Help me understand intergral and proportional airflow tables

[Allen Vos]

Hi guys, I am looking for a thorough explanation on the integral and proportional air flow tables and specifically how they work and translate the numbers into the required air flow.

I am not new to tuning, but I am new to Gen 4 stuff and all the different strategies used to control air flow and fuel.

I do appreciate all the help I have been getting here, however I want to understand how these tables work so I have a better understanding and not making changes blindly on what others recommend.

Here is a pic of my Integral and Proportional air flow tables for my 2009 LS3 Vette/Manual and I've included the proportional table from the same car with an automatic which appears to have different settings.

Attachment 142316

From what I understand the normal section of both tables are used in park or neutral. The coast down part of the table is used on coast down while the car is slowing down?

I am assuming they work the same as the over and under speed spark tables where as "0" is reference to for the target idle?

as soon as it deviates from the target plus or minus whatever the RPM is at on the table, it uses that number along with the multiplier table based on ECT to calculate the necessary airflow to add in my case
as Grams per Second.


I could be way off here, so please correct me if I am wrong.

I noticed on the proportional air flow table that the coast down only in the negative RPM's are all Zeroed out on the manual transmission.

I'm assuming this is because most people just put the clutch in when slowing down so coast down is not needed?

How does this work for guys who slow down and use the engine as a brake to help slow the car down?

[GHuggins]

I understand where you're wanting to go with this and it's always good to know how things work, but the majority of your idle tuning will be on the airflow side. I can give you the basics of how things work and you can take it from there.

In the adaptives you have your main two like what you're talking about. Proportional is immediate or your main big throttle controls. Integral is small and primarily ignition linked controlled. Both of these add on the right side of the table and take away on the left. The other important players that I think you might be missing are the airflow and rpm settings for those, not to mention the speed setting controllers for them.

At the top you have min rpm ref - this is the min rpm for the adaptives. A lot of people mistakenly raise this to the idle rpm with cams because it's set around the idle rpm with the factory settings, but as we know cams like to lope and the last thing you want is it jumping in and out of idle adaptives. This should be left down around 500 rpms.

At the bottom you have airflow limits - cams add airflow - pay attention to the g/s or lb/hr settings in this table and how that may have changed in comparison to how it was setup from the OE operations. Airflow Max under throttle follower needs to be observed for the same reason. Only set them a gram or two above observed idle airflow.

Decrement steps - higher numbers close the throttle quicker. Lower numbers slower. Typically only ever had to touch this on big cammed motors and even then it was rare. You can smooth up throttle plunges with clutch in's using this alone while keeping the airflow tables nearly stock.

There's a couple of other important players on top of this that people often don't even consider.

Effective area - min, max settings. Min combined with percent max will be your throttle stops. Big cam = more air = throttle needs to be held open more. This will be achieved not only here, but in your min air table. If you're increasing your min air ANY at idle above the 1000 or 1200 rpm columns then these settings need to be raised. Use the scanner and set these so at a hot idle you can still force it to idle at least 150 to 200 rpms below commanded. This way you still have a safety net designed into it for vacuum leaks or whatever. The min settings should never need to go above 80 from everything I've ever seen. Min brake is how much the throttle will close for engine braking.

Throttle follower. Not entirely sure, but I've always set it just under delivered torque. Again, this way the throttle has the capability of closing if needed. I like to think of this table like how the pedal progression combined with the DD tables are used in the newer stuff, so set under delivered and mostly negative everywhere. The torque model on it's own will help A LOT with throttle and idle control. 4th gens respond well to plugging delivered right back into the models even though it's not correct.

Then since it's all airflow controlled and airflow = fuel, fueling needs to be dialed in at full operating temp before touching anything including the very basics such as inj timing and O2 settings.

[TragicMike]

At the top you have min rpm ref - this is the min rpm for the adaptives. A lot of people mistakenly raise this to the idle rpm with cams because it's set around the idle rpm with the factory settings, but as we know cams like to lope and the last thing you want is it jumping in and out of idle adaptives. This should be left down around 500 rpms.

Use the scanner and set (effective area) so at a hot idle you can still force it to idle at least 150 to 200 rpms below commanded. This way you still have a safety net designed into it for vacuum leaks or whatever. The min settings should never need to go above 80 from everything I've ever seen. Min brake is how much the throttle will close for engine braking.

Hero status. Thank you.

