Prop design

[nhnifong]

How does one make a high-thrust electric propeller in KSP?

You may have noticed that if you just stick some blades on an electric motor it does generate thrust, but not very much. I'd ballpark the one I made with a medium motor and three small blades at 10kN.
So here are some thoughts about propellers.

The propellers will products a constant max torque and consume a constant maximum amount of power. It's not very much power so I won't focus at all on how to make more efficient propellers.
There is no way to make more thrust by consuming more power from a given motor. It's as if they're driven by a constant current supply (real motors can consume more power until they overheat)

The thrust generated is maximized when the prop is spinning at it's maximum RPM, all other things being equal. Max RPM is limited in two ways:

1. Spinning too fast may destroy blades because they stretch away from the hub and either hit the ground, or break off. The craft does not self-collide though.

2. There is a point where the counter rotational drag on the blades exceeds the torque coming from the motor and RPM decelerates.

At rest, for a given pitch (discussed below) the counter rotational drag on the blades is proportional to the thrust they're giving you. So you want to add more of it until it's almost equal to the torque your motor can apply. You'll know you've succeeded at this if your propeller takes a very long time to spin up.

The counter rotational drag on the blades will increase with the total amount of wing area of the blades, the distance from the center hub of that wing area, the air pressure, and the angle of attack of the blades, and the speed of the blades through the air.

Thus if you have enough wing that counter rotating drag is nearly equal to torque, you'll find max RPM limited not by the slider, but by the drag of the blades themselves. As the blades spin faster, this drag increases until it exceeds torque and your blades slow back down. This is what you want your RPM limited by for maximum performance. 

There's another irritating limit on max speed - the fact that the blades will stretch away from the hub and either hit the ground or break off. You can limit this by using lighter blades, making blades out of a s few parts as possible, and keeping blade COMs close to the hub

Amount of wing area can be increased with number of blades or bigger blades. consider these interchangeable. You can see that there is a tricky design trade off between having a lot of blade wing area far from the hub, and minimizing stretch. Especially since it's not just stretch you're probably concerned about but the effective diameter of your prop so it doesn't hit the ground. 

The angle of attack of the blades is very important. At a stop, blades at a high angle, say 45°, would create the most force (and thus require less blade area to do the same job) but as your craft speeds up, the effective angle of attack of the blades decreases. Imagine them screwing into the air. at some point, your blades are spiraling through the air at exactly the angle they're mounted at and applying no force (but there is virtually no counter rotating drag either) Then, if you were to go faster than that, say by diving or turning on a rocket engine. They'd actually start causing drag on your craft, even when the motor is spinning at max speed. Worse, the air spins up the propeller faster than it's designed to go, and they usually destroy themselves past 500 RPM.
 
Instead of measuring the pitch of the blades in degrees, it's helpful to measure it in the distance in meters that the blade would screw through in one revolution, a blade mounted at 20°, with it's center (the place where KSP applies all forces) 1m from the hub, would travel a distance of 2π meters, up at a 20° slant. The forward travel is then tan(20°)*2π = 2.29m. So if it spins at 360 RPM, (six revs per second) then it wants to screw through 2.29*6= 13.74 meters of air per second. if the craft moves forward at 13 m/s, it produces a little bit of thrust until the craft moves forward at 13.74 m/s and it produces no thrust. If we reached this speed and want more trust we have to either spin the prop faster, slow the craft down (booo) or change the pitch. My formula says increase it it but this is the part I don't quite understand. What would a pitch of 89° give us? at 360rpm the blades would want to screw through 2159 meters of air, which is a respectable flying speed, but somehow I know it's not going to work. they're practically paddleboards at that angle and most of the force they make would be counter rotational drag, which would limit the max RPM we can reach, probably nowhere near 360.

This craft airspeed at which the blade's angle of attack becomes 0, and we can't increase the RPM any more will end up being our cruising speed. The counter rotating drag on the blades at this speed should be zero, since they're just cutting through the air without doing anything. Since blades seem to function this way, it would seem they're much better suited to hovering than to winged flight.

