What is dynamic airflow

[SiriusC1024]

Ok it's not that it's too sensitive. The filter isn't working right.

[SiriusC1024]

Nvm maybe it is too sensitive. The problem I was having was a parenthesis that I seemed to miss while copying the formula. Wouldn't correct without restarting the application.

[SiriusC1024]

This is the filter (pretty sure):

(ABS([50040.71]-[50040.71.shift(-12)])/.012) < 10 & (ABS([2311.71]-[2311.71.shift(-12)])/.012) < 10

filter_mod.png

This is simply slope. Rise/Run. I'm trying to rationalize why slope of 10 is the same as a gain constant of 0.1. Made sense to me when I came up with it. I'm burned out right now and really need a break. Boolean calc a bit ago yielding a 4 (wtf) got to me lol. Plus I just got off a 12-hr shift at 5:30am.

[hjtrbo]

I guess your next thing to solve is baking in the transport delay to the O2 sensors at various exhaust gas velocities.

Maybe the "simple" filters plebs like me use are good enough to get the job done...

[SiriusC1024]

I think k = 0.020 and Gain = 20 is a good filter balance. .010 and 10 tighten things up a bit.

If the Boolean filter is overlaid on TPS it shows how effective this filter is. Throttle transients are filtered. What differentiates this method from filtering TPS is that focus is on airflow. While TPS may be relatively stable, one of the airflow models may be in flux. Also, while TPS may be filtered out in the TPS filter method, there are still opportunities there to collect data.

Here's k = 0.020 and Gain = 20
throttle_ss.png

Of course, the slight decrease in filter effectiveness from adjusting the values can be seen.

Think of it this way:
- Lower k means higher resolution.
- Lower G means more sensitivity.

***All units must me set within the filter to g/s for these values of k and G to work. Logged channels can be whatever. It's automatically converted.

[Cringer]

Update:
hjtrbo reminded me something I encountered a while back and forgot about. In the scanner MAF maxes out at 512 g/s and VVE maxes out at 655 g/s. This screenshot was taken from a log on the 2 bar OS, but appears to also be an issue on the standard E38 OEM OS. So if you are moving a lot of air, this is something to be keenly aware of!

Not sure if all Gen 4 and 5 OS's are the same in this regard though.

There are some work arounds for this, I will post them in a bit....


max.jpg

[SiriusC1024]

Oh FFS. I think I figured it out. Gotta d/dt both sides (ish). That's what I couldn't rationalize. The units weren't canceling. A convenient factor of 100 had me hung up between percentage change and slope.

So it works a little differently here if it's assumed GM is using Gain Constant of 0.1.

Filter with a sample interval of 12ms becomes.
k = 12ms (.012s)
Gain = 0.1

(ABS([50040.71]-[50040.71.shift(-12)])/.012) < .1/.012 & (ABS([2311.71]-[2311.71.shift(-12)])/.012) < .1/.012

Try a few sample intervals. Around 10-20ms sampling interval gets the best results.

[SiriusC1024]

Update:
hjtrbo reminded me something I encountered a while back and forgot about. In the scanner MAF maxes out at 512 g/s and VVE maxes out at 655 g/s. This screenshot was taken from a log on the 2 bar OS, but appears to also be an issue on the standard E38 OEM OS. So if you are moving an air, this is something to be keenly aware of!

Not sure if all Gen 4 and 4 OS's are the same in this regard though.


max.jpg

So will it work with lb/hr then? If logs can be done in lb/hr it shouldn't matter if the filter math does the conversion from lb/hr to g/s without the limit.

Log in whatever. Graph whatever. Filter formula in g/s.

The g/s is used because I'm inferring GM did all this stuff in metric. lb/hr readings in the tune I figured was to keep it consistent with convention on HPT's part.

[Cringer]

So will it work with lb/hr then? If logs can be done in lb/hr it shouldn't matter if the filter math does the conversion from lb/hr to g/s without the limit.

Log in whatever. Graph whatever. Filter formula in g/s.

The g/s is used because I'm inferring GM did all this stuff in metric. lb/hr readings in the tune I figured was to keep it consistent with convention on HPT's part.

The units don't matter, it caps at the equivalent unit conversion. I did edit the post to indicate there are work arounds. I will post those as time permits later today.