One question:

Uhm... what Torque model?
Did you mean the 12123 Throttle follower table itself, or something in the Torque Model tab?

At first I thought you were refering to something in the software, or an outside program.

[Cringer]

Hi guys, I am looking for a thorough explanation on the integral and proportional air flow tables and specifically how they work and translate the numbers into the required air flow.

I am not new to tuning, but I am new to Gen 4 stuff and all the different strategies used to control air flow and fuel.

I do appreciate all the help I have been getting here, however I want to understand how these tables work so I have a better understanding and not making changes blindly on what others recommend.

Here is a pic of my Integral and Proportional air flow tables for my 2009 LS3 Vette/Manual and I've included the proportional table from the same car with an automatic which appears to have different settings.

Attachment 142316

From what I understand the normal section of both tables are used in park or neutral. The coast down part of the table is used on coast down while the car is slowing down?

I am assuming they work the same as the over and under speed spark tables where as "0" is reference to for the target idle?

as soon as it deviates from the target plus or minus whatever the RPM is at on the table, it uses that number along with the multiplier table based on ECT to calculate the necessary airflow to add in my case
as Grams per Second.


I could be way off here, so please correct me if I am wrong.

I noticed on the proportional air flow table that the coast down only in the negative RPM's are all Zeroed out on the manual transmission.

I'm assuming this is because most people just put the clutch in when slowing down so coast down is not needed?

How does this work for guys who slow down and use the engine as a brake to help slow the car down?

The Proportional and Integral settings are part of a tuned system. You cannot just go in and willy nilly start changing and experimenting with random values and expect an improvement in your situation. Return to stock and do not touch them. Forget they exist.

You can look up a PID controller to learn more...but this is just a PI controller...or if the Derivative component exists, then HPTuners has chosen to not exposed it.

Proportional responds quickly to large errors. The bigger the error, the harsher the response. It's behavior is in proportion to the error...big error = big response.

Integral does not care about the size of the error, it only cares about how long the error persists. The longer the error = the greater the response. It is slow to react, but it adds more and more correction the longer the error persists.

So you can see there is a system of checks and balances going on.

During an error, Proportional (P) makes a quick change to knock the error condition down quite a bit. But after the initial error correction P loses interest and make less and less of a contribution to error correction...to the point that error correction is not going on.

It is at this time we expect a perfect hand off then to Integral (I) who sees the lingering error condition has been going on longer than expected and starts to work on it.

Derivative (D) then is the final check and balance to ensure we do not over or undershoot the collective correction from P and I.

So if you change P to be more of less active, then when you hand the error condition off to I...he is not going know how to correctly respond and end up over/under reacting and throw the whole error system management off. It is also important to understand that the [Total Error Correction] = [P] + [I] + [D]. So while I am making it seem like P works, then I, then D...that is not the case. All 3 (or 2 in our case) are active at all times working on an error condition.


If you change P then you have to change I. If you change I then you have to change P. And since this is a tuned system, you cannot arbitrarily change one and then tune the other. There is a very narrow set of acceptable values that works in controlling a system. To further complicate things, GM has added A LOT of other ancillary values to this PID controller (for under/over, max values, coast/idle, steps, etc). All these values work with the physical throttle body (diameter and TAC tables, etc).

This is a case of less is more, as in the less values you change from OEM the better.

[rabbs88]

Hero status.
Uhm... what Torque model?

Under Edit -> Virtual Torque

https://www.hptuners.com/help/VCM-Ed...ual_torque.htm

[Allen Vos]

I understand where you're wanting to go with this and it's always good to know how things work, but the majority of your idle tuning will be on the airflow side. I can give you the basics of how things work and you can take it from there.

In the adaptives you have your main two like what you're talking about. Proportional is immediate or your main big throttle controls. Integral is small and primarily ignition linked controlled. Both of these add on the right side of the table and take away on the left. The other important players that I think you might be missing are the airflow and rpm settings for those, not to mention the speed setting controllers for them.

At the top you have min rpm ref - this is the min rpm for the adaptives. A lot of people mistakenly raise this to the idle rpm with cams because it's set around the idle rpm with the factory settings, but as we know cams like to lope and the last thing you want is it jumping in and out of idle adaptives. This should be left down around 500 rpms.

