Mark & True,
Thanks for your thoughts and comments. If you don’t mind, I’ll clear up a few points which may yield more ideas.
My primary problem is not with downshifts followed by WOT (although I sometimes get it then as well), it is with all fast up-shifts under full acceleration.
As noted somewhere above, my current MAP cutoff is a very simple vacuum actuated micro switch that operates an electric solenoid valve. To be more specific, the switch is supposed to trip as manifold pressure passes through atmospheric, and it in turn closes the solenoid which is in-line with the vacuum to the MAP sensor. Theory being that the line between solenoid and MAP sensor remains at atmospheric (equivalent to WOT and hard acceleration on an NA engine) until vacuum is restored to the switch, at which time the solenoid opens and allows vacuum to the MAP sensor.
Although I lack the equipment to measure the exact pressure at which the switch operates, by sucking on the line I can tell that it takes only the slightest vacuum to trip it. Also, I can avoid the problem by initially rolling the throttle on more slowly and then everything is fine to redline in that gear, which suggests that it functions as designed under those conditions. That leaves the obvious question of what’s different immediately after a shift. Perhaps the momentary vacuum in the line between MAP sensor and solenoid during the shift gets trapped there because the return to boost is very fast, and the lines are not free of vacuum before the valve closes. In other words, maybe the line to the switch sees zero vacuum faster than the one to the solenoid, which equates to something true said. Given the circuitous routing of lines to keep things tidy, that just might be possible.
I just went out and tried to re-plumb and re-wire things to test the theory by having the solenoid open the MAP sensor line to the atmosphere when the switch sees zero vacuum. Unfortunately, due to the single pole nature of the solenoid I can’t do that without also leaving the feeder line open, which freaks out other sensors. With a little help from a friend I can probably simulate it manually.
As for putting it on a dyno and exhaust analyzer, I did that. I managed to get it dialed in pretty well with no sign of detonation on the dyno. Unfortunately, each run was done with gentle up-shifts followed by the WOT run from a low RPM, so the critical conditions of this problem, which I hadn’t yet grasped, were never met. Later that day I was shocked to hear it hammer worse than ever as I hit 2nd on a hard run. I’d like to go back and watch the mixture during a hard shift on the dyno, or get a portable wide-band to play with. (Yes, I am well aware of how deadly detonation is to a boosted motor, but you are well justified it repeating it at as often as the subject arises! My ears are finely tuned and I lift NOW when I hear the dreaded ping.)
I am currently using the Greddy Rebic IV controller on two auxiliary injectors, and I have replaced the fuel pump with a high flow unit. Although I got it dialed in pretty well on the dyno, the reality of turbos compared to an SC is that you can get the same max boost across a wide range of engine RPM. That’s what I like best about turbos (especially the little Aerodynes) but it also makes simple mapping by RPM and boost inadequate. At the very least you need throttle position, but it would probably take something like a mass airflow or hot wire sensor to truly get it right. The Haltech E6K appears to have all but the airflow part available so it might get the job done, but I would want to use it only on boost so the stock ECU can do what it does so well the rest of the time
Meanwhile, the smartmap might be well worth a try.
More ideas or suggestions?
