PDA

View Full Version : AFR Commanded/ Actual



zapp168
July 31st, 2007, 09:37 AM
Could someone shed some light on this for me? Why or what is commanded AFR if the AFR is mainly controled by the VE table? What does changing this value do? Does this value matter in a OLSD situation? and since I am :blahblah: you all to death, what is the best AFR to try to maintain? I know 14.7 is the best burn, but I remember reading that the LS1 liked low 13's for a ratio, is that just cruising around? or under load? The more I learn....the more I need to know....:rolleyes:

SSpdDmon
July 31st, 2007, 10:45 AM
Commanded AFR is your target AFR. Pretty straight forward. In open loop, it's determined by the Commanded Fuel in Open Loop table, the PE table, and any other adders that may become active. In closed loop, commanded AFR is the value represented in the stoichiometric AFR B3601 cell and again, the PE table plus any adders that may become active.

As for the rest of your questoins, the MAF and VE tables are responsible (along with the input of a few other sensors) for trying to calculate the incoming airflow. For steady state throttle and RPM's greater than the threshold of B0120, the MAF reads the incoming airflow represented by a frequency (Hz). The PCM then converts that frequency into an airflow number. For transitional throttle and speed density applications, the VE is used to calculate the airflow based on readings from the MAP, ECT, and IAT sensors along with engine RPM and some other known constants.

Once the PCM knows airflow and the desired (aka commanded) AFR, it can then determine the right injector pulse width based on what it knows about how the injectors flow.

When tuning, we reverse engineer our VE & MAF tables by reading the actual AFR and comparing it to the commanded. That's it in a nutshell. :)

Rules of thumb: 14.7:1 is a good AFR for cruising, 12.5~13.0:1 is a good AFR for N/A @ WOT, 11.6~12.1:1 is a good AFR for boosted/N2O @ WOT, and 14.7~15.5:1 is a good AFR for idle and light load cruise. Again, these are general rules of thumb...

zapp168
July 31st, 2007, 11:40 AM
Thanks again :) Soooo.... when if ever is it good to change the commanded AFR?

SSpdDmon
July 31st, 2007, 01:19 PM
When you're tuning WOT or when you're running open loop. The stock formation in the PCM tries to maintain stoich until PE becomes active (based on PE Enable table referencing TPS% position). The factory PE table is rather rich and doesn't follow the rule of thumb I mentioned. So, when you're tuning you'll have to lean out the PE.

If you're tuning the VE in open loop, you have to adjust the commanded fuel in open loop table so that it keeps you at stoich for your operating temperature range. Just make sure you don't change PE to keep it from activating. There's some info floating around recommending that....and it's highly recommended here that you don't do that. :)

joecar
July 31st, 2007, 04:38 PM
The VE table determines airmass, AFR is not determined nor controlled by the VE table.

B3605, B3618, B3647 set the AFR.

Changing these tables is meaningful if the VE table is dialed in (which means you reverse engineered it so that actual AFR equals commanded AFR).

You change those 3 tables (and ignition timing) to get best torque/power thru-out the RPM range (without knocking)...

e.g. people have been setting 12.5 at peak torque, and then lightly lean out to 13.0 at peak power, as Jeff said.

zapp168
August 3rd, 2007, 01:34 AM
The VE table determines airmass, AFR is not determined nor controlled by the VE table.

I have been adjusting the VE table, and it seems to be changing my AFR? Should I be changing the commanded AFR then trying to make the VE table match, meaning show all 1.00 logging BEN. I am running CO3 with SD there is a boost VE table, but no commanded AFR in boost? I was adjusting the boost VE table that seems to be changing AFR? :nixweiss: LMAO two steps forward three steps back......

SSpdDmon
August 3rd, 2007, 01:49 AM
You set commanded fuel and adjust VE when in SD. You're reverse engineering the VE table based on the % error between commanded and actual AFR's. Commanded fuel for the boost table should be handle by PE. If not, it should take the last row of the commanded fuel in open loop table......although, I'm not 100% familiar with the COS's. But, it should be similar. I really have to get some hands on time with those. :Eyecrazy:

zapp168
August 3rd, 2007, 02:14 AM
LMAO....by the time I know what I am doing you will be better at my OS than I am..I have to go look into this a little deeper and really look at the tables Joe mentioned. I think I am starting grasp the reverse engineering thing though. I am more "correcting" the VE table, not adjusting the AFR.

gto_in_nc
August 3rd, 2007, 02:31 AM
Exactly! You are correcting the VE table so that the PCM knows how much the engine really pulls in, versus the "safe" factory settings. This allows actual AFR to be close to commanded AFR in OL. This is kinda like getting the foundation level when building a new house...

zapp168
August 3rd, 2007, 07:06 AM
So if I log and see that BEN is mostly 1.00 and I also notice the car is also running lean I would then need to adjust commanded AFR? Yes? and sorry, I am at work right now....still need to look at the tables more so I have a visual. New laptop up and running today though :)

joecar
August 3rd, 2007, 08:02 AM
So if I log and see that BEN is mostly 1.00 and I also notice the car is also running lean I would then need to adjust commanded AFR? Yes? and sorry, I am at work right now....still need to look at the tables more so I have a visual. New laptop up and running today though :)Yes, that's the idea. How are you noticing it's running lean...? :)