At the bottom you have airflow limits - cams add airflow - pay attention to the g/s or lb/hr settings in this table and how that may have changed in comparison to how it was setup from the OE operations. Airflow Max under throttle follower needs to be observed for the same reason. Only set them a gram or two above observed idle airflow.

Decrement steps - higher numbers close the throttle quicker. Lower numbers slower. Typically only ever had to touch this on big cammed motors and even then it was rare. You can smooth up throttle plunges with clutch in's using this alone while keeping the airflow tables nearly stock.

There's a couple of other important players on top of this that people often don't even consider.

Effective area - min, max settings. Min combined with percent max will be your throttle stops. Big cam = more air = throttle needs to be held open more. This will be achieved not only here, but in your min air table. If you're increasing your min air ANY at idle above the 1000 or 1200 rpm columns then these settings need to be raised. Use the scanner and set these so at a hot idle you can still force it to idle at least 150 to 200 rpms below commanded. This way you still have a safety net designed into it for vacuum leaks or whatever. The min settings should never need to go above 80 from everything I've ever seen. Min brake is how much the throttle will close for engine braking.

Throttle follower. Not entirely sure, but I've always set it just under delivered torque. Again, this way the throttle has the capability of closing if needed. I like to think of this table like how the pedal progression combined with the DD tables are used in the newer stuff, so set under delivered and mostly negative everywhere. The torque model on it's own will help A LOT with throttle and idle control. 4th gens respond well to plugging delivered right back into the models even though it's not correct.

Then since it's all airflow controlled and airflow = fuel, fueling needs to be dialed in at full operating temp before touching anything including the very basics such as inj timing and O2 settings.

Thank you for that explanation. It makes a bit more sense to me now.

Here is a screen shot of my proportional and integral idle tables.

Attachment 142407

My first question or assumption is that the "0" in the center of each table represents target idle RPM?

Im sorry if that seems like a dumb question, but I just want to make sure.

If this assumption is correct then postive rpm numbers get refereenced for over the tarrget RPM and ngative RPM numbers are referenced for under the target rpm?

Now my biggest question is, what exactly do the numbers do?

How do those numbers get converted into the air flow(G/S) that we see being added in on the scanner on the integral and proportional G/S values?

I had also noticed the negative rpm numbers on the normal scale from the proportional table are negative. Is this normal? and why?

The proportional table numbers on my negative RPM on coast down are all "0"'s. Is this because I have a manual transmission?


Here are some error/multiplier tables from the adaptive idle settings.

Attachment 142408

Are these the multiplier tables for proportional and integral tables? I have noticed these tables and wondered what they are for?

Or are these tables for the spark over and under speed tables?

[Allen Vos]

The Proportional and Integral settings are part of a tuned system. You cannot just go in and willy nilly start changing and experimenting with random values and expect an improvement in your situation. Return to stock and do not touch them. Forget they exist.

You can look up a PID controller to learn more...but this is just a PI controller...or if the Derivative component exists, then HPTuners has chosen to not exposed it.

Proportional responds quickly to large errors. The bigger the error, the harsher the response. It's behavior is in proportion to the error...big error = big response.

Integral does not care about the size of the error, it only cares about how long the error persists. The longer the error = the greater the response. It is slow to react, but it adds more and more correction the longer the error persists.

So you can see there is a system of checks and balances going on.

During an error, Proportional (P) makes a quick change to knock the error condition down quite a bit. But after the initial error correction P loses interest and make less and less of a contribution to error correction...to the point that error correction is not going on.

It is at this time we expect a perfect hand off then to Integral (I) who sees the lingering error condition has been going on longer than expected and starts to work on it.

Derivative (D) then is the final check and balance to ensure we do not over or undershoot the collective correction from P and I.

So if you change P to be more of less active, then when you hand the error condition off to I...he is not going know how to correctly respond and end up over/under reacting and throw the whole error system management off. It is also important to understand that the [Total Error Correction] = [P] + [I] + [D]. So while I am making it seem like P works, then I, then D...that is not the case. All 3 (or 2 in our case) are active at all times working on an error condition.


If you change P then you have to change I. If you change I then you have to change P. And since this is a tuned system, you cannot arbitrarily change one and then tune the other. There is a very narrow set of acceptable values that works in controlling a system. To further complicate things, GM has added A LOT of other ancillary values to this PID controller (for under/over, max values, coast/idle, steps, etc). All these values work with the physical throttle body (diameter and TAC tables, etc).

This is a case of less is more, as in the less values you change from OEM the better.

