http://forum.vi-pec.com/forum/viewto...3&p=9187#p9187

 

Hmm. Well, I'll have to sign up for that forum so I can see the attached files.
But off the top, I don't see how what Dave did could possibly be right.
Seems to me he only used 2 of your values (one from each end of your MAF table) which is all you can do with Cal 4. So all the rest of the volts to g/s translation between those 2 points is just linear, not following your table at all.
He is treating it like it is the same problem as setting up your wide band which it's obviously not, because your wide band is linear and the MAF is not.
That's how it looks to me anyway.

 

Agreed.

 

Heres how I did it Talon. Just tinkering around. I posted these in my thread @ vi-pec hoping they can comprehend what Im after a little more.

 

Well gee that plot in ecuflash looks real sweet.
In the Vipec picture, how did you get it to make the plot?

 

View>table graph

There's a shortcut but I don't recall it.

I hate to say it but Subaru oem computers are bounds ahead of Mitsu. Engine hardware and all, Subaru has it on lock down. IMHO of course.. I liked my subies much more than the Evo .. The sti 6mt is amazing

I hope that Dave sees my post and has an ahah! Moment.. He says I didn't provide him with example voltage curve but I clearly uploaded my spreadsheet lol

 

The Help doesn't come right out explicitly enough for me on the subject of the number that is in each cell of the Fuel Table. So maybe you guys can clear me up on this.
I know this number is a % and that it can range from 0 to 100.
I gather it is related pretty directly to injector pulse width, but it is not actually pulse width in ms obviously.
So let's say you have a cell where this number is 20, and another cell where this number is 40. Does it just mean that the "40" cell has exactly twice the pulse width of the "20" cell?

Wow I just read the last few posts in your thread over in the Vipec forum. Dave just doesn't get it at all does he? I don't think he knows what a curve is!
This really makes my skin crawl, because for the last 23 years that I worked at Boeing I was a "Lofter", and curves are what it's all about in Lofting. Curves and Surfaces.
Back in the old days, before I was even in Lofting, the Lofters of that time did make point tables to approximate a true curve, the idea being that an NC machine would go straight-line from one point to the next, and if the points were close enough together, the path you get by following all those short little straight lines would always be within the "CHT" or Chord Height Tolerance of the true curve. The CHT we used was usually .001". We had software that ran on a mainframe to do all this stuff. When you would plot out a table of points to see how they looked it was easy to see how they would bunch up where there was more curvature. There would be thousands of points generated to define a surface for some area of the airplane.
Believe it or not, some of the parts made for current production airplanes are still made from this ancient data, because the ancient data is what is called out on the Engineering drawings to this day. It costs too many man-hours to go back and change all that Engineering if the shape of the part hasn't changed.

When you make a table of points like you did in ecuflash, does it tell you how many points you can use max? As far as you know, would ecuflash be able to use that table you made no problem, with 40-some points in it?

 

Its % off your materfuel number. So say my master fuel is 9ms then the numbers in the fuel table are a % of that 9ms.

 

The spreadsheet I posted is all ecu flash data

 

Aha! I had not read about "Master" which does kind of explain it - but with the condition that MAP is at 1 atmosphere.
So then re-reading "Fuel Equation" it comes together a little more. Sounds like the final pulse width at any given instant is a product of Master in ms, times the current fuel table value in %, times the current fuel equation value which is based mostly on the MAP value.

This all explains why the span of the main fuel table numbers is so compressed, and also it explains why a "Master" of say 9ms is enough even though actual pulse widths could go as high as 20ms or more, probably 25ms if your name is Curt Brown!

 

So probably looks something vague like this ?

((Master)x(fuel table))x((fuel equation)x(trims))

I should have logged ipw on the stock ecu and made a 3d table. Infact I bet I still could populate a 3d table based on existing fuel table .. Agh except I was maf and load means squat to me now. Well it would have been a good idea

But quick tune is so efficient. Hard to beat that, creative road loading and lfb can cover a lot

 

Yeah, I forgot "Fuel Corrections" which I see includes the IAT Fuel Trim Table, so that would be another term I guess in the whole product to get actual pulse width.

On the quick tune, I've been wondering what you guys think of it. How far up the rev and boost range can you go with it? Can you go all the way up - 30, 35 psi, 8000 rpm? I suppose you would have to creep it up there, and probably use "Tune ahead" and "Tune Ahead Trim" feature?
One thing that worries me about it is that at high boost things start happening real fast and I'm not sure yet if the widebands are fast-responding enough to keep up with it accurately. The LC-1 is supposed to be real fast, but I think I see some lag between my LC-1 and my stock narrow band. Not sure, I'm trying to get a handle on that lately.
Now if I had a test track where I could do all this in 5th gear at 100+ mph, that would slow things down enough I probably wouldn't be worried about it.
In Europe I guess you wouldn't need a test track. Highway would be fine.

What do you think?

 

If you can load the car up it shouldn't be a problem. You have to be fairly steady with load and rpm sites. Works very well as far as I've used it. I agree that it's no substitute for a well educated tuner and still requires good practice. But you can do a column in about 10 minutes if your left foot brake and right throttle technique is well rehearsed.

 

Just use the mixture map if you are street tuning.

 

Hmm, brake loading, hadn't thought about that. Maybe a nice long uphill climb somewhere.

Mixture map, yes I've been looking at that. Target AFR, actual measured AFR, number of samples based on, error between target and actual, yeah that's great stuff. What I haven't figured out yet is - isn't that what Quick Tune does? I mean, when you do a "Quick Tune" you are using "Mixture Map" aren't you? Or are they different things? Maybe another way of asking is, can you use "Mixture Map" without using "Quick Tune"?