Basically, in OL:
1. first, the PCM looks up VE table to compute [using MAP, IAT] the cylinder airmass,
2. then it looks up the OLFA/PE tables to get the desired (commanded) AFR,
3. then it computes how much fuelmass is required such that airmass:fuelmass equals AFR,
4. then it looks up the IFR table to get the injector flowrate,
5. then it computes injector pulsewidth to spray the fuelmass;

the commanded AFR is determined in step 2;
the measured (actual) AFR is influenced by steps 1, 2, 4...

so if the VE and IFR tables are "correct" then the actual AFR will equal the commanded AFR...

the IFR table can be computed (by you) from the fuel pressure and the injector rated flowrate; so what's left is the VE table... you "correct" the VE table by multiplying each cell by the % error (BEN) between the commanded AFR and the actual AFR for that cell; after doing this, the actual AFR's will equal the AFR's set by the OLFA/PE tables (i.e. BEN will be 1.00 for each cell covered)...

this makes it easy to set your AFR for best torque and best power as you sweep up the rpm range
(you set the AFR and that's what it will actually be, no need to do trial and error to achieve a particular AFR)...

so it becomes important how you're measuring actual AFR.

:)

Redline Motorsports
August 3rd, 2007, 08:17 AM
The whole pupose of the VE table adjustment is to properly "model" the airflow requirements of the engine. The reference to "reverse engineering" is because we don't know how GM came up with the estimated airflow values in the VE table. What we do know is if we command a given AFR and the actual is different, the error is in the VE table cell at that time.

Example;

commanded AFR is 14.7
actual AFR is 12.5

If you use the ABC method which is Actual Before Commanded or Actual/Commanded you can determine the "percentage of error" in that cell.

So 12.5/14.7=.85 or 15% in this case to the rich side

If the example was the opposite;

commanded AFR 12.5
actual AFR 14.7

The error would be 14.7/12.5=1.18 or 18% to the lean side.

This is where the BEN factor comes into play as the calculation for the BEN is the ABC math.

Once this is done correctly, you have now allowed the PCM to determine where the volumetric efficiency of the engine at all ranges which is the foundation for all fueling of the engine.

Howard

SSpdDmon
August 4th, 2007, 04:02 AM
So if I log and see that BEN is mostly 1.00 and I also notice the car is also running lean I would then need to adjust commanded AFR? Yes? and sorry, I am at work right now....still need to look at the tables more so I have a visual. New laptop up and running today though :)

If you log and see that BEN is 1.00, that means the AFR you're commanding is equal to the AFR the WB is reporting. That means the VE is accurate.

If you think that AFR is too lean, then why are you commanding it?

zapp168
August 4th, 2007, 01:14 PM
Guys, this is opening my eyes more than anything I haveever read!!!!:)
As for the running lean comment sorry, my bad it was as an example.
BUT as an actual issue. I have an AEM AFR gauge (uego controller) and it is in my face on the pillar. When I went into boost it just went down to 10.0 and stayed there. ( I thought it was wrong, I realize it was bottoming out)When I set up the SD tune (OS3) for boost it just said start with the last row of the VE table then increase 300 percent and fill in the middle, my thought was they just started super rich so as to not damage the engine, as I started lowering the VE table it started to lean out. That is where my VE table controlling AFR misconception came from. I maybe should not have started off tuning with updating a 98 to a 02 PCM, swapping to a custom operating system, then tuning a head, cam, turbo, car.....Thanks guys for being real cool about my ignorance. The car seems strong as hell until I get to the track :( (pre turbo 12.5 @ 110 post 12.3 @ 120) I am trying to work it out.....like a one legged man at an asskicking contest...

zapp168
August 4th, 2007, 01:50 PM
If you think that AFR is too lean, then why are you commanding it?[/QUOTE]


I Have not really worked at this yet, but wouldn't I be commanding it to a richer ratio?

SSpdDmon
August 4th, 2007, 03:42 PM
I Have not really worked at this yet, but wouldn't I be commanding it to a richer ratio?
I didn't realize it was just a made up scenario. Ideally, you command the AFR you want and dial it in. A BEN of 1.00 means it's dialed in. So, it looks backwards to me to see someone post about having a BEN of 1.00 and still be lean.

zapp168
August 5th, 2007, 08:29 AM
Ah ok.....sorry....my bad.....Now going to retreat to my car and my laptop for some more button pushing.....EFIlive just needs to add a "faster" key :)

FreddyG
August 11th, 2007, 07:42 AM
I just found this thread and WOW, does it make tuning a little more understandable! I did a search for commanded AFR and found this thread and it even helped a Block Head like me out! Thanks Guys! :cheers:

vatman02
September 16th, 2007, 01:41 PM
i am running cos 5 and have a fueling issue. i have done the auto ve and got the ve real close, only problems were at closed throttle and wot. all the rest of the ve was .99-1.01 bens. i have since changed over to E-85. i am now seeing lots of extra fuel being dumped in the middle ve areas that had been good before. if i revert to the stock os it goes away. can someone please explain what in the cos5 would cause this. i copy and paste the tune from the stock os the cos5 so everything is the same, except the fueling. any help is greatly appreciated.