Hi Gordon,

          I like it. But from the sketch it looks like the cord that pulls the launch system into launch position will also prevent the launch system from pivoting when the direction of the wind changes.

PeterS

Hi Gordon,

          If the wind direction changes 30 degrees (and gusts come from a different direction up to 30 degrees too, which means the change in direction could then equal 60 degrees during landing), the kite tether will place an enormous leverage on the kite end of the lever arm, thus bending it or snapping it off at the bottom hinge point. If it is made strong enough to withstand that bending/twisting force, then the kite itself will land mostly off of the platform. So then you have to make the platform much bigger to prevent that.

If you want to accommodate changes in the wind direction up to 90 degrees, then the landing platform will need to be a large circle so as to also anticipate wind gusts during landing. But the circular platform will be tipped at a large angle that won’t work because the kite may become snagged on the platform.

So then you need to add a tipping and pivoting mechanism to the circular platform to align it properly with the direction of the kite. At that point, it begins to look like a different basic design is called for. Your design is OK for launching, and if the wind direction doesn’t change substantially. But otherwise, it is not going to provide for reliable retrieval.

          Storms often have rotation to them, which can greatly alter the wind direction a couple of times within a relatively short period of time. How much labor expense will your device require to reorient it to accommodate those changes in wind direction? How often will that labor be required? Landing and relaunching a kite when the wind is strong, as during a storm, is the worst time to do it because of great deal of energy will be lost.

          So is your launching mechanism intended only for places with a stable wind direction? If so, then it should work as intended. If not, then it might end of costing too much for labor and lost energy.

But I’m talking about energy kites. If you are not, then there is no problem with your lever arm.

I don’t fly kites much at all. So maybe I don’t know what I’m talking about. You can be the judge.

PeterS

Hi Doug,

          I’ve used curved rods to transmit torque. I found that they create very high friction losses. How do you intend to avoid that problem while still transmitting high torque produced by a lot of rotors along the curved shaft?

          And what about the extreme cantilever of the shaft and its extreme pressure on the bearings?

PeterS

Hi Gordon,

          You are right. I did not understand. That is because your drawing neither showed nor mentioned additional kite restraints. You expected viewers, like me, to keep in mind the principles and details of the overall system you are working on. That is expecting a bit too much of my feeble brain.

PeterS

Hi Dave,

          I’m not sure, but my guess is that the power to rotate a Magnus effect cylinder increases by the square of the surface speed. That is because the surface acts to drag air with it, and drag is proportional to the square of the velocity. So for a given spin ratio, the lift and drag of the balloon increase by the square of the apparent wind speed, and the power to rotate the balloon will also increase by the square of the wind speed.

          There is no apparent reason for a scaling limit. Where there will be a limit is on the aspect ratio of the rotor. If it is too long and skinny, it will bend more easily.

          Magnus effect kites are just beginning to be explored. Most rotor kites, such as Savonius rotor kites, use the Kramer effect. They use a spin ratio of about 1 and are auto-rotating. So it’s too soon to draw conclusions about the limits or potential of Magnus effect kites.

          Hydrogen will, of necessity, be used eventually as the buoyancy gas.

PeterS

Hi Doug,

          The type of friction I am referring to probably has a specific name, but I don’t know what it is because it is usually avoided in the first place. Let’s call it “internal friction due to rotating a bent rod”. I’ll describe it again, but in a different way.

Take a long thin rod (perhaps spring-steel music wire, or a long piece of thin PVC pipe) that can be bent 90 degrees without breaking or deforming. Put a ball-bearing on both ends of the rod. The rod is still straight. Mount the ball-bearings in vices or the equivalent. Near one of the bearings, wrap a string around the rod and tie a small weight to the free end of the string. When the weight is allowed to drop, it will pull the string and spin the rod. Reduce the weight until it is just enough to spin the rod. That is your baseline measurement for how much torque is required to spin the rod when it is straight.

          Now bend the rod 90 degrees (or 45 degrees, or whatever you like) and again hold the two end bearings in a vice or the equivalent. Repeat the string and weight procedure as before.  What you should find is that the weight must be much larger to produce enough torque to rotate the bent rod. 

          The greater the degrees of the bend, the more torque will be required to rotate the rod. I assume that the internal friction increases linearly in proportion to the degrees of bending, but I don’t know.

          If you do the experiment, I would love to hear your findings.

The reason for the increased torque resistance is the internal friction of the bent rod. If you attach an electric drill to one end of the bent rod and run it for a while, you will feel the rod get warm due to that internal friction.

          When a metal rod is simply bent, it produces internal friction and heat. When a bent rod is rotated, that produces a continual bending of the rod because it is undergoing compressing and stretching at every point along its surface. So it’s just another way of bending the rod.

          There is probably a mathematical formula for determining the internal friction that includes the amount of curvature per unit length relative to the diameter of the rod. Also, the stiffer the material, the greater the internal friction.

          I am not aware of any material that can avoid that type of friction. But some complex structures may be able to reduce the friction. Wood dowels may generate less heat because wood is designed by nature to flex a huge number of times without causing fatigue. But even wood is not designed to both bend and rotate at the same time.

          Stiff tubing should have the same type of internal friction as rods because most of the internal friction for both rods and tubes is near the surface where the greatest compression and stretching occurs.

          Inflatable tubes, if they can be made stiff enough, might have less internal friction when they both curved and rotating. I don’t know.

          One solution to the problem would be to separate the bending and rotation functions. I think that it could be done in a number of different ways so as to allow rotors to be mounted on straight rod sections that rotated but didn’t bend. What I envision is a long or plank that only bends, with bearings mounted on its upper side and far enough from the plank to allow clearance for rotor blades. Then the short shafts with the rotors on them would be joined using universal joints to other sections of shaft that went through the bearings. It sounds heavy, complex, and expensive. But maybe it can be improved. At least we know there is a way to do it that can minimize rotational friction while bending.

PeterS