ORIGINAL: 123Cat
Would it be a good idea to have a go at dynamic balance by inserting weights in the prop driver, to offset the counterweight end ?
I can see the merit of placing a heavy blade of a prop on the same angle as the piston at TDC but this is because of the moment arm or distance that the blade length can provide will mean that any amount of offset weight will be at its most effective.
With a prop driver the small amount distance or moment arm length will provide very little meaningful gain no matter what amount of weight you could apply.
The point with single cylinder engine balance is to try and cancel out out the linear travel of the piston and con rod with the circular motion of the crank and although you 'can' achieve 100% primary balance (at TDC and BDC or 0º and 180º

with just a standard crank web, but the secondary balance (90º and 270º

will be shot to ribbons by doing so - you will double the secondaries by achieving a perfect primary that way. Then there is the combustion pulse to consider that only exists at TDC, so the best most designs can aim at is something under 60% but even that lives only in text books.
The 'heavy blade' idea could be visualized like an inline three cylinder engine with the both the left (heavy blade) and right pistons (the real piston here) rising and falling together and the middle piston (the crank web) providing the missing 'balance' but all of that really needs the moving parts to have exactly the same weight, height, distance apart, acceleration etc and this is never going to happen if the moving parts all vary greatly as in a single cylinder model engine.
The fact that the pistons travel is not in the same plane as the spinning crank web will give rise to a rocking couple, and this same mechanism of rocking is far worse if you then view how great in comparison the prop driver is from the piston - get the prop end balanced wrongly and things can get pear shaped rather rapidly.
Iwould theorize that you could indeed get your 'three cylinder' balanced almost perfectly but it would only apply for one engine speed due to the differences mentioned before and probably why engine manufacturers have simply gone for more of an average across the usable rev range.