How to: Startup Airflow and AC Airflow Table Tweaking

[jonofmac]

Hello All,
About me: I picked up a 01 C5Z with all bolt ons, heads, and cam (MS4) a while ago and had it tuned. There were some startup issues I was having and AC issues as well. Rather than driving all the way back to get it retuned (3 hour drive one way), I bought HP Tuners and decided to mess around with tuning for the LS platform. Now, I may be new to HP Tuners, but I'm quite experienced with tuning on the other side of the car world. I have a supercharged 09 Civic Si as a daily that I have been tuning myself for about 6 years now.

Abstract: I found that when I tried to learn HP Tuners, a lot of the tables were not intuitively named and it was kind of confusing what exactly the tables did. My AC compressor seems to be slowly dying and has an increased load on my engine, causing startup issues and idling issues. Me trying to save some $$ now, figured i could use this as a reason to jump in the tune and tweak things.
I did not find much information making about how to tune startup airflow or AC airflow, so I'll highlight some of the things I learned below to hopefully help any other newbies like me.


Startup Airflow:
Under Engine > Idle > Airflow, you'll see some tables under the "Startup" section on the left:
Screen Shot 2015-05-26 at 6.46.55 PM.png
Here are the tables/values and my description of what they do (at least as far as I observed):
Off Idle Scalar: This is used when your car has just started and is still using the startup airflow tables, and you do something to make it leave idle conditions (this could be pressing the gas pedal at all, or moving the car, etc. Anything that is not idling). The value ranges from 1 to 0 (and anything in between). A value of 0 means that if the car is using the idle airflow tables and something happens to exit idling, immediately stop using the tables (all airflow table values are immediately stopped from being referenced). A value of 1 means that even if something happens to exit idle conditions, continue using the startup airflow tables as normal. A value of 0.5 would mean to keep using the airflow tables, but use half of the value thats in the table. In my car, it is set to 1, and I decided to keep it that way, because I want the car to add the additional air shortly after a startup regardless of what i'm doing.

Underspeed Time: If startup overspeed flare has not occurred within this amount of time, then enable adaptive idle. Probably fine to just leave this alone, it doesn't particularly affect startup airflow or AC.

Friction Airflow Initial:I'm starting with this table, because it makes the most sense to. Friction airflow is additional air that gets added to your desired idle airflow during the first few seconds after startup only. Cars typically require a little extra air when they get started. This table describes how much extra air (per ECT temp) gets added. Friction airflow starts at this value, and then will stay here for a certain amount of time (defined in friction airflow delay) and then will decay to 0 at a rate defined by the friction airflow decay table.

Friction Airflow Delay: This tells the ECU how long to keep adding this extra air after startup (in cam revolutions) vs ECT. IE, if ECT is 198F and at that cell, it says 5. That means that the extra air will be held at a stable value from the frictional airflow initial table for 5 cam revolutions before it starts to decay to 0 g/sec. For a reference, a 1000 rpm idle has ~8 cam revs per second. so 16 cam revs would be about 2 seconds.

Friction Airflow Decay: This describes the ramp out rate of the friction airflow. Higher values correspond to the extra air ramping down to 0 in a shorter amount of time. Lower values correspond to a slower ramp down to 0.

Friction Airflow Graphic:
The below graphic illustrates an example of what the ECU does with the values from these tables. Keep in mind that normally the cam rev delay is so low that this entire process takes place in less than a second. (You can log this PID in Scanner)
FrictionAirflow.jpg




Startup Airflow Initial: Very similar to Friction Airflow above, except that this normally takes place over a longer period (IE ~10 seconds or less) with higher delay values (50 to 100 cam revs). This is intended for the slight increase in air required by a car when it first starts.

Startup Airflow Delay: Same thing as the Friction airflow delay, except that the values typically are larger to last for several seconds. For a reference, a 1000 rpm idle has ~8 cam revs per second. so 16 cam revs would be about 2 seconds.

Startup Airflow Decay: The decay rate of startup airflow once it hits the delay value. Higher values mean that the additional airflow decays to 0 faster, and lower values slow the process down.

