AFR drift vs IAT

[cc-rider]

Tonight I was searching the web for information on IAT effects on dialing in VE & MAF. I was wondering if I could use the cold Minnesota spring air (~50ish F) to to tune in my MAF/VE via AFR error datalogs...or if I should wait until temps reached a more average level in which I'll drive the car (~75ish F).

I apologize if I'm breaking a rule and posting something from another forum...but I found this write up and wanted to share it: http://download.efilive.com/Tutorial...20Tutorial.pdf
In particular read the section near the bottom about AFR drift vs IAT and how to correct via the Flow rate vs IAT multiple table.

I haven't seen any mention here on HPT forum of making IAT injector drift part of the standard process of VE or MAF tuning. But it makes perfect sense after reading this article. Unless I'm missing something....this seems like something that could really help dial in VE/MAF fueling consistency across the wide range of IAT's we see (especially forced induction builds). I wonder how many people have chased their tails trying to dial in VE while unknowingly fighting this in the background?

Anyway....I wanted to share and get some feedback.

[schpenxel]

I did not use that full method but basically did the same thing after reading that tutorial. I've seen factory tunes on some newer cars make use of the IFR vs. IAT mult table too.

I see huge temp swings here.. in the winter as low as teens and in the summer as high as 90's. It would go lean in the summer and rich in the winter and I got tired of constantly trying to adjust for temp changes.

-Making the multiplier less than 1 makes it think the injectors are smaller, meaning it opens them longer-->more fuel/richer
-Making the multiplier greater than 1 makes it think the injectors are larger, meaning it opens them less-->less fuel/leaner

[cc-rider]

-Making the multiplier less than 1 makes it think the injectors are smaller, meaning it opens them longer-->more fuel/richer
-Making the multiplier greater than 1 makes it think the injectors are larger, meaning it opens them less-->less fuel/leaner

Thanks for sharing. I would have thought it to be the opposite.

What is the general drift that you have seen in both factory tunes and other testing you've done on this? Is there some place to start that gets it close (like x% for every 10 degrees)? Or is this something that just needs to be logged because it's vastly different for each car?

[schpenxel]

Yeah it's kind of backwards from what you'd initially think.. have to remember it's changing how big the computer thinks the injectors are, NOT the pulse width itself like fuel trims would do

I haven't spent a ton of time on it but I basically made the 86* cell the starting point and left it 1.0. For 68* I used 1.01, 50* 1.02, 32* 1.03. Other way.. 104 I used 0.98, 122 0.96, 130 0.94, then 0.94 the rest of the way up

It still goes richer in colder weather but I just wanted to get rid of some of the differences I was seeing and this seemed to help

Not sure how much it varies car to car. I've always thought when using the MAF the differences due to temp changes would be much smaller but I haven't tested it enough to find out.. takes too long to test the same car on cold and hot days unless it's your own

[cc-rider]

Yeah it's kind of backwards from what you'd initially think.. have to remember it's changing how big the computer thinks the injectors are, NOT the pulse width itself like fuel trims would do

I haven't spent a ton of time on it but I basically made the 86* cell the starting point and left it 1.0. For 68* I used 1.01, 50* 1.02, 32* 1.03. Other way.. 104 I used 0.98, 122 0.96, 130 0.94, then 0.94 the rest of the way up

It still goes richer in colder weather but I just wanted to get rid of some of the differences I was seeing and this seemed to help

Not sure how much it varies car to car. I've always thought when using the MAF the differences due to temp changes would be much smaller but I haven't tested it enough to find out.. takes too long to test the same car on cold and hot days unless it's your own

Cool. Thanks again for sharing your experiences here. I think I am going to give the test method shown in that article a go. Seems easy enough. Just let it sit and idle and heat soak. Which is no problem for me at all given my Kenne Bell. Let it sit and idle and the temps will most certainly climb.

I tend to average about 30-35 degrees above ambient. So my base set point (where I'll have a 1 in the mult setting) will probably be somewhere around 110 degrees IAT.

I'll give it a shot and post some data.

[schpenxel]

Sounds good. Yeah just letting it sit and go from cold to hot/heat soaked is probably the data you'll get, I just never got around to spending the 20 minutes it would take to do it for some reason.

Let us know how it goes.

[realcanuk]

I hope its not a bad thing to post this... but page 7 of this has a good write up on tuning the iat compensation.


http://download.efilive.com/Tutorial...20Tutorial.pdf

[cc-rider]

I hope its not a bad thing to post this... but page 7 of this has a good write up on tuning the iat compensation.


http://download.efilive.com/Tutorial...20Tutorial.pdf

Yep. I found that as well (and posted it in my first post on this thread). Good write up.

