speed density

[mikesp1]

is it posible to convert a 2006 e38 to speed density
Mike

[ringram]

Yes, but its controlled via formula, constants and coefficients instead of a VE table. Mr GMPX is working on something custom. But to see how it really is done read this http://www.ls1tech.com/forums/showthread.php?t=710290

Im guessing that will keep you busy for a while

[mikesp1]

I think ill just spring for the ls2 maf

[redhardsupra]

The end is near! There shall be no fear cause I is here!

It's really not that hard, I just had to catch up on all the math classes I slept through in the past. For now I got the methodology down for E40, E38 is actually simpler in some aspects (less variables to optimize and they yield greater precision). That's the next thing to attack for me.

[geoff]

We are already doing it. The math has been reverse engineered and we can basically put whatever VE values in the E38 we see fit.

[wait4me]

I dont think it is as complicated as it looks. changing 2 values in each zone that you are operating in will make an infinate amount of adjustment even in boost with a standard 2 bar map.

[redhardsupra]

We are already doing it. The math has been reverse engineered and we can basically put whatever VE values in the E38 we see fit.

yea but that's going back to having one VE value for a whole area/cell/zone whatever you wanna call it. the idea behind the parametrized zones is to have precision at every spot within a zone, not just at the centroids.
and even if you put in a fake VE table on top of it and then do a multiple regression to find the parameters, isn't it gonna miss a lot of useful data, and basically just use few averaged values instead of the precise ones?
sounds like a lot of effort to do few steps backwards in evolution...
or am i missing something here?

[geoff]

yea but that's going back to having one VE value for a whole area/cell/zone whatever you wanna call it. the idea behind the parametrized zones is to have precision at every spot within a zone, not just at the centroids.
and even if you put in a fake VE table on top of it and then do a multiple regression to find the parameters, isn't it gonna miss a lot of useful data, and basically just use few averaged values instead of the precise ones?
sounds like a lot of effort to do few steps backwards in evolution...
or am i missing something here?

Basically we can create a fake VE table with as much or as little precision as needed. It can be every 1 RPM and every 1 Kpa and a value assigned to such or a VE table with only 4 values (although that would be quite useless). We aren't just putting VE values in the 30 zones. We have as much precision as is needed or wanted.

The bottom line is that we have figured a way to do 'conventional' VE tuning and convert those values to the coefficients that the E38/E67 needs for SD tuning. We can convert the coefficients into a table or take a table and convert it into coefficients with a good amount of accuracy. We have also verified that the coefficients that we are inputting are resulting in the desired VE values.

[ringram]

Jolly good guys. Geoff are you working with Ross on this? Was this feeding into the quasi VE table Ross posted up from the E38?
Or have you guys sorted this out independantly?

[GMPX]

We are all collectively banging heads (literally).

Cheers,
Ross