When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
just dialed my truck in.. LQ9 42#@43.5psi, 224/230 113 .560, 92mm, NNBS intake, valvetrain, exhaust, modded intake.
overshoot/undershoot like a b***h.. wouldn't drive right, couldn't get it running well enough to save my life. THEN, I started thinking of ways that I could tune the IAC effective area.. MAN.. Now my Dynamic air (Calculated from VE) and MAF (measured air) And DESIRED (Calculated from IAC Area and required air tables) all match up
How about posting the info as text? I can't open the attachment....
If people in the forum are having problems opening the text file, I'll gladly post it in forum. It was just too long to post so I'll have to break it down.. (I tried once already). Lol
This was a very helpful write up. To be honest, I already had a pretty stable idle, but it was bugging me that the commanded air was always 2-g/sec lower than dynamic or MAF air.
I did a simpler version of your method, which for a quick and dirty at least got the commanded and MAF reading pretty close:
1. Log IAC steps vs MAF airflow in metric units
2. Log IAC steps vs Dynamic Airflow in metric units
3. Log MAF or Dynamic airflow vs commanded air flow in metric units
4. Cold start, so you get a wide range of IAC steps. No throttle.
5. Take the range of IAC steps (for example lowest was "4" and highest was "150") and the range of air flow from logs 1 and 2 (example, low is 11g/sec, high is 20 g/sec)
6. Divide (High IAC-Low IAC)/(max air - min air) = 146/9 (in this example) = 16.2. Think of this as "IAC counts per g/sec airflow"
7. Take log 3 and calculate your average airflow error - example commanded air is 2-g/sec lower than MAF air
8. for a given effective area, the IAC counts in this case need to come down 2 (g/sec error) * 16 (IAC counts per g/sec airflow) = 32 IAC counts
9. Subtract this number (example 32) from the entire IAC table.
Fire it up and log some more.
Granted this is nowhere near as accurate as the OP's method, but it did put commanded air more in line with MAF airflow pretty easily.
Credit to the OP for figuring out how to do this at all. Very good detective work.
Its still pretty close to the same method.. And while I'm learning more about GM's algorithms, don't be dismayed about being off on your desired air flow AFTER its tuned..
What I'm learning is desired air will change from day to day depending on temp/humidity. So you really wanna knock out desired air flow calc AND IAC area in the same day hopefully as soon as possible.. Reason being, if you log your required air, you know that's "0" for right now. Get your IAC steps dialed in RIGHT NOW so you can dial them in to "0".. This establishes an air flow characteristic for your iac/throttle position setup.. Now, if "desired" air flow changes tomorrow or next week, stit/ltit come into play and fix the offset in desired air flow. While your iac still has the SAME flow curve since you never changed anything, and physics didn't change over night either. LOL
So key to this tutorial/writeup is to do desired air flow AND IAC effective area at the same time to get accurate results. Then don't worry when desired air is 9g/sec and maf/dynamic is 12g/sec, because more than likely your ltit is making up for it with -3g/sec. :-)
Very interesting. I'd still like to see the full document. DYNAIR (calculated air flow rate) works if the engine is MAF-less, right?
If your VE tables are dialed and are accurate, yes, this will work using dynair. What I've noticed is dynair will always slightly trail maf air, as dynair is an rpm dependant calculation. For example when the IAC opens to raise idle from engine load or an rpm drop. The maf will measure the additional air flow before the rpm catches up in the calculation. However, it is for the most part a mirror image of what the maf would read.
Touching on the subject of idle tuning related to IAC area/desired/required tuning.. I've been learning more and more and I want to share it with you so your car can idle awesome too..
Since desired changes from day to day, this is what learned airflow correction is for. Since the IAC area should now be tuned, and the desired is calibrated, the desired should be a rather fixed correction upon startup, but does change a little with weather etc.
