Bengt-M
commented
8 years ago I'm still not used to how things I do on my laptop propagates to the PR at GitHub/Triflight. After the last merge I did one rebase too much and all comments disappeared. Anyway, here is a small analysis regarding the limits:
A prop with a higher factor must have less drag and must produce more thrust per watt. That is not easy to do. The prop selected by David W is pretty good. Let's say we can find another one which is double good (but then I think David would have chosen that one already). If our prop have a factor of 14, I'd say its unlikely we find another one with a factor >30, because I think that would mean that the drag is half. That corresponds to an angle = 91.9 deg. Remove the misalignment from bad bench tuning. 1 deg is cleary visible. Gives us 90.9. So 90.5 (factor=114) should be on the safe side. Your current function stops at 90.15 deg but I don't see how that can happen. I don't think a prop can be that good. (if so, where can I buy it?) At even lower angles the factor grows very rapidly into indefinit at 90 (divide by 0).
In the other direction; how bad prop can we have? Four times worse? (Who would buy it?) A factor at 14/4 corresponds to 106 deg. Add one for misalignment gives 107 deg. Say 120 to be safe.
So 90.5 .. 120 deg is a reasonable and not unrealistic range, given that the bench tuning is good. If we put 90.01 .. 180 there would never be a failed tuning and we can remove the check completely.
Does this make sense or did I miss something here?
lkaino
I have mentioned this possibility before. Here it is. For me it passes the same unit test.