not really sure where to go from here. I know other people say they run OLMAF with no problems, so I would think it should be able to be done ok, but I don't see how!? The only thing I can think of is that maybe it has to do with my MAF placement. I swapped a LS6 into a FD Rx7 and built my own CAI that places the K&N filter behind the driver side brake duct. The MAF is at the very end of the CAI tube and the air filter is attached directly onto the end of the MAF sensor. I figured I might have some turbulance with it mounted there, so I did leave the screen in the MAF sensor. Maybe my MAF is bad?

I have attached my most recent tune and log file. Any help would be MUCH appreciated!!!!!

[WS6FirebirdTA00]

cold start fueling can be changed with olfa

[LS6FD]

The cold start isn't even my main concern right now--it's just general driveability at other frequencies.

The idle stuff I mentioned was kind of an aside and I rarely hit those cells while driving anyway. The 4000 hz range cells I do hit very commonly though and are creating horrid driveablity issues.

Basically I can either tune it so that when crusing at around 60 mph or so I'm in tune but will be reaaaaaly rich at low speed (0-30 or so), or be in tune at low speed but lean as hell crusing. It's hitting the same cells in both scenarios so it's impossible to tune both.

In really simple terms lets say a -15% AFR error for 4500 hz at 30mph, and a +5% AFR error at like 60 mph for 4500 hz. Should this ever happen!?

I can switch to OLSD and the car runs perfectly.

I'm assuming this is a very abnormal problem with MAF calibration?

As I mentioned earliar the air filter is on the end of the screened MAF. The bottom 1.5 inches or so of the air filter is directly exposed to incoming air through what used to be the oil cooler duct and receives direct air coming in, while the rest is concealed behind the front bumper of the car. This is my only guess as to the issue other than my MAF is bad!?

[WS6FirebirdTA00]

what is done to the car and what rpm/load is this happening at?

[LS6FD]

The motor is a basically stock LS6 (swapped into a 3rd gen Rx7). It has headers, a cold air intake, and ported throttle body.

I don't remember the exact RPM it was happening at, but from what I remember, it hits 4500hz at around 1000 RPM lower at higher speed/gears (around 2500 or so RPM) vs low speed (3500 RPM). Other than the load required to move the car, there is no load--I was trying to keep it at perfectly steady throttle position.

I have been logging my commanded AFR and it definitely was not entering PE.

[WS6FirebirdTA00]

When I say load I mean what g/cyl are you at.

Do you have any pics of the MAF piping?

[LS6FD]

Hey WS6, I can snap pics of the MAF piping when I get home.

I'll probably need to do another log this eve and set markers (tried to this last run but must have missed the "m" key) so that I can cross ref the VE table to tell you the g/cyl I'm at.

[redhardsupra]

your MAF curve is completely hacked up, it supposed to be a smooth 3rd order polynomial.

if for the same MAFfreq's you get different fueling, then it's gotta be some other modifiers muddying it up.

i'm assuming you get AFR through EGR voltage. what formula do you have for that?

[LS6FD]

I started with a bone stock Z06 tune and the VE and MAF table are the only ones I've messed with.

I realize the MAF curve is supposed to be smooth, but due to the weird issue I've been having that's how it ended up like so lol!

you are correct-I'm using EGR voltage to log, and my formula is: ([PID.2811]/.5)+10 (I have the PLX wideband)

I was able to dial my VE table in using the exact formula.

[5_Liter_Eater]

I donno that it's "hacked up". Mine looks similar. I don't think the condition you're seeing is that abnormal. If you look at the min and max AFR error across the MAF curve for any log I'm sure you'll find a big variation. Just so long as the averages pan out thats what you're looking for.

With that being said, with the advent of 2.2 we now have a PID called "Air Calc Mode". I would suggest you start logging that PID so that you can verify that you're referencing the MAF when you see these variations. Under some circumstances you may actually be referencing the VE table. If the PID says "High Speed" it's referencing the MAF. Low Speed = VE.

[Bluecat]

I agree that some other table has to be coming into effect. How does your open loop EQ ratio table look?

I see problems like your describing all the time, but only with big cam cars that dramatically change they way they breath over the rpm range. Say at a 900rpm/70kpa idle the MAF spits out 3000hz and to get the commanded air/fuel right you end up with 11 g/sec in that cell. Then again at a lighter load, but higher rpm you hit the same hz, say at 1600rpm/45kpa with just enough throttle to hold a cruising speed. Its lean and takes a value of 15 to make the Commanded AFR match the actually AFR. Obvioulys now at idle its rich. The large amount of overlap and duration just makes the car require different fueling at diffent times, even though the same amount of air being injested. Part of it, you just have to go SD and shit can the MAF unless you want to fudge numbers in other tables to make up the difference.

I can't image that being your problem. Its either a missed tuning problem or a phisyical issue like the way the air rolls through the MAF or just a bad MAF to start with.

[LS6FD]

Thanks guys. I was going to log the new "Air Calc Mode" PID but couldn't add it to my table!?

For my OL EQ table, everything 140 ECT and over is set to 1.

Being a basically stock LS6, I really don't think I should have any issues getting this thing tuned right hopefully.

I've been talking to a friend that does a lot with Z's and he said the way I have my filter/MAF situated will definitely cause problems on the Z. I think I'm going to try blocking off the bottom of my filter from the direct airflow it's receiving (the bottom inch or so of my filter is exposed to direct incoming air) and see if that helps the problem.

[LS6FD]

Well SONOFA!@#! I figured out the problem. My hypothesis was correct--the placement of the filter receiving direct airflow on the bottom and not top and having it directly attached to the MAF was totally throwing my reading off, even having the screen in the MAF.

I attached a before picture of my setup below. Basically, I just took some aluminum tape and completely sealed off the duct you see the airfilter sitting behind just now for testing. I'm now getting 11.5 - 12 AFR where I was getting 15-15.2 AFR before at high speed and it corresponds to the same reading at low speed now, that's how much of a difference it made! I was actually rich across the board after doing this.

Sucks that I spent so much time trying to get it calibrated, but at least got it figured out, and hopefully others can benefit from my mistake. Now I've just got to figure out an elegant way to permanately block off that opening.

Thanks for all your help everyone, it's been much appreciated!!!

[WS6FirebirdTA00]

I was going to say, that location of that does not look like it would help lol

[EC_Tune]

MAF's don't like to be at the end of a pipe in any application. Check all your GM trucks as well....

They tend to be most accurate in a nice straight piece of pipe and right in the middle of that straight piece.

I can't count how many "hot rod" intakes have the MAF attached within an inch or two of the throttle body and consequently have poor running issues. Hmmm wonder why...

Glad you found your issue though!

[LS6FD]

The reason I put the MAF where I did is because the integrated IAT sensor on it was getting BAD heat soak anywhere else I could mount it (basically within the engine bay).

I should be able to dial this in pretty easily though now--it stayed perfectly consistant last night across the board thankfully!

[EC_Tune]

Sounds like you got it under control. Happy tuning!