[realcanuk]

Yep. I found that as well (and posted it in my first post on this thread). Good write up.

Oh, sorry didn't notice... I used that right up to do it a few years ago.. never found anything better on the topic really.

[cc-rider]

Reading through that tutorial again...all they are doing there is calculating AFR error vs iat for a steady state idle reading. So I created a basic histogram to do this. AFR error vs IAT. I should be able to record this in open loop at idling....let the IAT's climb up through the range....and then I should have the data to plug into the injector mult vs iat table.

IAT drift histogram.jpg

[mourat]

I find the table doesn't do what you want if you wish to tune both SD and MAF because the fueling effects from IAT temp vary considerably from SD to MAF so if you're tuning for MAF mode and then switch to SD you'll be way off.

I did find that in the Custom OS HPT offers there is a VE vs temp table which I adjust for VE and that works wonders. You'd have to play with both right if you're running both VE+MAF however if you're running MAF only then use Injector vs IAT table.

The other issue I found is that OUTSIDE temp makes a difference and we can't tune for that. Here in Vancouver if we go up Cypress where elevation isn't that far off and MAP barely changes but the outside temp drops by a good 8* outside, I will see 5 LTFT split just from being in the cold compared to on the ground where it's warmer even though the IAT temps are the same that tells me that the air is more dense even though the temp is about the same by the time it gets into the manifold.

[schpenxel]

Is that VE vs. temp table on an E38? I don't see it on my E40

[Lakegoat]

It is under airflow/dynamic/ve correction factor

[schpenxel]

I see "VE Initial Correction Factor". Is that it?

The notes for that table basically say it's the ratio of MAF airmass to VE airmass?

[realcanuk]

I see "VE Initial Correction Factor". Is that it?

The notes for that table basically say it's the ratio of MAF airmass to VE airmass?

I have the same as you on my E38.

[cc-rider]

I think what Mourat is talking about is not the VE Initial Correction factor. I think the table he is mentioning is under Airflow-General....under the Main VE section...Charge Temp Mult table. It is right under the 3 main VE tables. This allows you to fudge the VE based on IAT (at least I assume IAT is the "Charge Temp" they are referring to). What is nice is that the temp axis is configurable to whatever you want it to be. Nice for superchargers that run higher IAT.

I think this table is only available on a custom 2 bar OS. So if youre looking at a stock OS...it won't be there.

[mourat]

Exactly like CC-Rider said has to be custom OS which are free these days. All you need is to convert Kelvin to Celsius. Like I said though, outside temps make a difference so it's not just what the IAT sees but the density of the air outside which I guess is also why GM went with MAF as it's less susceptible to air density changes I could be wrong but it's what I'm seeing consistently which I wish I did not.

[cc-rider]

Correct me if I'm wrong...but air density (really oxygen density is what we are getting at) is largely a function of barometric pressure and temperature. Both of those are accounted for in the tune (baro & IAT). Though IAT can be affected by so many more things than just the incoming air (such as a heat soaked supercharger). So fundamentally speaking...the ECU should be able to adjust for air density changes. That being said, I'm sure that the calculations and functions are probably hard to get perfect as so many different atmospeheric conditions exist.

In some ways, I wish the ECU used the ambient temp somehow (I am told it doesn't on the E38). That could help.

[schpenxel]

Yeah, it should be taking into account temperature when it calculates how much oxygen is in a certain volume of air.. Why it doesn't work quite perfect, I have no idea... above my pay grade I guess.

It's not like it doesn't do it at all.. cool the IAT down from meth or whatever and even 10-15 seconds after the meth is off if the IAT is still reading low then it will run rich. It think its colder/more oxygen needs more fuel, but it doesn't, so it goes rich.

Never have figured out why it changes in summer vs. winter

[cc-rider]

That must be the computer's way of adjusting for outside air density changes. That's why when you artificially cool down the intake air charge (via meth or intercoolers) it will make it run a bit rich. You aren't making any more oxygen in reality, but you just told the computer you did. Perhaps you can lower the temp adjustments in your tune to be less additive and it won't go as rich.

A lot of people make the mistake of thinking because their intercooler cools the air charge down that there is now more oxygen to make more power. That is simply a widespread misconception out there. More oxygen density can only come from the atmosphere, and oxygen cannot be "manufactured" in the intake system by cooling. But that's not to downplay the wonderful effects of good intercooling. A car can't make more power with a cooler charge no doubt. But it doesn't come from more oxygen. It comes from timing and even AFR.

Nitrous and meth, on the other hand, are a whole different ballgame. Can add combinations of cooling, oxygen, and octane to the charge.