Thank you for that explanation. It makes more sense to me now and I now know to just leave them the hell alone lol

I do have a question for you though.

I have seen in some of your videos and posts that you recommend using the 2014 Camaro SS proportional and integral idle tables.

Is it ok to use these as long as the tables are untouched? I'm not suggesting I want to do that, but just wondering if that could ever be an option for me once I get more comfortable with tuning?

I'm sorry its a bit confusing when people say to leave mine stock and then I see your recommendation.

I just want to make some sense of it all.

[GHuggins]

Again, the majority of your corrections will be elsewhere in a 4th gen and the "elsewhere" needs to be dialed in before even thinking about making corrections to any of this. However the airflow "limits" need to be set to your new idle airflow operating range. Read up on the min airflow thread, especially after Smokeshow says the operating strategy of these ECU's.

This is how I've taught myself to think of the adaptives. Are they pid controllers. Absolutely, but they're still dialed in for that engines current TB, airflow and torque. With that in mind, I truly believe there's a little miss-labeling on the adaptive tables. Yes they are representing rpm error and airflow, but they are in fact correcting for "torque errors" or at least they seem to be to me. With that train of thought re-look at the tables. To increase rpm and thus increase torque, you open the throttle plate (right side of the tables). If you need to decrease rpm and thus decrease torque, you then close the throttle plate (left side of the table). They're exactly the same going forward to the newer gens too, which funny enough almost everybody out there adjust. Go figure right Anyone can correct me if I wrong about this, but all of my own testing and tuning has shown just that. You can sometimes REALLY smooth out a big cam manual cars drivability with just lowering and smoothing about 4 cells to the right in the proportional table. Just depends on what's going on. Now am I saying go and adjust yours like that - NO, I'm saying if you still have problems down the road then it doesn't hurt to smooth things up a touch.

[Cringer]

Thank you for that explanation. It makes more sense to me now and I now know to just leave them the hell alone lol

I do have a question for you though.

I have seen in some of your videos and posts that you recommend using the 2014 Camaro SS proportional and integral idle tables.

Is it ok to use these as long as the tables are untouched? I'm not suggesting I want to do that, but just wondering if that could ever be an option for me once I get more comfortable with tuning?

I'm sorry its a bit confusing when people say to leave mine stock and then I see your recommendation.

I just want to make some sense of it all.

My LS3 car with a big cam ended up with the '14 Camaro SS PID values (this was the suggestion of the previous professional tuner for the car). The surging idle and stalling experience was a journey for me, and is the whole reason that got me into buying HPTuners and learning to tune. I thought that learning the PID's was the answer. It was not. Simply applying the Camaro PIDs did not solve anything for me. What did solve my issues was sorting out the min final airflow table, base spark, and OEM over/under spark values. I have made some very minor tweaks to over/under spark (as in .1 to .5 or so), not technically stock.

I still have the Camaro PID values in there, I should have taken them out, but they were forgotten over time as I was working out other more important issues...I suppose it would be a good experiment to roll back to my stock values (SS Sedan) to see what happens. So I guess I suggest the Camaro ones as an alternative since they seem to be working for me, it is an option and a config that is known to work. However, I think your OEM 'vette PID values will work just fine for you.

Bottom line though, I would just focus on the fundamentals for idle. The airflow PIDs are just a distraction from the finish line.

[edcmat-l1]

This is how I've taught myself to think of the adaptives. Are they pid controllers. Absolutely, but they're still dialed in for that engines current TB, airflow and torque. With that in mind, I truly believe there's a little miss-labeling on the adaptive tables. Yes they are representing rpm error and airflow, but they are in fact correcting for "torque errors" or at least they seem to be to me. With that train of thought re-look at the tables. To increase rpm and thus increase torque, you open the throttle plate (right side of the tables). If you need to decrease rpm and thus decrease torque, you then close the throttle plate (left side of the table). They're exactly the same going forward to the newer gens too, which funny enough almost everybody out there adjust. Go figure right Anyone can correct me if I wrong about this, but all of my own testing and tuning has shown just that. You can sometimes REALLY smooth out a big cam manual cars drivability with just lowering and smoothing about 4 cells to the right in the proportional table. Just depends on what's going on. Now am I saying go and adjust yours like that - NO, I'm saying if you still have problems down the road then it doesn't hurt to smooth things up a touch.

