Okay, this seems too easy to work.

I get myself all set up, and start playing around with Tunercat. I quickly find that the limit for VE numbers is 100. So, on a goof (no doubt inspired by a Killians), I go into the TDF editor and raise the max value to 145. Then, I go back to the ecm editor and find that I can indeed raise VE numbers to 145 with no complaints.

Of course, I have not tried this on the vehicle yet as my max VE in the OEM tables is 90.2 and even taking it to 100 would likely throw A/F ratios into the single digits. I am sure someone has already tried this, could this work in reality? Of course, I realize that it would also depend on IPW and duty cycle being low enough to allow that much more fuel. What is the trick to calculating my current duty cycle? I have max IPW from my scanner, but is there a way to calc all this from the VE numbers?

 

from what I know if the VE tables add up to more than 100, the ECM limits it to 100.

This is from a 7747 ARJU hack.

 

I had the same problems with my VE tables. I would alter my VE tables according to Datamaster test runs. From the test data I would generate figures over 100 trying to remedy an overall lean condition! After several test runs and increasing and increasing the VE tables over and over, the VE tables satrted looking like a table top (flat)and I saw a little but not much improvement in my BLM's. I went back to the factory VE settings and decreased my injector constant from 61 to 58 and the lean condition was eliminated. While I was adjusting the VE tables I would set some cells to 100, but after I burned those to a chip and would come back and read the tables, those 100 magnitude cells would be adjusted down to just under 100 (i.e. 96-98). I never could get the VE tables to burn 100 onto a chip!

This is just what I went through. Maybe someone else can tell us what is happening.

Tedd

 

Lower the fuel injector constant, this will richen your VE tables so you don't need to have numbers around 100 in the table.

 

Guess im not the only one who went thru that then huh :-)

when i had my formula setup with the stock inj during my tuning(never finished , tranny went then eng grrr) i had several values at 100 or very near.

When i had the second iterationrebuilt and the new tranny too i got smart and went with svo 24# inj and have the ic set at 24.1 i believe me highest value to date(thru about 2800rpm more or less dialed in on the ve) is just into the 60's. I dont even approach the 80's until the mid 4k range which i have not even begun to serioulsy tune for yet and is rich at this point still.

The way i did it was knowing that i had a somewhat stable idle and the lower end ve semidialed in on my old eng. i used that programming and bumped up the ic until i got it off pig rich(108) and the blm into the mid teen range(115 give or take) and then began to retune my ve curve from there.

As a side note to all this i know some of u out there dont have a laptop and are using a scantool for datalogging(as i am currently, awaiting laptops number 3 and 4) .... Find tpi_rocs post about the excel; conversion spreadsheet and make yourself a reduced copy of a blank VE table to fit on one page. then make yourself a bunch of copies and talk a friend into going out with u for a few hrs(my gf hates me now i think:-) ) the more u fill in the chart with your blm values vs load vs rpm the overall clearer picture that u have and using the graphing function like in TC u can smooth out the high and low points and get yourself a real nice starting point on your ve curve. i had about 50 values total and was amazed by how much else i could fill in just by using the graphing feature and estimating the missing blanks. I will be doing a second sweep this weekend hopefully so i can do some work chipwise while my sfc and exhaust is being done .....

It may not be much or help but it working for me:-)
good luck
jeremy

 

"i had about 50 values total and was amazed by how much else i could fill in just by using the graphing feature and estimating the missing blanks. "

I want you guys to understand something.
When you run out of VE, your running out of fuel. Either the injectors are too small, or the constant is not set up right.

Now for the filling in the blanks. This could not be more correct if you tried. That IS how the ecm uses the VE and all other tables.

The ecm uses interpolation to find the table values. TWO tables points are used (right next to each other) for a 8X16 lookup. The eight bit is the input value, the 16bit is the 2 table values. Other tables are a 16X16 lookup.

To understand how this works, you need to know how Polynomial reconstruction works. Just kidding, the ecm does the math for you. But it is something you might want to get a small grasp on.

What you do need to remember is that when you make a change to a table point, you affect the numbers above and below it. To find out which numbers you need to worry about, you do need to know which way the table is going. Almost all of GM's tables are RPM based.
This means you look at the left column on tunercat to find the rpm range, then the two map vaulues.
If a INT/BLM change was made to RPM block 1600, and MAP value 40, the table points came from Map value 40 and 50 (or whatever the next one UP is).


It is perfectly acceptable to map out a few areas on the VE table, interpolate your own with the 3D graphical maps, make changes and verify it with some more logged data.

I consider a well tuned motor to have Block Learn values between 134 & 122. Not every motor can achieve this though.

The rest of the work is to be done by the ecm with the INT/BLM learning. After all, that's it's job. Don't try and fire it.

Weather, ambient temperature, and most of all cylinder wall/head surface temps greatly affect the VE tables. ALuminum heads will have more variance than any other. They just don't hold the heat like cast iron does. Go drive around for a while, then let the motor idle for 2 minutes. You will see what I mean.