MAF Frq vs. G/sec at WOT

[PRAY]

I have been breaking in my new motor and trying to get idle, low rpm surging and WOT under control. What I am noticing now at WOT on the street is that my MAF G/S don't match my Frq/Hz at WOT. An example being my MAF F/H says 9,400 but my G/S in the scale says 485 but the log says only 422. The other extreme example of this is at 8958hz it is reading 489gs in the lof when the scale reads 351. Can someone explain what is going on. I don't have my WB hooked up right now to get the actual AFR for both readings. Maybe posting a screen shot could help explain what I am talking about. At normal driving and idle it matches.

[swingtan]

Which gm/Sec PID are you looking at?

Can you post the current tune?

[PRAY]

Maffreq2. It is an E67 in an 07' CTS-V. I will try to post the tune in the morning. I am running an LS7 MAF card as well. It is giving me fits. Car runs great in CLSD. Once I turn the MAF back on here comes the bucking. I have tried everything in all of the stickies. I had the injector timing advanced to match the cam and things were not that bad. Then I put it back to stock and the fuel leveled out with the maf on it just about rattles your teeth out. So, in the morning I am going back to SD with the injector timing stock to see what happens.

[swingtan]

I can think of a couple of things....

[list=1]
Charge temp blending may be altering the calculated value. Maybe.... maybe not.
Given things seem right in SD mode, you may be getting reverberation in the intake pipe. Again, a long shot....
[list]

You could lave the MAF in place and just set it to fail in the tune, then see if the values match up. You can still record MAFFREQ2 when in SD mode, if you set the MAF to fail at a very low frequency.

Simon.

[joecar]

Post log file.

[PRAY]

18066 18067 Here is the current MAF tune and the log. The WOT pull is around the 16 minute mark.

[PRAY]

I only fail the MAF and leave it in place when I go SD. I also turn off DFCO and CFCO to ensure I get the VVE table correct. So I ran it that way this morning and all my LTFT's were within +/- 3. Most under that. Car drove almost like stock for a car with 21* of overlap. It was wonderful. I am going to keep it in SD and turn DFCO/CFCO back on to see if that effects anything. It is doesn't, then I will turn the MAF back on and see what happens. Those are the only two things I change on the tune going from SD to MAF.

Just so the tune makes sense. I dropped in a LS3 427 and ported the heads, intake, headers with stock cat back and did a custom CAI with the LS7 MAF. The cam is a LSL/HUC 243/251 113+4 21* of overlap. MN12 car with 3.73's.

[PRAY]

18068 So this is a WOT in SD. If I look at it over all the car went beast mode. The MAFFRQ and G/S still don't match up but the AppCylDMA stayed high along with the MAFFRQ and the G/S maxes out at 511 instead of like the last on with the MAF where all that data noses over at 6,400 or so. The pull is at 8:26:25 in the log. It actually spun for a sec in 3rd on the launch and traction control kicked in which it has never done. It was colder this morning but I will have to go back and look at by how much.

Adding back in the DFCO/CFCO it didn't seem to work in SD mode. Does it only work with the MAF turned on? Car still drove great but was a little leaner. Still idling at 0 to -1 but was leaner at cruise by about 2-4. I hate how you have to manipulate this VVE table now. Minor corrections are about impossible if you have rich and lean conditions that span only a couple hundred rpm. The car seems to hate 5th gear for some reason. I think I have it narrowed down to my coast down spark table. Once timing drops out at light throttle or backing off the throttle to around 27* and below the buck fest happens. I think if I can get that worked out I will be good. All the other gears are money. Maybe I can mess with the Idle Air in 5th. I just know that as soon as I hook the MAF back up I am going to want to blow the car up. This MAF set up did the same thing in my cammed TBSS. When I went back to the stock 85mm MAF it drove like a dream. That car was an Auto with a much smaller cam but was comparative to this new set up as per overlap to cubes.

[statesman]

I don't think your MAF is working properly. Your Freq/Hz numbers are all over the place.

[PRAY]

I didn't either but I swapped MAF sensors and all of the wiring around thinking I messed something up with the swap. Everything is correct. The only thing I can think is that I need a different air filter with a velocity stack or perhaps I can find some honey comb to put in there to act as a flow straightener. It definitely does not act right. I can't find it in the tune either.

[statesman]

Ah, that just brought back memories for me. I read something a while back about someone installing a 'performance' air filter and having an erratic MAF as a result... going back to the original filter resolved his problem.

[PRAY]

Yeah, I have noticed with all the stock stuff and aftermarket stuff that use this MAF the air intake has a smooth transition into the filter. They say you want about 6" on both sides of the MAF on straight tube. None of the after market stuff has that nor does most of the stock stuff. My problem I believe is that I only have about 3"-4" in front of the MAF to the filter and no smoothing transition. Just pipe then filter. I just don't want to back it up and get it to close to the TB and expose it to reverb of the intake. Although since it is basically a closed system what you see at the manifold you see at the filter. I will have to play with it and see what I can do. On the TBSS I tried all sorts of combos with different filters bit nothing worked.

