I have a sponsor, North Birmingham Truck and Auto, 800-438-7158, that allows me to borrow several at a time for dyno work. I pay for what I end up using. They have lots of e30 parts, and so far, I have used 12 different DMEs from them. Chuck
I'm out for a while
Here is a representative plot from the end of the dyno session. Scott’s plots are very similar. [file name=dyno1.jpg size=66302]http://spece30.com/media/kunena/attachments/legacy/files/dyno1.jpg[/file]
jlevie wrote:
Scott’s plots are a little less. There’s hp loss associated with his 6 oil pumps.
That is a nice air fuel plot always between 12:1 and 14:1. Wish there was some way to fix that ugly dip at 5000 RPM though.
King Tut wrote:
My guess is that dip is due to flow dynamics. I can’t think of anything legal to do about it, though it sure would be nice to eliminate it.
King Tut wrote:
I heard a theory that the engine had a tendency to knock around 5krpm so BMW retarted the timing there. That gave then a choice of retarding or specing premium gas.
Anti-knock sensors came later.
That climb up to 14:1 after 5k seems very lean for full throttle high RPM conditions. This is the first time I’ve seen a stock ECU dyno plot for an M20. Do they all do that?
AndrewK wrote:
The curve is pretty cherry. You’re looking at the result of a lot of part swapping. Folks that haven’t worked this issue have A/F curves that look worse. Some a lot worse. My old A/F curve spent a lot of time scary lean.
AndrewK wrote:
In my experience, yes. And other than swapping DME’s there is nothing you can do about it (legally) if the fuel system is working correctly. I’d be a lot happier if the A/F ratio never went above 13:1.
Some time back I tested a half dozen or so junkyard ECT sensors and found one that had slightly higher resistance at 100C, which is what is in the car now. It wasn’t off much, roughly indicating 5F lower at median engine temps. That, or finding an IAT sensor that indicates a bit cooler, is the only legal way I know of to improve the A/F ratio above 5k at WOT. Maybe I’ll try that next…