|
With two cylinders not producing any power, that leaves much more loading against the two that are producing power, which changes the transitional piston speed during the power stroke, and results in the lugging.
Spark advance would need to be reduced to match this increased work load and slower piston acceleration. You will also likely need to increase the stall-speed of your torque converter and/or up the stall-torque ratio inside the converter. That part is not usually easy or cheap unless you can find one that already fits your need, and fits your vehicle.
IMO, it's best to keep the ECM in open-loop mode when testing a drastic change like you're doing. Even if closed-loop functioned perfectly during your test, stoich just isn't going to cut it when you're trying to accelerate with only two cylinders. Also, if you're not data-logging, then there will be far too many unknown variables in your experiment and probably a little too much speculation when it comes to the results.
I think this is interesting and I don't want to sound discouraging, but you know if it were very easy to do, ...
I like the idea, but it's hard for me to imagine being able to pull it off on a shoestring budget. If I were trying this experiment, I'd want to go through the engine thoroughly before-hand, eliminating the cam followers, rods, and pistons on the dead cylinders, and epoxying up the dead ports in the intake manifold, - make sure all open oil passages were either plugged or rerouted as necessary, and add weight to the flywheel or install a looser converter.
As for the crank, I have had to lighten the counterweights for performance rods/pistons before. It's relatively easy, though a little more weight would need be ground off in this case. I have my V8 cranks balanced for usually $150 or less - usually less if they don't require added weight. Tapping and plugging the oiling holes in the crank pins would also be easy. Oil pressure and delivery to the remaining bearings would actually be improved, possibly allowing a reduction in volume from the pump.
I would use a stand-alone for engine management, or an older GM ECM in case the required tuning parameters were such that the code had to be patched. The increased pulsing in the intake manifold would probably require extra smoothing (filter routine) to calm down the MAP output at lower RPM. The throttle body would also need to be physically downsized in order to preserve any sort of part-throttle resolution in both the spark and VE tables.
This is all bench-racing though. I have a couple of spare four cylinders, but no time for such an experiment in the near future.
__________________
#####################################
|