[statesman]

Have you got enough room to put a stock air filter back in?

[PRAY]

Not any more. The stock filter is a square type that sits in a box. The box won't fit with the way the piping is set up.

[statesman]

I thought that might be the case. You could always try to screen the MAF, it's cheaper than the honeycombs on the market and it should help to reduce the turbulence.

[PRAY]

I have to figure out something. I literally want to pull over and leave the car on the side of the road with the MAF hooked up. What screening would you suggest?

[statesman]

If you're using a stock MAF tube, then try to get a hold of a factory MAF screen. If you've got a custom tube then make one out of fibreglass flyscreen mesh. Just make sure whatever you put in there can't get dislodged.

[PRAY]

It is a 4" tube so something will have to be made. I will epoxy what ever I do in there. I haven't seen a screen on anything with a stock LS7 style MAF card.

[statesman]

I don't know much about the LS7 style MAF card... maybe with a stock set up the air filter does enough buffering and they don't need screening, but with your set up you need something to reduce the turbulence. Just looking at your log I can see how crazy the MAF readings are... so I can imagine what it's like to drive.

[PRAY]

Yes, cruise is the worst, especially light throttle. Idle is pretty good, acceleration is pretty good, and I can't really tell what is going on at WOT. Light throttle and deceleration are horrible. I never considered the stock filter creating a buffer but it makes perfect sense. I guess I need some sort of velocity stack or flow straightener to smooth things out.

[joecar]

Try here: http://www.saxonpc.com

More specifically: http://www.saxonpc.com/100mm-cells-for-100.html

[swingtan]

I haven't quite worked out the MAF gm/S issue yet, but WRT the light throttle issue, check this...

I've expanded the data to show the RPM bucking and relative PID data. I'd almost say that the TB blade is following the pedal movement a bit too quickly, allowing for larger than needed changes in airflow as seen by the MAF. Your tune also seems to be running some rather strange dynamics settings, are those stock? Can we get a log showing wideband data? If not can we get O2 voltage and IBPW?

https://forum.efilive.com/attachment.php?attachmentid=18071&d=1425260489

[PRAY]

I noticed the TB blade trimming while bucking also but wasn't sure if it was slight fluctuation of my foot. I tried keeping it as still as possible. Some times the blade trims and others it doesn't. I believe on one of those logs I was SD. I was thinking that the blade was trying to compensate for the MAF weirdness trying to stay in a steady state of fueling.

The Dynamics are stock. I am still trying to figure those out. I have read both of your threads pertaining to them but can't exactly wrap my head around what exact tables need to be changed. It seemed that in the end of the second thread Hymey and your self were disagreeing on some concepts of the fueling so I left them alone untill I could get clarification. I know they don't work in SD so I haven't really been messing with them. I have been trying to get the VVE table perfect so I can rule that out. I think I have it nailed down now. All LTFT's are -1 to -3 and WOT fueling should be around the 12.8 mark. I didn't get to WB log after the final corrections. I checked four or five other cars tunes that I have and all the dynamics are the same. I was going to apply Z06 data since the motors are the same size and I am running that MAF but there was nothing to change.

What do I need to do to correct the blade trimming issue? I have reduced a couple of the blade tables as per one of your threads and it helped with starting and bucking. Maybe I went too far.

I have some WOT runs with WB data from the other night that were in SD. I will say that I have never seen such uneven fueling. It bounces from 12.3 to 12.6 in a pretty smooth rhythm. Some of the runs it matches the MAF discrepancies even though the MAF is off. Not sure if it could some how be coming on at WOT. During normal driving it is failed for sure. I still feel that it is turbulance with the MAF. I can feel it and see it on the logs. I have ordered a new elbo that will allow me to pull the MAF back a couple inches from the filter and rotate it to horizontal along with rolling the filter to match. Maybe that will help. I am going to order some honeycomb also. I will post a pic of the intake set up as it sits right now. This morning I turned the MAF back on and on three different driving sessions I got three different sets of trims. I didn't change anything in the tune.

I will say the thing runs harder than I though it did. I was guessing it would dyno just over 500-520rw with the stock exhaust and intake but I did five tuning pulls lined up with an 07' Z06 I did that put down 515/505rw the other day and we were dead even with me pulling about half a car. I would jump out about a car and hold him there. 5 runs ensures it wasn't a fluke. My car weighs 4,100 with me it it. Most of them were one gear pulls to eleminate driver error. I will post the log as soon as I get home. I will set it up to reflect MAF g/s over WB AFR. The commanded fuel in PE is 12.8 across the board. Thanks for all the help.

[PRAY]

18074 This is the SD runs with the Vette the other night. Here are 4 of the runs. They start around 18:46.

[PRAY]

Try here: http://www.saxonpc.com

More specifically: http://www.saxonpc.com/100mm-cells-for-100.html

Thanks Joe. I never would have found that. I just placed my order. How far in front of the MAF Card would you place this? Not sure if right in front or at the filter would be best. The pipe it is in only has about 3" on each side. I guess I will just place it at the lead edge of that and see what happens. Do you guys think this will cost me any power? I am guessing 3-5rw at most. If the car drives like it is supposed to it would be worth 10rw. I can make that up elsewhere.

[PRAY]

Here is another strange thing. I can drive for days in CLSD with no codes. Turn the MAF back on and with in 30 miles I get insufficient switching codes from my 02's. I drive it yesterday an monitored TP% during bucking with he MAF in and couldn't see anything obvious. The % stayed the same.

[joecar]

You can place it close to the MAF (the stock MAF screens are typically close to the MAF element)... you may have to experiment (leading edge of pipe, trailing edge of pipe).

this screen is not a restriction since the hexagonal cells are much larger than the cells on the stock MAF screens... the 4:1 ratio refers to the cell size (4:1 is large, 7:1 is small, IIRC).


I'm hoping it works, otherwise you have a $13 coffee table ornament.

[PRAY]

I was wondering what the ratios were. I thought it was the depth of the cells not the actual size. They way they explained it on the site was 5:1 was the correct depth for the stock 78mm MAF. I am going to put it an inch or so in front of the element.

[joecar]

It says on that page, just in front of the ratio it says "cell size".

[joecar]

I have used a few screens from them, and the cell size is much bigger than stock, the screen is not going to be a restriction to airflow.

[PRAY]

I put the screen in, re did the 90* boot, and rolled my injector timing forward. The MAF is now pretty darn steady and I went from an average of 13 mpg to around 18 mpg so far. I am getting right around 23 mpg on the highway from 19 and the fuel smell has greatly diminished. It also starts and idles much better. I have much less bucking and what I do have is not from the MAF any more. I believe I may have hurt the motor. I am going to do a compression and leak down test tomorrow to see what I get. More on that when I have results. I thought about what Simon said about the throttle blade trimming so I returned all the throttle tables back to stock. I still have my PI tables zeroed up +/- 96rpm. I am going to mess with that next. It is nice having DFCO/CFCO again. That cured a bunch of the on/off throttle bucking. I still need to work on my Dynamics but don't know which direction to start. Simon, are you out there. Thanks for all the help so far. I will keep you all updated.

[swingtan]

Dynamics..... there's an interesting topic to get into.

The stock tables seem OK in some tunes but off in others. So I think there's a bit of hit and miss depending on particular engines and modifications. So here's what I think ATM. Working down the E38 tables....


B2021: Wall Wetting Delay - I reduce this to try and speed up the ECM's calculations. Currently I have it set on 30 while I'm playing around.
B2003: Stomp delta min - I have this on 3mg
B2002: Stomp delta min add - I have this on 0.5 to allow for a ramp out of the additional fuel instead of my stock settings that just stopped.
B2004: Stomp mass decay - lower in the warmer regions to slow down the fueling transients.
B2005: Stomp compensation - Higher in the warmer regions to add more fuel.
B2012: Impact Factor - Set lower at low MAP (hot and cold) and higher at high MAP when hot.
B2014: Evaporation factor - this will probably make the most change. I have it significantly higher at low MAP when hot, and lower at high MAP when hot.
B2015: Evaporation factor RPM - along with the previous, it has a big impact. I have if lower at low RPM and high at high RPM.
B2011: Wall Wetting Asymmetry - High at low temps and low at high temps


Tuning the dynamics is a bit of an art. I probably should write up something on that....

[PRAY]

Yes Sir, I think you should do a write up on it. Thanks for the breakdown. The EFI Live explanation's of the tables didn't give me a warm and fuzzy on what was going on and why. I will assume that I will have to bounce all of this off of my WB and the goal is to achieve steady consistent WB data at all transients and throttle positions in correlation to the commanded AFR? I do this after my WB data matches the above statement but with just the VVE table (since Dynamics are disabled when the MAF is disabled). Then enable the MAF and get it as close as possible then go after dynamics vs. the WB for the areas and transitions the MAF can't handle due to speed or it's one dimensional nature. I will also assume that my PE tables come into this and need to stay as one dimensional or consistent as possible. I.E. only commanding 12.8 afr at WOT and ramp in and out at .1. Am I tracking?

[swingtan]

The dynamics shouldn't be disabled in SD mode, they just need a lot more work....

In the E38 (and probably most other controllers), air flow is used heavily for dynamics calculations. With the MAF in place, you get an airflow spike when snapping open the throttle, as the plenum fills with air. The MAF reads this air flow and you can see it in the logged data, but you don;t see the same spike in SD mode. So you need to redo all the dynamic airflow tables to take into account there is no spike. See this example showing MAF airflow vs AIRpersec (axis shifted to allow the two traces to be seen).

https://forum.efilive.com/attachment.php?attachmentid=18119