Tuning MAF and VVE at the Same Time

[Cringer]

OK, it may have been mentioned in here already, but is there a way to work around the airflow limits of the scanner? I wasn't paying attention on one and made a 30% correction only for the next log to want the same thing over again at which time I realized I was hitting the limit and switched over to the airmass equation to finish out my VE table.

I did come up with these, haven't tested them...because I am N/A scum you know lol. But this is based on the fact that Cyl Airmass is determined by Dynamic Air, so we should be able to work backworks. From my initial testing, it is close, but not great. Most likely this is not a viable solution unless someone else can accurize this method, or perhaps my math has some issues that need to be ironed out.

https://forum.hptuners.com/showthrea...l=1#post743701

[jermeyg]

I have had my LTFT's disabled for a long time now. But I use the formula you posted above...however, the only way that works is that you have to have the LTFT channel added to your channel list. It will report 0% but that is necessary for the formula to work since it is look at both ST and LT.

The other option is, keep LTFT disabled and remove it from the channel list and use these formulas instead which only look at STFT:
(([2320.71]+([2320.71]*([50116.156])/100))-[50040.71])/[50040.71]*100
(([2320.71]+([2320.71]*([50116.156]+0)/100))-[2311.71])/[2311.71]*100



That filter looks like the one Sirius created, right? If so, I have not used it myself. However, I do not feel like it is needed, so I stick my mine:

CL
[2517.161.avg(1500)]=0 and [2517.161.avg(-200)]=0 and (abs([50090.156.slope(1500)])+abs([50090.156.slope(-500)]))<2 and ([6310]=9 OR [6310]=14 OR [6310]=15)=0

PE
[2517.161.avg(1500)]=0 and [2517.161.avg(-200)]=0 and (abs([50090.156.slope(250)])+abs([50090.156.slope(-250)]))<2 and [6310.avg(250)]=9

May I ask that you explain these filters and how to work please?

[Cringer]

May I ask that you explain these filters and how to work please?

They filter out noise from the data — noise from throttle movement, torque management, and they also use the fuel trim cell to infer open/closed loop and if the EVAP vent is open (engines will run richer when the EVAP is open).

[FetzerValve]

I have had my LTFT's disabled for a long time now. But I use the formula you posted above...however, the only way that works is that you have to have the LTFT channel added to your channel list. It will report 0% but that is necessary for the formula to work since it is look at both ST and LT.

The other option is, keep LTFT disabled and remove it from the channel list and use these formulas instead which only look at STFT:
(([2320.71]+([2320.71]*([50116.156])/100))-[50040.71])/[50040.71]*100
(([2320.71]+([2320.71]*([50116.156]+0)/100))-[2311.71])/[2311.71]*100



That filter looks like the one Sirius created, right? If so, I have not used it myself. However, I do not feel like it is needed, so I stick my mine:

CL
[2517.161.avg(1500)]=0 and [2517.161.avg(-200)]=0 and (abs([50090.156.slope(1500)])+abs([50090.156.slope(-500)]))<2 and ([6310]=9 OR [6310]=14 OR [6310]=15)=0

PE
[2517.161.avg(1500)]=0 and [2517.161.avg(-200)]=0 and (abs([50090.156.slope(250)])+abs([50090.156.slope(-250)]))<2 and [6310.avg(250)]=9

Yes, I think the filter came from Siruius' contribution to this thread. Logging and tuning is going great with your method. My CL fueling seems to be better than it ever has been with much less effort on my end. Many thanks to you and everyone else who has contributed to this.

I'm happy with CL, but need to address PE and VVT tuning. Do I need to park the intake cam while logging data for VVT tuning? Also, my application is a turbo'd truck. Does the PE calc work with forced induction and manifold pressures of ~145kpa?

[Cringer]

Yes, I think the filter came from Siruius' contribution to this thread. Logging and tuning is going great with your method. My CL fueling seems to be better than it ever has been with much less effort on my end. Many thanks to you and everyone else who has contributed to this.