I'd have to disagree. The GEN4 is airflow. It is blade control. The GEN5 is torque based which in that platform is timing. Adjusting the GEN5 adaptive idle to me is akin to the GEN4 over/under spark tables. You adjust them and it smooths out the saw tooth pattern. 2 completely different approaches with 2 completely different outcomes.

I know you're one of the smart guys, and I'm open to listen to a difference of opinion, but to me these are 2 completely different things here and you'd have to convince me they aren't. The GEN4 PIDs and GEN5 adaptive idle are NOT the same.

[GHuggins]

What does predictive do? AND I should have said that they are set up the same, not that they are the same. Immediate is timing. Predictive is throttle as proportional is largely throttle and integral even though it's an added learned value also seems to largely influence adaptive timing. Again, that's just my observations with them.

[Cringer]

I'd have to disagree. The GEN4 is airflow. It is blade control. The GEN5 is torque based which in that platform is timing. Adjusting the GEN5 adaptive idle to me is akin to the GEN4 over/under spark tables. You adjust them and it smooths out the saw tooth pattern. 2 completely different approaches with 2 completely different outcomes.

I know you're one of the smart guys, and I'm open to listen to a difference of opinion, but to me these are 2 completely different things here and you'd have to convince me they aren't. The GEN4 PIDs and GEN5 adaptive idle are NOT the same.

Greg's comment is reflective of some recent postings by smokeshow:

https://forum.hptuners.com/showthrea...l=1#post766182
https://forum.hptuners.com/showthrea...l=1#post766201
https://forum.hptuners.com/showthrea...l=1#post766205

From my experience it is all airflow too...and more air = more torque.

[hjtrbo]

Just jumping in to answer an earlier question...

Integral numbers are the increment / decrement per calc cycle.

In the (unusual) case where the error remains constant e.g -150 rpm and the PI control loop runs every 50ms, then every 50ms add the value from the -150rpm cell onto the previous integral airflow.
If this goes on for ever then the integral airflow will be huge (wind-up). So we limit how high it can go. And vice-versa.

P, a correction value proportional to the error
I, a correction value that if an error is present will increase or decrease over time
D, a correction value based on the rate of change of the error

Another thing to keep in mind specific to an engine when looking to modify a behaviour, slow and fast actuators. Spark = Fast. Throttle = Slow

Edit:
In your case with wanting to bleed the air off faster to close your throttle after a blip, I would first try and work out why the hell the thing thinks it's a good idea to preload the integral term with that much air in the first place.

Continue with your airflow / torque discussion gents

[GHuggins]

Integral numbers are the increment / decrement per calc cycle.

In the (unusual) case where the error remains constant e.g -150 rpm and the PI control loop runs every 50ms, then every 50ms add the value from the -150rpm cell onto the previous integral airflow.
If this goes on for ever then the integral airflow will be huge (wind-up). So we limit how high it can go. And vice-versa.

P, a correction value proportional to the error
I, a correction value that if an error is present will increase or decrease over time
D, a correction value based on the rate of change of the error

Another thing to keep in mind specific to an engine when looking to modify a behaviour, slow and fast actuators. Spark = Fast. Throttle = Slow

This actually clarifies some things for me too. Thank You

Ed this goes back to some of what you've explained in other threads too and actually something I've agreed on. Spark as we know is fast action or built into your integral and throttle is there when the spark runs out of it's reserve. Now if you really want to F some stuff up you can make throttle the first and have a nice bouncy puke car....

Also the way I worded it as "torque" is probably wrong as well. I keep getting yelled at for mentioning that word It's just how I started visualizing the representation of what's going on. Helps ME to think of what it's doing. BUT yes, in it's simplest of forms it is adding and taking airflow to correct rpm or bring it back to it's "0 = desired idle rpm setting". Sorry for the confusion. Just how I visualize what it's doing.

[edcmat-l1]

Just jumping in to answer an earlier question...

Integral numbers are the increment / decrement per calc cycle.

In the (unusual) case where the error remains constant e.g -150 rpm and the PI control loop runs every 50ms, then every 50ms add the value from the -150rpm cell onto the previous integral airflow.
If this goes on for ever then the integral airflow will be huge (wind-up). So we limit how high it can go. And vice-versa.

