Tracy,

Correcting for the effect of injector lag only under certain conditions is missing the whole point of making the correction at all.  The dead time is constant and as a result, it affects the amount of fuel delivered under all conditions.  The fact that the correction made by my hardware was so effective tells me that you have nailed the calculation of the amount of fuel required.  All we have to do is actually deliver that amount of fuel. 

The secondary injectors are affected by their dead time also, and the larger they are, the more of an effect that dead time will have because the pulse width is smaller for the same amount of fuel delivered.  The information that I’ve seen indicates that dead times are slightly different for different injectors but not by much.  The 1.2 ms that I used is not a magic number, but using something in the neighborhood of the actual value is much better than leaving it out.

Applying the dead time correction across the board removes the requirement to correct for it anywhere in the MAP table.  Mode 6 then is needed only to compensate for the different flow rating of the secondary injectors rather than using the MAP table correction to compensate for a variable combination of flow rating and dead time (amoung other things) throughout the whole MAP table.

I’ve attached an updated plot that clarifies the questions Ed asked.  As can be seen, it took less than 30 seconds to go from an essentially default controller condition to one that worked well over the entire power range that I could attain.  Adjusting mode 6 for the case of different sized secondaries would only add a few seconds to this procedure.  I would expect that the tuning procedure would be equally effective no matter what sized injectors (within reason) were used as long as their dead times were taken into account. 

If it doesn’t work this way, I’ll eat my socks.  What the heck…let’s up the ante… I’ll eat YOUR socks.  Know that I’m smiling while writing this even though I might get a package from Shady Bend in the mail.

Steve

Interesting results Steve.  I didn't think a simple across the board increase would be that effective.  In view of your earlier results I made a minor change to the EC2/3 software to bump up the injector pulse width but only when the staging point is reached.   This only applies to the setup for same sized primary & secondary injectors.  With the much larger secondaries on the 4 port Renesis the mixture goes rich instead of lean at the staging point (unless  Mode 6 is adjusted to compensate).

I have not flight tested the change yet.

Tracy