I'm happy with CL, but need to address PE and VVT tuning. Do I need to park the intake cam while logging data for VVT tuning? Also, my application is a turbo'd truck. Does the PE calc work with forced induction and manifold pressures of ~145kpa?

You don't necessarily need to park the cam. You would need to log the cam angle and use the cam angle as a filter to only show 0* cam angles. Then you need to tune the VVE for when the cam angle retards to 20* or whatever you are seeing. To do this, you will have to use the VVE Assistant I created and go into advanced mode. However, this is not really good now that I know what I am doing...meaning it is not really possible to manipulate the coefficients by hand and expect a real 3D shape. VVE Assistant v2 will fix this. Not sure if you have seen or used the VVE tool I wrote but advanced mode is covered in the first video in my sig (https://youtu.be/_Eiyy460C_k).

The Dynamic Air method will work with forced induction to a point. There is a defect or something in the HPT scanner that caps the reported MAF and VVE g/s values. The other mega thread covers this, but once you exceed these limits you will need to resort to other methods.

[FetzerValve]

You don't necessarily need to park the cam. You would need to log the cam angle and use the cam angle as a filter to only show 0* cam angles. Then you need to tune the VVE for when the cam angle retards to 20* or whatever you are seeing. To do this, you will have to use the VVE Assistant I created and go into advanced mode. However, this is not really good now that I know what I am doing...meaning it is not really possible to manipulate the coefficients by hand and expect a real 3D shape. VVE Assistant v2 will fix this. Not sure if you have seen or used the VVE tool I wrote but advanced mode is covered in the first video in my sig (https://youtu.be/_Eiyy460C_k).

The Dynamic Air method will work with forced induction to a point. There is a defect or something in the HPT scanner that caps the reported MAF and VVE g/s values. The other mega thread covers this, but once you exceed these limits you will need to resort to other methods.

Ahh, that makes sense to just log cam position and filter accordingly. Thanks! I've been using V1.5 so far and am very happy with the results. Assuming my "basic mode" VVE table is good, is there any risk of just leaving the VVT stuff alone until V2 is ready?

I've been through that particular mega-thread. After looking at my most recent log, it turns out I'm maxing out VVE airflow so I guess I'll re-read that thread now that I know it applies to me!

[GHuggins]

You don't necessarily need to park the cam. You would need to log the cam angle and use the cam angle as a filter to only show 0* cam angles. Then you need to tune the VVE for when the cam angle retards to 20* or whatever you are seeing. To do this, you will have to use the VVE Assistant I created and go into advanced mode. However, this is not really good now that I know what I am doing...meaning it is not really possible to manipulate the coefficients by hand and expect a real 3D shape. VVE Assistant v2 will fix this. Not sure if you have seen or used the VVE tool I wrote but advanced mode is covered in the first video in my sig (https://youtu.be/_Eiyy460C_k).

The Dynamic Air method will work with forced induction to a point. There is a defect or something in the HPT scanner that caps the reported MAF and VVE g/s values. The other mega thread covers this, but once you exceed these limits you will need to resort to other methods.

This is still WAY better than adjusting the coefficients by hand and then going to the VVE table, changing the cam position and seeing if it changed it roughly the right percentage and in the right area. That's how I was doing it

I prefer to lock the cam at 0 dial that in as a whole and then lock it at the max position and make corrections for that. Seems to work best for the torque model corrections and to get the VE table right in general. I've had issues with the scanner not reporting the cam positions correctly via the filters or what ever and cause problems trying to dial it in with the cam moving. May just be because of all the maths running at once bogging it down. Don't know.

[Cringer]

Ahh, that makes sense to just log cam position and filter accordingly. Thanks! I've been using V1.5 so far and am very happy with the results. Assuming my "basic mode" VVE table is good, is there any risk of just leaving the VVT stuff alone until V2 is ready?

I've been through that particular mega-thread. After looking at my most recent log, it turns out I'm maxing out VVE airflow so I guess I'll re-read that thread now that I know it applies to me!

....

