Novel FI Tuning Idea, I'm sure it's been done.... Options

[McMark]

I was thinking about graphs (see my Excel help thread) and I go to thinking about how to use this datalog information to help me see where my FI map is wrong. It dawned on me that the two major input for this KitCarlson system are TPS and MAPS. The result or output would be RPM. So if you graphed TPS on the X-axis, MAPS on the Y-axis, and RPM on the Z-axis you should be able to see where you've got problems. Of course you would have to average your readings over time since you would have a lot of repeat data in the cruising RPM range. What are your thoughts/responses to this idea?

[Jake Raby]

What I have learned in the past few weeks has taught me that the resolution of Dave's system increases the tuning difficulty a bit and that would make this much harder to do. I have tried it when I had the system in the 912, while I was very happy with it I didn't care for the values now that i have experienced SDS, Autronic and Haltech like a mad man for the past 7 weeks of DL-1 datalogging.

With the SDS I was able to only set a MAP value every 3rd RPM fuel value and then interpolate the ones in between to the point that the entire fuel MAP was done in only one dyno run. The MAP values were the same from the mid 14s to the low 1s range but on the dyno we can't set high MAP numbers because we lack a way to force the engine to decelerate at closed throttle.

Hopefully the 3D graph I am sending you as I type this will help.. It helped me to add my current values and then smooth the 3D graph and then upload those changes into the EMS. This is much the way autronic tunes their map values and I think MS as well..

[Mueller]

Mark,

The Link has the 3D mapping as well (still figureing it out).....feel free to drive down here today to help me tune my car (IMG:http://www.914world.com/bbs2/html/emoticons/wink.gif)

[DNHunt]

Mark

Is it possible to log O2 readings also? To get a rough map I use the FI in closed loop with a really large correction and look at the MAP and RPM points where its really working to correct. I then change those points. Finer tuning I use a viewer like you are talking about. The graph makes it easy to see where to change. My only problem has been avoiding the temptation to use readings that lie between the RPM and MAP specified values. Also, O2 values are different at the same RPM and MAP points depoending on whether the engine is under load or compression. I tend to ignore the p[oints that are under compression.

Jake

I have found that my injection goes way lean >18 on decel when it's pulling high vacuum. I've ignored that cause I really don't think I can do anything about it without something like a decel valve. I then set the acceleration enrichment to give a small squirt as soon as the throttle cracks. it seems to work. Incidentally, I should be starting on my rebuild soon. I wanted to use the old engine as a test mule for the FI and ignition so i won't repeat my problems.

Dave

[Jake Raby]

Dave,
I found that a richer mixture on decel at high MAP made for a much better deceland transition when you get back on the throttle..

At anything above 14:1 mine wants to hunt around at decel.... when its a bit richer its really smooth... With SDS setting those values is a cinch.