[sgod1100]

Update:
hjtrbo reminded me something I encountered a while back and forgot about. In the scanner MAF maxes out at 512 g/s and VVE maxes out at 655 g/s. This screenshot was taken from a log on the 2 bar OS, but appears to also be an issue on the standard E38 OEM OS. So if you are moving a lot of air, this is something to be keenly aware of!

Not sure if all Gen 4 and 5 OS's are the same in this regard though.

There are some work arounds for this, I will post them in a bit....


max.jpg

Yeah, I've seen this as well with my ctsv

[GHuggins]

4th and 5th gens both cap out. It's a scanner issue that no one seemed to care about fixing ;/

I like the idea of using airflow as the filter. This will especially become necessary on the 5th gens where you can be at 50 or 60 % throttle and the tb will open or close based on torque request and since there isn't much actual airflow change from say 70% tps to say 100% tps on a NA vehicle I would think this would be more effective. Would like the formula to work on both

[Cringer]

So there are several different tactics to solve the data cap. One question remains unanswered, if MAF and VVE have caps in the scanner...what is the cap for Dynamic Air? If we look in the editor, the MAF airflow caps at 4,000 g/s and who knows what the VVE limit is (probably way more than 4,000 g/s). So I would *assume* Dynamic Air caps at 4,000 g/s. I think we are all safe LOL



For MAF, there is only one decent solution that I can think of. Basically multiply the dynamic air by the fuel trims to obtain the corrected airflow in grams per second, then you would just do a normal copy paste from the scanner to the editor for the MAF table. Again, this is NOT percent, do not attempt to copy paste special %.

MAF - CL (STFT Only)
[2320.71] + ([2320.71] * (([50116.156]+[50114.156])/100))

MAF - PE (replace [50127] with your wideband PID)
[2320.71] + ([2320.71] * (([50127.238]-[50118.238])]/[50118.238]/100))

VVE poses a challenge though because its native values are not in grams/sec (I think its the molar mass of something or other). However, the scanner does reveal some key information when you log the "volumetric efficiency" channel.
VVE (mg*K/KPa)

That is cylinder airmass (in milligrams) times the intake air temp (in Kelvin) divided by the MAP (in KPa) = the VVE literal values in the editor.
So since we know cylinder airmass is derived from dynamic air...then we can calculate the new VVE literal values (again, not percent, but the final numbers).

VVE Option #1: CL STFT + LTFT
([50041.224]+([50041.224]*(([6.156]+[7.156]+[8.156]+[9.156])*.5)/100)) * [2126.240] / [50030.91]

VVE Option #2:
Basically the same thing except rather than converting cylair to VVE (mg*K/KPa) terms, you would convert VVE (mg*K/KPa) to grams/sec to directly compare to the dynair value. One way may be better than the other, IDK.

I will say I think the VVE side is going to be sketchy. However, it can also be converted into a percent by dividing against the VVE (mg*K/KPa) channel...and I am too tired to do that...my eyes are bleeding from looking at all the parenthesis at this point! So if someone wants to flesh this out more, please go for it.

Assuming there are no errors in these formulas, they would need to be built out for STFT only, STFT+LTFT, and wideband only.

Here are some general formulas in case anyone else wants to experiment or make corrections.

1) convert [grams/sec] to [cylinder airmass] (4 is the constant for 8 cylinders)
    [cylinder airmass.grams] = [dynamic air.grams/sec] * 60 / 4 / [RPM]
    
    [cylinder airmass.grams] = [2320.71] * 60 / 4 / [50070]




2) convert [cylinder airmass] to [grams/sec] (4 is the constant for 8 cylinders)
    [dynamic air.grams/sec] = [cylinder airmass.grams] * [RPM] / 60 * 4
    
    [dynamic air.grams/sec] = [50041.223] * [50070] / 60 * 4




3) convert [cylinder airmass] to [VVE (mg*K/KPa)]
    [VVE Literal Value] = [cylinder airmass.milligrams] * [MAT.Kelvin] / [MAP.KPa]
    
    [VVE Literal Value] = [50041.224] * [2126.240] / [50030.91]



4) convert [VVE (mg*K/KPa)] to [cylinder airmass]
    [cylinder airmass.milligrams] = [VVE Literal Value] / [MAT.Kelvin] * [MAP.KPa]
    
    [cylinder airmass.milligrams] = [2312] / [2126.240] * [50030.91]

[SiriusC1024]

I'm not quite following. How have people been using DynAir before if there's a cap? It's not just the filter itself that wouldn't work.