Startup Airflow Graphic: Below illustrates the startup airflow process and where the ECU gets the values.
startupairflow.jpg

Things to note about Startup Airflow and Friction Airflow: They are additive! This means that the ECU adds them all together on top of the base idle airflow (RAF) table to get the desired airflow for startup. The friction airflow is supposed to be very short to compensate for the fact that no air is flowing, an there is extra friction in the intake manifold during the first second of startup. The startup airflow is designed for the slight increase in airflow needed by the engine during a startup before it will settle and require only the value in the RAF. So don't just duplicate values into both the startup and friction airflow tables, because you will essentially double what you put in one needlessly.

How to tune startup airflow: Normally, you don't need to touch Friction airflow because stock values should be ok and it's just a short transient. If you did the proper RAF tuning and find that when you start your car up, it still stumbles a little bit (commanding higher STITs when it first turns on and then the STITs fall to 0 somewhat quickly), this could be a result of not having enough startup airflow. In this case, I would open the Startup Airflow Initial table and increase the airflow values (check to make sure the units are what you are expecting by right clicking on a cell, going to units, and selecting the appropriate unit. I prefer g/sec. To make my car happy, it seems to like 2-2.25 g/sec throughout the ECT range. Tweak this value until you find a point where the car doesn't stumble any more. I would log the idle startup airflow PID and watch a startup process when the car stumbles. If the idle startup airflow starts to decay to 0 too soon, you can increase the delay values (remember that 8 cam revs = ~1 sec). If the airflow ramps down to 0 too fast and causes hesitation when this happens, you can decrease the value in Startup Airflow Decay to make this ramp out process slower.







AC Idle Airflow
Please see post #2 as I can't add anymore pictures to this post, I need to make another post.

[jonofmac]

AC Idle Airflow
Some of you might have your idle and startup idled in perfectly when the AC is off, but as soon as you turn it on, the car stumbles and might just stall if you try to start the car with the AC on.

I have a large cam and also think that my AC compressor might be starting to go out as the load it exerts on the engine seems to be increasing over time, so I find that I need to tell the ECU to add more air when the AC compressor is on.

Under Engine > Idle > Airflow, there is the "Adaptive Idle Airflow", in here you can set the limits for the AC offset minimums and maximums. For large cams, you might need to set the AC Offset Max/InGear to 4 g/sec (you can change the units by clicking on the unit next to the box). If your car is an automatic, you will likely need to make the same changes to the AC Offset Max/PN, but I have a manual, so my ECU always refers to to InGear tables. These are the only real changes that need to take place here.

The AC Compressor in most of these GM vehicles is not quite like other compressors. Instead of having a switch that cycles the AC compressor on and off, the AC compressor is always engaged and there is a valve that controls how much freon flows to control the pressure. So what this means is that the AC compressor has a varying load on the engine, based on how much pressure is in the line. There is a PID that you may monitor in the scanner to see the voltage of the pressure sensor the ECU uses to see how much pressure is in the system (see image below):
Screen Shot 2015-05-26 at 7.45.14 PM.png

The problem is, you can only view the voltage, and not the actual pressure. With some searching, I was able to find an excel document made by a user on this forum (my apologies but I do not have a link to the post as I have lost it , but if you know who it is, please message me and I'll give you your due credit!) which could calculate the voltage to pressure value.
Please see the image below, you will need it to know where you must add values in the tables I will talk about next:
Screen Shot 2015-05-26 at 7.50.06 PM.png

The ECU does not have a table that tells it how much g/sec of air to add when the AC is on. Instead, there is a table that says how much torque the AC compressor requires from the engine for the psi in the line. The ECU will read this torque value, and then use its own calculations in order to determine how much airflow the AC needs (which is the value you see in the AC Idle Airflow PID, and it gets added to the base RAF for the desired idle airflow).
The table we need to edit is in Engine > Torque Management > Engine > AC Torque.
Here you will see a lot of values, and I will explain them again.
Screen Shot 2015-05-26 at 7.53.50 PM.png

Let's go through them:
vs. AC Pressure: This is the main table we need to edit in order modify the amount of air given to the engine when the AC is turned on. It lists the amount of torque needed by the AC compressor for the range of pressures in it. Remember that the compressor used on a lot of these GM vehicles is ALWAYS engaged when you turn it on, and there is a valve that will change the amount of flow, which changes the pressure in the line. This pressure in the line affects how much torque the AC compressor needs to turn. Normally, you will see lower pressures while driving, and higher pressures when standing still because the condenser is not able to shed as much heat when you're standing still, so the ac pressures are higher.

vs. IAT: This is similar to the previous table, except that this is a table that tells the ECU so much EXTRA torque the AC compressor needs depending on the intake temperatures. Normally this can be left with stock values, but it is possible you might need to tweak these. I did not have to though.