The idea is to set the learning trim in such a manner that the learning air flow rpm is set within reasonable bounds of a proper idle. My cam varies in idle about 50-60 rpm +/- target. So I set my learning rpm threshold to 60 (B4508)
Next, you want to set your learning timer to a long enough delay that there's DEFINITELY a target rpm error. Every once in a while my can will deviate idle +/-70 rpm but its so brief that I don't mind and it doesn't affect idle quality. However if it were to STAY +/- 70 rpm from target, that would NOT be acxeptable .. My idle learn timer is set to 4.3 seconds (B4507). This means that if the rpm is below my set 60rpm deviance from target for more than 4.3 seconds, adjust my idle air requirement (trim desired air a little to bring it in spec for proper idle) you will have to play with time and rpm to see what works best. I'd recommend using an rpm error that's outside of what sounds/feels like an acceptable deviance in rpm that does not effect idle.
Next, learned airflow correction (B4514). For the most part, I used stock calibration for this and added a small percent of correction since big cams tend to need more air than a smaller cam to get the same effect. I also zeroed out 0-40 rpm correction factors. I do not want it to make any adjustments to idle air while its within my ideal +/- 50rpm target. Correction for 40-100rpm error for low rpm is higher than the same rpms for high rpm error. All other corrections match values. The idea here is that its favorable to have too high of an idle than too low of one. So if there's an error where the rpm is too low, I want it to add more correction than it takes away from being high. This so far has produced a VERY stable idle with minimal hunting or searching for target idle if any at all.
Hope this helps.
Desired air compensation by short term and long term idle trims. This can is kinda choppy, but on average, the correction is accurate.
Learn up faster than you learn down, just don't get too aggressive.
So tonight I've been driving around dialing my tune in.. I've managed to eliminate most all of the surging and bucking between 25mph-dead stop.. The steps up to now turned an in bearable tune into something much more manageable with some faults, but most of the big stuff is fixed.
Tonight I was looking at logs and noticed my IAC would open to compensate for below target idle, then get slightly over target and take off and then overshoot back down past target all while 0%tps at 15ish mph in 2nd gear. I also noticed my timing jumping up after IAC raises rpm to about +40 of target. I went into BASE IDLE TIMING tables and followed the cell being read by scrolling the data log. Found that the overshoot was causing some timing from the 1200 rpm cells to be added, making my overshoot way worse than it needed to be. All cells up to 1200 rpm and 1 column just outside the cell that was being read at idle all got set to 20* base timing and smoothed the transition between cells while not touching values in cells I just set.. This helped A LOT.
Now I still have hunting, but its super manageable. It doesn't make the truck want to run off. Playing with rpm set point adjustments, and rpm set point timer, I got some more if not almost all of the bucking to quit and the surging too! My setpoint adjustment is set to 8 (if someone can chime in and explain exactly what this number represents and how it works, I'd be greatly appreciative) and noticed a bunch of big cam cars have more aggressive set point adjustments as well.. That, combined with a set point time of 0.250 sec worked really good for both park/neutral and in gear. (I'm automatic) I know the set point adjustment has something to do with the adjustment to idle speed, whether it be 8 adjustments per .250, or 8x(adjustment amount) per .250 sec, I don't know. What I do know, is once I set those 2 numbers in the calibration, I had almost NO undershoot and overshoot was so mild that I can't complain about it. If anything, I should probably adjust my direct air flow adjustment.. Its very possible that I'm adding too much air per rpm of error which is causing the super mild overshoot. I did add 20% to throttle cracker between 0-40mph and 400-1200 rpm between tunes as well. I increased cracker gradually until I had a slight cruise control effect and took 20% out of the same cells I was adding to and blended the map. This smoothed the idle transition from 25-0mph a lot as well. I'd say this part of the tune is now 98% of what you'd expect a factory stock tune to feel like.
