not saving ve values

scott0999 · Post by **scott0999** » Sun Dec 17, 2006 12:38 pm

im trying to tune my "FUEL Low RPM VE % vs RPM" table. I type in the value I calculated, press enter, it shows in the table and is red. I click save, close the table and go back into it. then the values are back to what they were before. it will only save certain values. how am I supposed to tune around this?

edit:

just noticed, if I right click on that table and goto edit xdf item info, then conversion. it has "0.390625 * X + 0.000000" in there.

then I went back and tried adding .39 to whatever value was in the table and it would save that. but from 5.00 to 5.01 it wont.

I can edit the PE air fuel ratio's by .10, but it doesnt make sense how your supposed to fine tune the VE tables

im using 7730 ecm with $8f

Post by **Mangus** » Sun Dec 17, 2006 10:00 pm

Raw values in a bin are saved as integers. When converting from binary to engineering values, there is a minimum delta that must occur before incrementing to next available raw value.

Let's say I give you an equation, X * .5

Plug in 1 for X, you get .5
plug in 2 for X, you get 1
plug in 3 for X, you get 1.5

etc.

Now go the other way. What value of X does the result 1.1 represent? The answer: it's closest to X = 2. So if you typed in "1.1", it would get saved as 1, because the value of X closest to 1.1 is 2. Our conversion equation (X * .5) says that for X = 2, the output is 1.

Confused?

scott0999 · Post by **scott0999** » Mon Dec 18, 2006 5:57 am

for instance in a datalog if the blm was 133 at 1800rpm & 20kpa I would do 133/128=1.039

then I would look in the table under 1800rpm@ 20kpa and it would have 5.86 as the VE

so I would go 5.86*1.039=6.09 (rounded up)

if I try and enter 6.09 it wont save that. it would only save 6.25

I understand why its saving them only in .39 increments, but I dont understand how im supposed to tune the VE tables like that.

am I using the wrong formula to calculate VE, or are you supposed to go up/down by .39 in all the VE tables?

Post by **Mangus** » Mon Dec 18, 2006 7:43 am

There is only so much resolution in the tables. 255 "values" to be exact. In general you shouldn't need any more resolution than that.

scott0999 · Post by **scott0999** » Mon Dec 18, 2006 7:55 am

ok so essentially once I get this tuned, making VE changes by .39, I should be able to get a BLM of 128

when I look at the table it seems like .39 would be too much of a step up/down since the values are small to begin with

so im assuming once tuned, the INT should adjust enough not to change BLM from 128?

Post by **Mangus** » Mon Dec 18, 2006 7:57 am

You should be able to tune with the resolution available. You'll never be able to lock the BLMs exactly at 128 in all conditions, but you should be able get within +/- 4 or so.

scott0999 · Post by **scott0999** » Mon Dec 18, 2006 8:02 am

ok thanks for clearing that up

time to burn 100 chips to get this thing right

scott0999 · Post by **scott0999** » Mon Dec 18, 2006 8:19 am

in the datalogs its hard to find a map reading exact to the rpm like it has in the table anyways. so that makes sense why theres .39 resolution. it would take alot of datalogging and searching to match up to the tables so exact.