Inertia Torque: This table is similar to the vs. AC Pressure table, except that this models the initial amount of torque used by the AC when the compressor clutch engages. The idea is that this initial torque surge will be higher because additional torque will be needed to bring the compressor from a stand still to moving at engine speed. I will explain later how this is used, but know that this is the first table that the ECU references in process of turning the AC compressor on, and then it switches to the vs. AC Pressure table.

Ramp: This table defines how fast the inertia torque values will decay. Higher values mean faster decay, lower values mean slower decay.

On Hold Time (Table): This table defines how long the ECU will refer to the Inertia Torque table before it will ramp out and switch to the vs. AC Pressure table. It is a value in seconds. Once the hold time has been hit, the ECU will ramp out the inertia torque value (at a rate defined in the Ramp table) and then use the vs. AC Pressure table.

On Hold Time (Single Value): I think this is a typo in HP Tuners, and should actually say Off Hold Time, but this is a delay for the ECU to hold the AC torque value after the ac compressor clutch has been commanded to disengage before it will fall to 0. What this is really for, is a delay to account for the fact that when the signal to the AC clutch to disengage happens, it will take a small amount of time for the clutch to actually let go and stop loading the engine. If this hold time is too short, the extra air for the AC will be taken away too fast and you might have hesitation or rpm dip when the AC compressor disengages. Too high of a value and you'll experience a spike in RPM.

Delay: How long to wait between commanding the AC clutch to engage and adding extra air to the engine. I have mine set at 0 seconds.

Ramp In: This value describes how fast the AC compressor can build up the torque requirement. IE: if the line pressure is high when you engage the AC compressor, this will slow down the air added to the engine, as the load on the engine should not be an instant increase. Stock values might work, but for my car, I found that I needed to increase it to 8-12 ft lb in order to make it ramp fast enough, otherwise I would get hesitation and RPM dips because the ECU was not adding the extra air fast enough.

Ramp Out: Same thing, how fast can the torque/air needed for AC ramp out. Stock value of 4 ft. lb. is probably fine, it was for me.

Duration Max: This defines the longest amount of time allowed for the ramp out process to happen. IE: if the Ramp Out value is set too small, and it takes longer than the time defined in Duration Max to hit 0, it will just cancel the ramp out, and immediately go to 0 g/sec of Idle AC Airflow.

ETC Max: Describes the maximum amount of additional throttle the AC Airflow can command.

On Delay: It is used for the spark retard tables used when the AC Compressor engages. I've 0'd out my spark retard tables because I don't want the engine to retard spark. So these values are kinda useless for me. Stock values would be fine though.

Off Delay: Same thing as above, except used when the AC Compressor is being turned on, this is another mechanism the ECU uses to try and prevent any RPM spike when you turn the AC off.

Max Retard: This table tells the ECU how much spark to remove when the AC compressor engages to prevent a dip/spike in RPM. I 0'd out all of the tables, as I did not want any spark retard.

Spark Off Delay: A delay for the spark ramp out when the AC Compressor disengages. Not used by me because I don't deal with the spark retard tables.



Understanding how the AC Compressor process works:
Ok, so let's break this down into several steps to try and help you understand what's going on (at least as far as I've noticed when I was messing around with the settings).
1) Car is idling with AC off, you command the AC to turn on.
2) At the moment the compressor engages, the ECU will wait for the amount of seconds in "Delay" and then it will ramp up to the "Inertia Torque" value, which will come out to some AC Airflow value.
3) Once at the defined "Inertia Torque" value, the ECU will wait for the amount of seconds defined in the "On Hold Time" table, and then it will ramp out at the "Ramp Out" rate. Once ramped out, the ECU will refer to the "vs. AC Pressure" table for the remainder of the time the AC Compressor is engaged. The load will change and the table will change the AC Airflow depending on the line pressure.