As of right now, I have a STOCK converter/4L60E behind a LQ9, 224/230 113* cam with 46# injectors 3.42 gears with a 28" tire on a 4000# truck.. It comes to idle perfect, no cruise control, almost no surging (no surging that you can feel. Its so minor that its only audible, maybe +/- about 20rpm while decel.. Could be from o2 switching?), starts up instantly, and smoothly comes to an idle with no drama or overshoot/undershoot. Idle rpm is stable, long term and short term idle trims are solid and don't barely move if at all. In gear to neutral and back into gear has no flare at all, and no drop in rpm. I'm having a LOT more confidence now that this is almost done! So very excited!
I'm curious about how much of this is truly about IAC
effective area, and how much is laying related-ish but
not specifically effective area "stuff" onto the IAC area
table out of convenience or no better idea.
I've played with this some, chasing similar things, and
found that for me it wanted the IAC channel that I had
epoxied and drilled, drilled bigger. I scanned as you did,
airflow vs steps and plotted that over the stock curve
and saw it was "laying over". Rather than model this
new state, I drilled the channel out to 3/8" (from the
bad.net.advice of 3/16") and everything trued right up
and the surging (which was IAC winding up, trying to
get the airflow through that soda straw) went away.
That everything lined up again, makes me think the
stock model of the stock IAC is pretty right. But I see
more than a few folks having problems with aftermarket
TBs, and maybe there's the discrepancy.
Everything relating to idle is based on the foundation of IAC effective area and desired air flow.
The IAC effective area is a map of how many steps it takes the IAC to change the effective flow area for idle air, and how much impact it has on that area.
For example. At target idle, if the butterfly valve in the throttle body flows 5g/sec and the IAC at 40 steps flows 5g/sec, the IAC has a significant contribution to idle air flow, and therefore a large impact with minor movements.
You crack your throttle body open a little more, same target idle. The butterfly flows 7.5g/sec but now your IAC is at 20 steps and flows 2.5g/sec. The total idle effective area is the same, however the IAC is contributing much less per step, and therefore the IAC effective area is now changed.
When the pcm knows the IAC effective area to some degree of accuracy, it cuts down on "finding idle" in a very big way. Rather than the computer looking for 40 steps because that's what the iac EFF area map says is x g/sec and is wrong, then it has to use direct idle air correction to actually locate where idle physically is in g/sec of air flow. After it establishes idle and you romp on the throttle again, it hunts all over for the correct airflow in the IAC area chart and uses direct airflow to compensate. This is what causes a lot of problems. If you're running a stock TB and haven't adjusted the stop screw, and its cleaned, you should never have to adjust IAC area. However, if you touch that screw, bore the IAC port, port the TB, or modify it in any way that changes airflow characteristics, you need to recalibrate the effective area.
Its not hard, just do it. LOL
BTW, my tuned IAC effective area map looks NOTHING like the stock area map.
I'm curious about how much of this is truly about IAC
effective area, and how much is laying related-ish but
not specifically effective area "stuff" onto the IAC area
table out of convenience or no better idea.
I've played with this some, chasing similar things, and
found that for me it wanted the IAC channel that I had
epoxied and drilled, drilled bigger. I scanned as you did,
airflow vs steps and plotted that over the stock curve
and saw it was "laying over". Rather than model this
new state, I drilled the channel out to 3/8" (from the
bad.net.advice of 3/16") and everything trued right up
and the surging (which was IAC winding up, trying to
get the airflow through that soda straw) went away.
That everything lined up again, makes me think the
stock model of the stock IAC is pretty right. But I see
more than a few folks having problems with aftermarket
TBs, and maybe there's the discrepancy.
I think in general you are correct. The one exception might be the Nick Williams throttle body. I checked the IAC port, and I fit a 3/8 drill bit all the way in with no resistance. It might be 7/16" to be honest, as easily as the 3/8 fit.
It's funny how sometimes we find out the GM knew exactly what they were doing...
01-17-2016, 07:42 AM
IAC Effective area tuning.. ITS A MUST 