[SiriusC1024]

I will say I think the VVE side is going to be sketchy. However, it can also be converted into a percent by dividing against the VVE (mg*K/KPa) channel...and I am too tired to do that...my eyes are bleeding from looking at all the parenthesis at this point! So if someone wants to flesh this out more, please go for it.

Notepad++ helps with that. It highlights the other parenthesis of a pair.

notepad.png

Back in the day one of the nice things about the upgrade from Maple 9 to Maple 10 was that when a parenthesis was typed the formula would automatically adjust cursor point as if writing by hand. I'll see if there's something like that for free.

[Cringer]

I'm not quite following. How have people been using DynAir before if there's a cap? It's not just the filter itself that wouldn't work.

I don't think people ever used dynair before. It was always some point of curiosity that made people ask questions, and then move on to something else and was never used.

However, as far as I know, we have not found the cap for Dynamic Airflow in the scanner, just for MAF g/s and VVE g/s. And don't ask me why they are 512 and 655 g/s and not the same value.

The reason why this is a problem with the formulas I developed is because that is how the % is calculated. Prior, to that it was basically drive steady state, assume MAF=DynAir and then apply the fuel trims back to MAF with a copy paste special. Then go into SD to work on the VVE table. Neither of those solutions required the MAF or VVE values from the scanner.

[Cringer]

I'm not quite following. How have people been using DynAir before if there's a cap? It's not just the filter itself that wouldn't work.

Just to be clear, if you open the tune file, the editor shows the MAF caps at 4,000 g/s. But when scanning an air hungry engine, the MAF seems to stop logging beyond 512 g/s. The ECM is still able to make the engine run with the proper air calculations, there is just a disconnect between the ECM and Scanner. I am not sure it is HPT's fault, it just may be a built in limitation of the E38, or canbus, but something is just limiting it to 10 bits of data.

maf 4k.jpg

So 512 makes sense in terms of 10 bits of data. Now where the heckin does 655 come from?

[SiriusC1024]

I don't think people ever used dynair before. It was always some point of curiosity that made people ask questions, and then move on to something else and was never used.

However, as far as I know, we have not found the cap for Dynamic Airflow in the scanner, just for MAF g/s and VVE g/s. And don't ask me why they are 512 and 655 g/s and not the same value.

The reason why this is a problem with the formulas I developed is because that is how the % is calculated. Prior, to that it was basically drive steady state, assume MAF=DynAir and then apply the fuel trims back to MAF with a copy paste special. Then go into SD to work on the VVE table. Neither of those solutions required the MAF or VVE values from the scanner.

Wait what? Haven't you been tuning with DynAir? What were you guys going on about on the first part of this thread. Post#22?

For what it's worth, I originally tuned my car (2013 Camaro w/headers + intake) doing MAF and VVE separately with my wideband in open loop. Fuel trims back in closed loop after were within 1-2%. All was well. About a month ago I found Cringer's method using dynamic air to do MAF and VVE simultaneously so I flashed my stock MAF curve and VVE table back in and tried his way out of curiosity. Within 2 logs (Steady state with minimal transients) my fuel trims are all within 1-2% again and took significantly less time to dial in.

Smokeshow, to be fair I haven't tried your method to tune MAF and VVE simultaneously yet but at this point I almost feel obligated to see how well it works. If I get some time and good weather next week I might give it a shot and report back. Regardless, if both methods work well enough then who cares? Unless someone is smart enough to figure out exactly how Dynamic Air is calculated then there won't really be a right way to do anything.

I'm no pro tuner by any means. Just figured I would share my experience so far.

[Cringer]

Wait what? Haven't you been tuning with DynAir? What were you guys going on about on the first part of this thread. Post#22?

I have been using dynamic air for quite some time now. But I was responding to your question: How have people been using DynAir

And "people" have not been using it. It was never an issue for me since my engine run's the way god intended it to (N/A) and I still have some breathing room (pun intended) before I hit the MAF/VVE cap in the scanner.

[hjtrbo]

512 is an unsigned byte, 65535 is an unsigned word. Picking that's where these scanner limits come from.

[GHuggins]

Dyn air doesn't have a cap in the scanner to my knowledge or at least one I've seen so far.

I was also told that a regular scan tool will display g/s airflow on up past this, so an hpt issue. Seems kind of stupid to me, but whatever... I don't personally recall ever having a gm tool hooked up that it capped out personally, but it's been a few years too.