Tuning the AC Airflow
Fine and dandy, but how do you use this information to fix your car's idling with the AC on?
Pretty simple, your RAF should first be dialed in with the AC off to where STITs are idle are within +/- 0.5g/sec. Also it would help to turn off your LTIT by setting the Max InGear/ACOff and Min InGear/ACOff values to 0 (also the PN values if you're an automatic, remember my car is a manual, so the ECU always uses the InGear tables)
1) Make sure you're logging the AC Pressure and Idle AC Airflow PIDs in the scanner, and log the car with the AC off. Take note of the STIT value.
2) Turn the AC on, you'll see the AC Airflow PID increase and change. Give it a few seconds and watch your STIT values. Depending on if they are positive or negative, you will need to add torque or remove torque from the vs. AC Pressure table at the corresponding pressure (and likely those around it).
3) In my case, my STITs went WAYYY positive to almost 2 g/sec! I also saw that the pressure (when I compared the voltage to the pressure table above) was ranging between 150 psi to 250 psi. So I went to the vs. AC Pressure table (Engine > Torque Management > Engine > AC Torque) and increased the torque values. This took several attempts to get it dialed in, but I changed my table values in 10% increments, and made sure I made the table look about the same as it was before. So if increased the 250 psi value, I also increased all the values above 250 psi the same amount since the load increases with psi.
You would do the same thing, but remove torque if the STITs were negative when you turned the AC on. Of course, pay attention to STITs. If you are at -0.5g/sec with AC off, and go to -0.25g/sec with AC on, this means that you should still add a little bit of torque, but your RAF tables really need to be dialed in correctly since you want to get the STITs as close to 0 as possible.
4) You will likely have to change your inertial torque table values as well if you end up making large changes. So if you add torque to the vs. AC Pressure table, you might have to add a little torque to the inertia table to make it proportionate.

Here is my vs AC Pressure table, but I would not recommend you start at values as high as mine. My compressor is really loading my engine, and I'll need to replace it.
Screen Shot 2015-05-26 at 8.39.10 PM.png


There are other scenarios you might run into, which requires changing other tables. For example, with my AC torque values this large, I had to really increase my inertia torque values to get the car to start with the AC on, since the ECU will reference the inertia torque tables first, which means it will use this table during startup.

So in short: If starting with AC on causes stumbling or if engaging the AC when car is running causes stumbling, you might need to change the inertia torque values depending on your STITs and make sure that the inertia torque values are a little bit higher than their corresponding vs. AC Pressure values .

If idling with the AC on has a large STIT value when the AC off has a very close to 0 AC value, you need to tweak the vs AC pressure to dial in the amount of extra air needed by the AC.

[elementaltoad]

Awesome write up and answers some questions I had about why my hot start seems to be slower compared to a cold start. I'll have to give this a try, thanks!

[smokestack]

Thank you.

[mowton]

OK and so where does afterstart enrichment play in all this....3 separate startup tables all doing about the same thing? All controlled via Engine Temp with afterstart enrichment including an IAT variable.....

Ed M

[jonofmac]

That's a good question. I don't know as I haven't played around with those tables at all to see what it does. If you know or play around with the tables and learn how they come into play, please share. I am no expert by any stretch, but I do like to learn and try to share what I learned.

[RonSSnova]

Well, after start enrichment is a fuel adder.

[RonSSnova]

BTW, thanks for sharing this knowledge! Going straight into my tuning bookmarks.

Ron

[JakeFusion]

I had some fuel delivery issues that Afterstart Enrichment fixed... But then I fixed the fuel issues and put everything AE back to stock. IMO, it's one of those tables that shouldn't be messed with. If your injector data is good, you'll get proper fueling once you get airflow dialed in for startup.

And this is good stuff. I just multiplied my A/C Torque table by 1.8 and that seemed to fix my issues.

[mowton]

Well, after start enrichment is a fuel adder.

Wow, what a senior moment I just had.....thanks Ron

Ed M

[jonofmac]

Haha. Glad to help out. The guide i read on RAF tuning helped me a lot, and this part of the tune confused me so when I finally figured it out, figured I'd share.

We need people to make helpful posts lol. It helps make us a bit more independant. Plus I like this stuff

[mowton]

Haha. Glad to help out. The guide i read on RAF tuning helped me a lot, and this part of the tune confused me so when I finally figured it out, figured I'd share.

We need people to make helpful posts lol. It helps make us a bit more independant. Plus I like this stuff

And props to you for a great writeup....s/b a sticky

Ed M

[RonSSnova]

Right behind you Ed! Ha!
You help us all a lot too. It is much appreciated.

[Ghostnotes]

Yeah my idle issues are kicking my butt now. It will cold start fine and run.After about 5 minutes it starts to labor a little but will still idle. If i kill it then try to restart, it immediately dies unless I hold the throttle open.

