This thread is following the usual course of discussions of thi
subject. It never fails to amaze me how much misinformation most of th aviation community seems to hold with regard to props. It's as bad a the old "downwind turn" debate!
I've covered this subject of optimum number of blades in depth in th "Ask Joe and Don" section of our website at:
formatting link
(for that matter, I've also discused downwind turns, but that's anothe subject for a different category and thread)
However, in a nutshell, it's fundamentally a lot simpler than what thi thread so far would seem to indicate.
Just as a wing makes lift by grabbing chunks of air and shoving the down, a propeller makes thrust by grabbing chunks of air and shovin them aft. For a given amount of thrust, you can take a small chunk o air and give it a violent shove, or you can take a bigger chunk of ai and give it a more gentle push.
The production of lift of a wing results in a by-product we cal "induced drag". Likewise, the production of thrust by a propelle results in "induced losses", the natural by-product of making thrust Grabbing that larger chunk of air and giving it the more gentle pus results in smaller induced losses.
There are also what are called "profile losses" for a propeller analogous to an airplane's parasite drag. This is the energy consume just from the act of moving the blades through the air. It include things like the skin friction of the air scrubbing against the blades surfaces.
For a given diameter, if we add more blades (all other things bein equal), we can spread the horsepower more evenly over the surface o the propeller disk, which reduces the induced losses. However, we'r moving more stuff through the air, so adding more blades generall increases the profile losses.
Now for the tricky part. The induced losses reduce when we add blades the profile losses increase, and the total losses are the sum of thos two. Whenever we have a mathematical situation like this, where on variable is continuously decreasing and the other continuousl increasing, the lowest total of the two will occur at the point wher the two are exactly equal. Always. Just as the best L/D of an airplan occurs at the airspeed where its profile drag and induced drag ar exactly equal, the most efficient number of blades occurs where th profile losses and induced losses are exactly equal.
So, we do better with fewer blades when the "disk loading" of th propeller (horsepower per square foot of disk area, sort of th propeller equivalent of wing loading) is low; in other words, when w have plenty of disk area and not all that much horsepower it has t absorb. This is usually the case for model airplanes, which is why w usually get better efficiency from a 2-bladed prop. When you have mor power, such as the big Allison turboprops on a C-130 Hercules, or thos Russian Kuznetsov 14,000 horsepower turboprops on the Bear bomber, si or eight blades might be more efficient.
The other thing that plays into this is the flight condition. Imagine cylinder of air equal to the diameter of the prop, and with a heigh equal to the distance the plane moves in one second. The mass of th air inside this imaginary cylinder is a representation of the size o that "chunk of air" per second that the prop is grabbing to make it thrust. If the cylinder is larger in diameter (i.e.: a bigger diamete prop), we increase the cylinder's volume and help minimize the induce losses. Likewise, if the cylinder is longer (i.e.: faster airspeed), w increase the cylinder's volume and help the induced losses.
Thus, for slow speed/high power flight conditions like takeoff an climb, more blades tend to be optimum. In cruise, where the airspeed i higher and the horsepower is lower, fewer blades are optimum.
For example, I recall one regional turboprop airliner application w studied at the prop company where I used to work (before foolishl going into the model airplane business full-time!) where a six-blade prop was best on takeoff, five blades and six were about equal i climb, but a four-blade was best in cruise. Designing airplanes is all about managing your compromises effectively.
When it comes to noise, similar principles apply. The noise of the propeller is the sum of the "pressure noise" (the noise that results as a by-product of making thrust, analogous to induced losses), and "thickness noise" (the noise created by the disturbance of moving the blades through the air, analogous to profile losses). Once again, the optimum number of blades from a noise standpoint occurs where the pressure noise and the thickness noise are exactly equal.
As far as this discussion about tip vortices interfering with each other, yes they do, but interference between blades is not a significant factor in propeller performance until you have a really large number of blades, like more than six or so.
Don (former full-scale prop engineer)