[dassaur]

Been doing some great vve and maf in CL using cringers formulas so far so good, I'm definitely noticing improved drive ability and economy!
My car was dyno tuned so I don't think WOT will need much work, but one thing I am a bit hazy on:
These formulas only work if the data is collected at steady state, as far as I am aware. So my question is how does WOT come into the equation in relation to steady state?

[Cringer]

Been doing some great vve and maf in CL using cringers formulas so far so good, I'm definitely noticing improved drive ability and economy!
My car was dyno tuned so I don't think WOT will need much work, but one thing I am a bit hazy on:
These formulas only work if the data is collected at steady state, as far as I am aware. So my question is how does WOT come into the equation in relation to steady state?

The formulas work no matter what. They are just crunching numbers and spitting out a result. The best outcomes occur when you are in steady state, or near steady state, so you prevent the ECM from using transient fueling. One of the hang ups that a lot of people have when tuning this stuff is that they expect/assume you have to be in steady state for this to work (using Dynamic Air or any other method). To be fair, prior methods did rely on steady state. However, with Dynamic Air you do NOT have to be in steady state. All we need is the DynAir value and the Fuel Trims that go along with it and then we can apply that to the MAF and VVE. We really just want transient fuel to be removed from the equation (no pun intended).

At WOT the ECM uses MAF. Everything I am seeing of late is pointing me in the direction that once you hit 4,000 RPMs, the ECM completely ignores VVE and relies solely on MAF (regardless of where you have the Dynamic Air RPM enable/disable set to).

[lsxjunkie]

The formulas work no matter what. They are just crunching numbers and spitting out a result. The best outcomes occur when you are in steady state, or near steady state, so you prevent the ECM from using transient fueling. One of the hang ups that a lot of people have when tuning this stuff is that they expect/assume you have to be in steady state for this to work (using Dynamic Air or any other method). To be fair, prior methods did rely on steady state. However, with Dynamic Air you do NOT have to be in steady state. All we need is the DynAir value and the Fuel Trims that go along with it and then we can apply that to the MAF and VVE. We really just want transient fuel to be removed from the equation (no pun intended).

At WOT the ECM uses MAF. Everything I am seeing of late is pointing me in the direction that once you hit 4,000 RPMs, the ECM completely ignores VVE and relies solely on MAF (regardless of where you have the Dynamic Air RPM enable/disable set to).

Using your tools is it ok tht LTFT is disabled or should it be enabled for the formula to work?

[Cringer]

Using your tools is it ok tht LTFT is disabled or should it be enabled for the formula to work?

There are two versions of the formulas, one for STFT only and one for STFT + LTFT. Use the correct one and you will be fine.

[dassaur]

The formulas work no matter what. They are just crunching numbers and spitting out a result. The best outcomes occur when you are in steady state, or near steady state, so you prevent the ECM from using transient fueling. One of the hang ups that a lot of people have when tuning this stuff is that they expect/assume you have to be in steady state for this to work (using Dynamic Air or any other method). To be fair, prior methods did rely on steady state. However, with Dynamic Air you do NOT have to be in steady state. All we need is the DynAir value and the Fuel Trims that go along with it and then we can apply that to the MAF and VVE. We really just want transient fuel to be removed from the equation (no pun intended).

At WOT the ECM uses MAF. Everything I am seeing of late is pointing me in the direction that once you hit 4,000 RPMs, the ECM completely ignores VVE and relies solely on MAF (regardless of where you have the Dynamic Air RPM enable/disable set to).

Super helpful, thanks!

So you're saying that even in stop start traffic, assuming all the usuals are disabled and the car is sufficiently warm (DFCO, COT, desoot etc), the logs could be used for adjusting maf and VVE?

I also should've been clearer - rather than saying these formulas possibly only work in steady state, I meant the entire methodology - but you've clarified either way, thanks!

[Cringer]

So you're saying that even in stop start traffic, assuming all the usuals are disabled and the car is sufficiently warm (DFCO, COT, desoot etc), the logs could be used for adjusting maf and VVE?

Yes, that is exactly the power of using this method. Prior to this, guys would drive their best to be in steady state to tune MAF only, then disable MAF to tune VVE. It was a two step process and you would have to modify the tune. It meant a lot of driving wasting gas to collect logs. Then along came the GMVE method would uses sensor data and math to predict what the ECM determined the cylinder airmass to be so you could tune VVE without disabling MAF. Using DynAir is the next evolution which allows you to do both at the same time, and do it faster and more accurate than the good ol days.

Once you combine these formulas with the VVE and MAF programs I developed you will spend even less time driving around to collect logs.

[dassaur]

Great! I have been using all 3 of your programmes, very thankful to you.

[hjtrbo]

Hear ye, hear ye

I've made a minor adjustment to my filters that are included in this method. I have been using these for the past 6 months. Whilst not 100% perfect, they capture more fringe area steady state data which over the course of a decent street tune session produces a smoother grouping of error values. In my opinion the new filters are superior. I've added extra conditions to allow the VVE table to capture coasting events hence the split up of the filters.

My old closed loop filter
CL: [2517.161.avg(1500)]=0 and [2517.161.avg(-200)]=0 and (abs([50090.156.slope(1500)])+abs([50090.156.slope(-500)]))<2 and ([6310]=9 OR [6310]=14 OR [6310]=15)=0

My New VE Table Filter
CL: [50017.241]>80 and [2320.71.slope(-500)]<4 and [2320.71.slope(-500)]>-6 and [2320.71.slope(-1500)]<999 and[2320.71.slope(-1500)]>-999 and [2320.71.slope(500)]<8 and [2320.71.slope(500)]>-4 and [2320.71.slope(1500)]<999 and [2320.71.slope(1500)]>-18 and [6215.241]<60 and [2517.161.avg(500)]=0 and [2517.161.avg(-500)]=0 and ([6310]=7 and [6310.slope(2000)]>0.1)=0 and ([6310]=9 OR [6310]=14 OR [6310]=15)=0 and [50072.slope(2000)]<0.1 and [50072.slope(-2000)]<0.1

My New MAF Table Filter
CL: [50017.241]>80 and [2320.71.slope(-500)]<4 and [2320.71.slope(-500)]>-6 and [2320.71.slope(-1500)]<999 and[2320.71.slope(-1500)]>-999 and [2320.71.slope(500)]<8 and [2320.71.slope(500)]>-4 and [2320.71.slope(1500)]<999 and [2320.71.slope(1500)]>-18 and [6215.241]<60 and [2517.161.avg(500)]=0 and [2517.161.avg(-500)]=0 and ([6310]=7 OR [6310]=9 OR [6310]=14 OR [6310]=15)=0 and [50072.slope(2000)]<0.1 and [50072.slope(-2000)]<0.1

Same random old large assed data log
OLD FILTER


NEW FILTER


60% INCREASE IN CAPTURED DATA!!!

[hjtrbo]

EDIT 2: THESE FILTERS WILL BOG DOWN THE SCANNER WHEN OPENING LARGE FILES. THEY'RE NOT VERY EFFICIENT.

OOH, FORGOT TO REMIND AND MENTION AGAIN, IF BOOSTED OR SOME CRAZY N/A BUILD, USE CAUTION WITH THIS TUNING METHOD. (smokeshows method does not have this limitation)

512g/s VE AIRFLOW LIMIT!!!
655g/s MAF AIRFLOW LIMIT!!!

[PGA2B]

Have you changed your open loop filters too?

[GHuggins]

I assume you corrected for the neg MAF values during transients a different way to not need the ABS or absolute (positive values used only) function? Just glancing quickly at it.

I'll give it a try. I would recommend using a higher temp filter for coolant for anyone using it on multiple platforms. Just another one that jumped out. Have to plug in to see the rest

I use cylinder airmass in the instances where the formula maxes out. Surprisingly only really happens in boost even with the 427's I've done.

[PGA2B]

50017.241 is Engine Oil Temp in Celsius and 6215.241 is Injector Temp Temp in Celsius for me.