P, a correction value proportional to the error
I, a correction value that if an error is present will increase or decrease over time
D, a correction value based on the rate of change of the error

Another thing to keep in mind specific to an engine when looking to modify a behaviour, slow and fast actuators. Spark = Fast. Throttle = Slow

Edit:
In your case with wanting to bleed the air off faster to close your throttle after a blip, I would first try and work out why the hell the thing thinks it's a good idea to preload the integral term with that much air in the first place.

Continue with your airflow / torque discussion gents

PIDs need to be considered in their most simplistic form. It is a system to smooth or modulate oscillations, in anything really. In this case here it's about blade control, period. In the GEN3 and GEN4 platforms, to me the PIDs are basically throttle body characterization. For the most part they don't need to be changed and you're unlikely to improve on anything by doing so.

When it comes to idle speed, not return to idle, not cold start idle, not anything else but at rest idle, I want to use the least amount of throttle control. Ideally I want to hold the blade still. If I need additional air on decel, dashpot, etc. or additional air when you turn the AC on, or when it's cold, etc. there are specific tables for each of those scenarios.

[GHuggins]

Edit:
In your case with wanting to bleed the air off faster to close your throttle after a blip, I would first try and work out why the hell the thing thinks it's a good idea to preload the integral term with that much air in the first place.

Usually in my experience with this particular issue it's either an airflow mismatch/miscalc or torque model in general. VE and MAF will both correlate to those along with the torque models, baro tables and min air. If tb is original then don't mess with the tb baro tables. Fix VE. Fix MAF. Fix min air if too high. Then torque model. Just my .02.

[LS ROB]

Just jumping in to answer an earlier question...

Integral numbers are the increment / decrement per calc cycle.

In the (unusual) case where the error remains constant e.g -150 rpm and the PI control loop runs every 50ms, then every 50ms add the value from the -150rpm cell onto the previous integral airflow.
If this goes on for ever then the integral airflow will be huge (wind-up). So we limit how high it can go. And vice-versa.

P, a correction value proportional to the error
I, a correction value that if an error is present will increase or decrease over time
D, a correction value based on the rate of change of the error

Another thing to keep in mind specific to an engine when looking to modify a behaviour, slow and fast actuators. Spark = Fast. Throttle = Slow

Edit:
In your case with wanting to bleed the air off faster to close your throttle after a blip, I would first try and work out why the hell the thing thinks it's a good idea to preload the integral term with that much air in the first place.

Continue with your airflow / torque discussion gents

With PIDs you will always have an over/undershoot correct? The objective is to minimize the deviation from setpoint? When adjusting a PID controller, lower or 0 out the Integral & Derivative if you have a Derivative, until you get as close to the setpoint as possible. Then bring in the Integral until oscillations minimize then do the same for the Derivative?

[hjtrbo]

There shouldn't be over/under shoot in a correctly tuned system. Tuning PID is complex and there is not one size fits all method. Proportional–integral–derivative controller - Wikipedia

I haven't tuned a e38 / e67 throttle PID as I've never needed to make adjustments to it. I must have got lucky. However, I have tuned numerous first order industrial control systems, so I know enough about it to comment. 2nd order and high dead time systems I leave for the legit control engineers.

[smokeshow]

2nd order and high dead time systems I leave for the legit control engineers.

Hey, that's what I do. I've had my hands on controllers that existed in the double-digit order range at full fidelity. That's nasty stuff

Gen4 is a bit of a bastard when you look at the throttle control strategy. While Gen5 is true coordinated torque control, the models in Gen4 exist only for speed control. That means very little use outside of idle. 6 speeds will utilize speed control for certain shifts like TUTD, but only on a few applications. Where Gen5 transitions from speed to torque control when leaving idle, Gen4 utilizes the throttle follower sort of like a fudge factor to bridge the gap between speed control and open loop throttle. But at idle...the base desired airflow is determined by the torque model. Corrections are done independently by air and spark, but that initial airflow target is all APC torque.

I definitely agree that the idle airflow PID tuning doesn't need a whole lot of tweaking to bring idle under control...but to say the throttle should be frozen at idle is leaving some stones unturned. Take a look at just about any return to idle maneuver and watch integral airflow discreetly step in to prevent stalling...it is doing more than you might realize.

[07GTS]

was there a conclusion with the early e38 that had the integral and adaptive transition, normal, coastdown tables ? was this mislabeled and the transition should be normal ? as the normal values from factory are all zero but just transition and coastdown are populated, i see later e38 on there is just normal, coastdown