Are you logging fuel pressure as well to make sure that it is not shifting from pull to pull? That's one of the "constants" the ECM doesn't have control over that could play a big part in your variance.

I would take Haxby on his offer, he is an EFI and Holley EFI genius. I'm on my 3rd season and 2 motor with Holley efi, and it's been nothing but amazing so long as the initial settings/base Map are correct. Send him your global folder, I be you have an answer by the end of the day. Good luck

You are saying that in open loop your reading will change 2.5 points in the exact same fuel map cell? That would be my question. Most likely you are getting into a different cell in the map that is not correct because of load, rpm or whatever reason. Get some data logs and study them extensively, then smooth out the map. I'm currently building a Cobra replica with stack injection and Holley HP efi. I had some similar issues with tuning below 2500 rpm. I finally turned off closed loop and mapped it manually to get it to run right. I'd like to see a copy of your tune file and some data logs if you get a chance to send it.

[email protected]

Bob

I hope to get back up there this weekend. Just too busy to spend a bunch of time online with it etc. but i'll get the file and logs when i go back. Bascially all was equal and 15 mins or less apart. When you get it like you want it, transfer learn to base & smooth the table then go out of closed loop (zero percentage corretion) it will not repeat itself but it should. I wonder how many users have spent the time on dyno to watch the open loop runs several times after taking it out. Ultimately in my boat, i can never run correction and need to count on it staying where we set it. So far, that has not been the case and it's not only on one ecu.

Also, there are other bugs in it as well. .. Like regardless of how many times and when you cycle ignition it has saving issues as well.

This could well be your problem.

If you smooth the table, and you have some adjacent values that are significantly different, it will skew your "learned" data.

I can put some random values in my Dominator tonight to illustrate this if it helps.

I always have had to do some manual mapping. I get them to where I can run open loop before they come off the dyno. Then retune in the boat. You just can't simulate all the load points on the dyno. Closed loop is a nice tuning aid, but you can't keep it there in a boat. Even for tuning I seem to always turn it off at lower rpm and tune it manually. I've got it turned off below 2000 rpm in my Cobra. It just wasn't happy with the stack injection at very low throttle settings.

Bob

I would take haxby up on his offer!
He won't even return my calls related to remote tune! :-(

It sounds like they aren't very familiar with how to use the software. Also, I'm not sure how your dyno shop not being able to get the tune dialed in on the dyno translates to You don't have to call me, but call someone who is familiar with the system before these guys waste a whole bunch of your time and money and talk you in to putting a MEFI on it.

I thought you ran FAST EFI Larry?