[jonofmac]

I'm not sure but that sounds more like your RAF table needs tuning. Are your STITs high when it warms up?

Yeah my idle issues are kicking my butt now. It will cold start fine and run.After about 5 minutes it starts to labor a little but will still idle. If i kill it then try to restart, it immediately dies unless I hold the throttle open.

[RSTinOKC]

Awesome guide man! I've been shooting in the dark for weeks now trying to tune A/C airflow.

[Jason W]

I know this post is old now, but it's where I am currently at with my tune. After building a 402 stroker for my WH2, I got it running, bumped up the timing and the PE tables and left it there for ages! It was fun, but not the most efficient and pretty terrible at idle!
I have now pretty much dialed in my VE and idle RAF tables and have flattened out timing at around 25 degrees across the board to sort out other issues, but now we are getting into the mid 30 degree days my warm start idle is starting to surge real bad.
I am going to start playing with my after start airflow and friction airflow tomorrow to try and sort this out.
I can't wait to get all these basics sorted and to kick my timing back up and enabling PE again cause I am missing the pull when sinking the boot

Thanks so much for an awesome write up. It's always handy when people put into words what they have found when playing. I just wish I had some knowledge to give back to everyone I have read from in the past!

Cheers,

Jason.

[jonofmac]

Late reply, but I'm glad you liked the post. The HPTuners forum has helped me immensely throughout the process of learning how GM PCMs work (I've got experience mostly with Honda based ECUs, and a lot of these finer functions are hidden from the user). When ever I get the chance to give back to this community, I'll do what I can.

For what it's worth, I managed to nuke my engine racing it (lifter gave out) and only just yesterday got to the point where everything is back in the car (I custom built a 416 CI LS3 to replace my LS6), so I've been having to tune it on the street to get drivability and everything dialed in so I can break the trans and diff in gently before it gets a dyno tune for WOT. Looking back at this post reminded me of a few things i forgot, since it's been 1.5 years since I've touched HP tuners!

[Ricardo]

Gret thread.
Trying to solve some cold start issues right now. Will give it a try.

[kyotey1693]

I know this is a pretty old thread but it seems to focus on what I think is causing me the most issues. Jonofmac, if you're still around, thanks for the insight on AC tuning. I've got a 99 WS6 h/c/i car that's pretty well dialed in with the exception of having the AC on. It's good most of the time, but occasionally it will dip and die when coming to a stop.

I've played with just about everything covered in here and I'm still scratching my head. I've been fighting the common issues of idle hangs or idle dips with AC on or off. I've increased and decreased the AC torque numbers and at idle I can get STIT near perfect but once I start moving and then try to stop again it all changes. One thing I have realized is the AC pressure is obviously changing so that requires you to make adjustments in other areas of the AC torque settings, but it's never consistent.

I'm mostly focused on the low speed areas below 40mph right now. I though throttle follower or throttle cracker might be causing me some issues. So those are zero'd out. The surprising thing there is when I log them, they're still showing activity. Can anyone explain this? It seems like there must be another table(that I must not be aware of) controlling IAC once speed is involved. If the AC is on the car doesn't want to return to idle until I get below 4mph. In the log, TF is showing activity even though it's zero'd out below 13% TPS. What's causing that?

Since 4mph seemed to be an enabling/disabling point, I zero'd TC assuming since it has MPH built into that it must be causing the idle hang. No change. Interestingly too, it really doesn't seem to matter what I put in for TC, I can't tell that it has much bearing on anything.

My last thing to try was zero'ing out the AC torque numbers for kPa and IAT. At first I only drove a short distance in a parking lot and came to a stop, not very much MPH, and it didn't seem to want to idle down. So I headed on back to the house and it drove pretty much perfect all the way home. And I really have no idea why. Anyone care to shed some light?

I've read countless threads on TF and TC. It seems like there is some voodoo going on with them on the older 6 speed cars. Since mine is a 99 I don't have the rolling idle setting and clutch transition seem to enable/disable a whole different set of parameters that I don't access to. I have attached the current tune and a recent short log. I also made a spread sheet that expands on the one above that makes it easier to reference
PSI-kPa-Voltage.

99 30thAnni- Full Tune 3.9.9(RGT-REBOOT)No TF,TC,AC Touque.hpt

REBOOT No TF,TC,AC Touque.hpl

LS1 F-Body AC PSI-kPa-Volts Chart.xlsx