So, After countles prop iterations in my foil assist I was finally able to get one that was pushing as I wanted.
Thing is, the RPMs were much lower than anticipated. I started with a value close to the free rpm (kV*V), but eventually had to reduce it a lot, to almost 50% of the free spinning one.
So I started looking for info on the internet and found this Brushless Motor Power and Efficiency Calculations - Tyto Robotics. Which is pretty interesting. Basically you have to force the motor all the way to 0.7 or less of the free rpm if you want to get the max mechanical power out of it (as long as you don’t care about efficency)
But I had to push my rpms even further. I think this is due to to battery sag. My battery pack is really small, so I could well be asking 30A per cell, which reduces the voltage from 4.2 to 3.6 under load (see Test of Molicel INR21700-P42A 4200mAh (Gray))
So there it is, if I now do (63.6)V170kV*0.7 = 2600 rpm, which is the value that was working for me.
Honestly I don’t think that prop design is “rocket science”, as Dynamik loves to call it. Maybe it was in the early days of aviation. But nowadays, builing on previous knowledge, and being so easily accesible as javaprop /B-series designer, It’s more a matter of knowing how the whole setup will interact. But since this is actually complex, I guess we’ll just have to stick to trial and error. After all not even the best steady state simulation would tell you how loose/fix a prop will feel based on its pich.