The ideal pitch depends on many factors that change during flight so aim to use the movsble fins as your prop parts. How can a control input be mapped to the blades independent of their orientation on the craft? airbrake function?

I want to increase the number of blades until the counter rotating drag produced is enough that it begins to limit my max RPM, not the slider. At craft speeds close enough to my pitch-determined maximum, ultimately the slider will limit my max RPM no matter what, and I want that speed to be high, which means I want the pitch to be close to 90, which would mean high counter rotating drag.

What exactly is the drawback to having a pitch close to 90 anyways?

Edited June 20, 2019 by nhnifong

[Snark]

Moving to Breaking Ground Discussion.

[AngrybobH]

3 hours ago, nhnifong said:

How can a control input be mapped to the blades independent of their orientation on the craft? airbrake function?

put a servo motor on the rotor first then attach your blades. map the target angle to the translate up/dn or l/r or f/b. set your min and max desired angle and translate speed. I've been messing with helicopters with the new robotics parts. I'm really not sure KSP can handle the physics fast enough or is not modeling props quite right. A couple of notes:

FAR ruins propellers, they don't provide meaningful lift/thrust at any pitch. Avoid this mod for now.
Rotors behave very strangely and often do not produce the power they should. minor changes to prop design produce great differences.
~300 RPM is about the safe limit before the Kraken starts to take notice. not that spinning higher always causes RUD but things can get strange especially when changing directions.
the rotor RPM limiter seems to also limit torque even before the RPM limit is achieved, so I recommend doing what you have planned, let the drag control max RPM and leave the RPM limit at max.

3 hours ago, nhnifong said:

What exactly is the drawback to having a pitch close to 90 anyways

I don't know the exact relationship (or terms) but the more pitch you have the less efficiently you produce thrust(or something like that). Think about a plane flying near level. It produces enough lift to stay flying when at the right speed. Take that plane to 45 degrees nose up at the same speed. You will climb but also lose speed. If you where to go to 89 degrees(instantaneously) you would produce nearly all drag with your wings and nearly no lift. You don't need to worry too much about this in KSP because the rotor will stall long before you get to 45 degrees. Most of my helicopter designs rarely get to 7.5 degrees blade pitch before the G11 rotor runs out of torque. Granted a helicopter main rotor is much bigger than an airplane prop but I was also having similar problems with the tail rotor with ~20 degrees pitch.

[nhnifong]

Here are some lessons I've learned from prop design - 

• A prop where the blades are connected directly to the spinning collider of the motor, and are one part only is the lest likely to self destruct, though it still won't hold up under 2x time warp.
  • Multipart blades stretch too much at max RPM, it doesn't matter what kind of parts they are, cubic strut, servo, fin, etc. Stick to a single part for the blades.
• The part which makes the best blade for me is the basic fin because
  • It's light and therefore doesn't stretch out as far
  • has the best lift/mass ratio of any lifting part
  • is small to fit in more fairings. Since it's small, use a lot of them per motor, maybe 12 per prop or more.
• Unfortunately putting props anywhere except inline with the center of an aircraft causes disastrous unbalanced forces as soon as the plane is flying non-straight.
  • Two nacelles, one under each wing, does not work.
  • clipping 2 or 4 props inline in the center of the plane body works well. Have half of them spinning clockwise and half spinning anticlockwise.
• The best pitch for me was about 60. (90 being parallel to the direction of flight).
• Doubling up motors to get higher RPM isn't effective, it's basically as destructive as 2x time warp.

[KerikBalm]

Seems redundant with this thread (despite that thread getting a little sidetracked):

We're getting new prop parts, so this discussion will soon be obsolete.

For now you really need variable pitch props for good thrust throughout a wide speed range, up to high speed flight.

Due to low RPM, you really need to offset the lifting surfaces (so their radial velocity is higher for a given RPM)

To keep part count down, you need something bigger than a basic fin.

Observe the steep pitch here:

or here:

And shallow pitch here:

Variable pitch also helps when going to other planets:

^reentry configuration

And a configuration for gliding when prepping to land: