Rotary MAWES concept: (rigid or soft) kites arranged into a (ground-controlled) bol kite suspended from a (flygen) carriage suspended from a (multiple-line) lifter kite (train)

The goal seems to be to make power from a tarp. Any curves in the tarp seem to be created by tether attachment points, not from trying to cut and sew back together different pieces of fabric. But I haven’t looked at that

Anyway, the kite is as simple as possible. I think that pretty much precludes doing anything other than flying circles under lifter kites. Otherwise you’re probably just doing vortex shedding, which doesn’t seem promising.

Assuming you don’t like torque transfer and assuming you want to put several pulleys on the ground, let’s say 3 so you can orient the system into any direction. No, using something other than a rotating ground station would tangle the tethers, so let’s do away with the 3 pulleys.



Here is the beginning of a probably dumb idea: something that looks something like the Pyramid (or just a number of tarps connected in such a way that they self-launch and then spin, something like a bol kite, but spinning much faster) but now some distance below the kites there is either the triangular bridle or a secondary triangular bridle. A carriage rides along this triangular bridle, or along a secondary line that goes around the outside of the triangular bridle (a single line comes from each main tether that then connects to a loop that goes around the triangular bridle and the tethers).

You could also add the secondary triangular bridle above the kites, lines coming from the top of the kites that then connect to a loop. That’s likely better, if you can prevent the kites from striking the line/cable from the carriage to the ground.

This carriage is lifted by a lifter kite. In the carriage there is a system that converts the motion of the line into either something for a generator or a rope drive to use. You’d probably start with the generator as that allows you to respond quicker, and so on.

1 Like

A good question might be where you would put the carriage and how that would affect the kites in the bol kite. You don’t want to compress them from the tension from the lifting line, and maybe you would like some bank angle variation to help the bol kite support itself more.

A good place I think might be on or close to the inside edge of the kites. Or maybe above the kites but the loop would have a lower radius than that of the inner wingtips of the kites. The kites currently at the top of the loop would be more horizontal, generating more lift upward, and the kites currently at the bottom of the loop would be more vertical, generating more lift downwind, would be the idea. For that the rope drive or cable would probably need to go through the inside of the bol kite though, so you can’t just use a swivel to connect the bol kite to a single line going to the ground, you’d need to connect it to something like a ring instead. On the outside of that ring would be two or more lines going to a rotating ground station, or rotating arms. The carriage would act like the lifting line attachment point and would align the elevation angle of the bol kite closer to that of the lifter kite.

1 Like

Or I don’t know about dumb really. The known difficulties would be:

  • You don’t have a design for a carriage yet, maybe you could take inspiration from ski lifts. The carriage would also cause wear on the loop so you would need to design the carriage to minimize that, and the loop would be need to replaced after some time of operation.
  • You’d have to design a bol kite that spins quickly, unlike the bol kites that we know of. Maybe you’d use rigid kites instead. But to use rigid kites, over some weight, you’d probably give them some help to start spinning. The Enerkite wing weight should probably still be fine I think. The generator/pulley in the carriage could act like a motor during spin up
  • The torque harvested by the carriage would have interesting effects deforming the individual kites in the bol kite and the bol kite itself, which I think rigid kites would more easily absorb with their higher flying speed and mass, and rigidity.
  • Managing the elevation angle and cyclic bank variation of the bol kite passively. That is needed so it generates some of its own upward lift, taking some or most of that off from the lifter kite.
  • Making a lifter kite train that is passively stable throughout all operational wind speeds. You would start by limiting the operational wind speeds.

You’d also need to design the ground station and so on.

I think making a stable lifter kite train and a fast-spinning bol kite are good things to focus on for Dave.

1 Like

The polygonal bridle could connect to tethers coming from the inner wing tips from the kites in the bol kite for example. The carriage could ride along this. The carriage could look like a cross, one rod being aligned with the lifting line / electrical cable axis, the other rod being aligned with the polygonal bridle. You would size the length of the rods so that the tension in the lines keeps the carriage aligned correctly.

You would add bridles to the kites that in turn get connected to the outside of the ring. You could do ground steering by adding control lines that go through the outside of the ring. The cable or rope drive coming from the carriage would go through the inside of the ring.

While the bol kite is not spinning, you could imagine it is probably draped from the carriage. Or not if the wind is blowing and the kites are light enough and the control lines are adjusted so that the kites have the right orientation, and then start spinning, with or without help from the motor in the carriage.

You could make a 2/3/4-line lifter kite train instead of single line so you could control it from the carriage. You’d do that probably as I don’t see a simple way to do ground control for the lifter kite train, and you probably need electronics in the carriage anyway to quickly vary the torque harvested with the generator.

2 Likes

For the carriage the lines from the polygonal bridle to the kites would be a problem. For that you could make the intake for the bridle in the carriage like a funnel, with a slot along the long axis, along the top side of the carriage, for this line. If the funnel were positioned right side up, the top of the funnel would not be a circle or horizontal but be cut at an angle or make an ellipse to gradually guide the line into the correct orientation.

The carriage would have a kind of U-shaped channel along the top for the polygonal bridle to rest inside. Wheels inside this channel would be clamped onto the bridle from either side, which would be powered by or power the bridle.

You would need to make it impossible for the line coming from the kites to go over the lifting line and carriage while the kites go under them, or get stuck on them. For that I think you could add another polygonal bridle that connects the inner wing tips together. That line should make sure the inner wing tip at least goes over them. Or maybe you’d just make the lines coming from the kites very short. Maybe that’s enough if the kite doesn’t have sharp corners. Maybe you could shape the carriage like (part of) something like an ovoid for example. Maybe you’d add ovoids to the lifting line and cable wherever a kite might hit it to guide it around them.

Before spinning up, it would be good if you could make the kites in the bol kite have a higher elevation angle than the lifter kite. Maybe you could do that by adding energy to them by reeling them in, or starting to rotate them with the carriage, or changing the bank and or pitch angle of the kites. Maybe you could let out the lifter kite train or steer it to a different position.

1 Like

It’s probably much simpler to suspend the carriage somewhere above a loop that goes around the kites. The loop then hooks onto the carriage from below. The loop can be attached wherever it makes sense, to tethers coming from the outer wing tips, to tethers coming from the top of the kites, or to tethers coming from the bridle lines for example.

The opening of the U-channel would still face mostly upwards or perhaps be parallel with the lifting line so that the loop can’t slip out. It probably curves from being in line with the tethers connected to the loop when the tether enters the channel, to upwards in the center, to down in line again. The channel itself perhaps also curves upwards in the center so that the pressure from the channel on the loop is more evenly distributed instead of concentrating at either ends of it. So instead of a rod with a straight U-channel it’s more likely to be something like an ovoid shape, with on its surface a path for the opening of the U-channel, with its changing angle. The ovoid further likely has a hinge connection to the rod for the lifting line to make it lie on the rotation plane of the loop, something like a rod attached to the center of the ovoid connects to a hinge on the lifting line rod.

You’d probably attach smooth rods along the side of the U-channel that the tethers rub against to reduce contact area, friction, and wear on the tethers. And perhaps you could just use pulleys in the bottom of the channel instead of clamping the loop between wheels. Or if you do need to clamp the loop, the tethers could perhaps lift the clamps when they go through.

You’d perhaps control the bank and pitch angle variation of the kites with ground control instead of passively.

Now you’d probably add a tower for the electrical cable or maybe rope drive to increase the distance between that going to the carriage and the rotating tethers going to the ring.



Or if you do want the cable or rope drive to go through the inside of the ring and with that eliminate the extra loop around the outside of the kites and the extra tower and so on, you could only extend the lifting line rod upward and not also downward from the hinge for the ovoid and make it so that the cable or rope drive exits the carriage somewhere below where the kites are likely to impact it. There are lots of ways to do that. You’d probably start by not extending the lifting line rod beyond, below, the hinge and then changing the geometry of both this rod and the ovoid. At extremities of these you’d probably have one or two, or more, pulleys for lifting lines, an attachment point for the hinge, and an attachment point for the cable or a pulley for the rope drive.

1 Like

Some older ideas:

And some newer ideas:

Another carousel idea somewhat inspired by the nodes / towers in this comment: Economical assessment of the scalability of the ground station in pumping mode - #5 by Windy_Skies

Now towers are arranged in a circle or ellipse. The carriages are mounted on top of the towers. The idea now is on how the tethers are arranged. You could have 3 loops: one would go through the carriages, connected to this like before with lines is a lower radius loop, with a radius at least half a wingspan lower than which the towers make. The outer wings of the kites could rest on this during rest. Connected to this is another loop with an even lower radius for the inner wings of the kites to rest on. These two shorter loops hook over and around the carriages and towers. These two loops make a perch for kites to sit on, or mounting points for perches for the kites.

The perch could be something like a rod between the two loops with fairlead(s) for the tether(s) to go through. These tethers could go to a ground station in the center of the circle, letting them in and out. The carriages could drive generators in the towers or via rope drives drive a generator in the central ground station. Or it could make a yoyo system for example.

The carousel would be horizontal. Some thoughts from this comment on cams for example could be relevant: Claim: Kites in a MAWES need to be able to: CLAMP onto the tether and LATCH onto the previous kite + a discussion on how to do this + a first mention of stackable kites + a concept for automatic launch of a soft kite train - #67 by Windy_Skies

Arranging the towers into an ellipse and putting it on a ship sounds like a fun idea.

Or having the towered carriages as a set of foiling ships with hydro regen which take the squishy diamond course around a central buoy.
(The buoy being far more upwind than central as the main pull balance tends downwind)
The advantage being you can use the 2 opposing yacht tethers to wrap around a the buoy/ capstan as a complimentary winching to assist with upwind speed using downwind travelling.
You’d want to coordinate the kite yacht tacking across the 2 opposing sides

1 Like

Somewhat like Kite Power Systems (KPS) but with ships with hydro generators instead? If you’re using hydro generators you don’t really need to travel upwind, just like with flygen where the tether length is mostly fixed? What would be the efficiency and expected power output of using hydro generators vs powering the drum on the buoy?

Another idea would be to use buoys instead of towers and maybe use ships, but then probably a problem is that the ships would want to always be downwind of the buoys. Maybe that’s solvable by connecting them in a lower radius loop and letting them go around the inside of the buoys for example, or not connecting them and doing something else and letting them go around the outside of the buoys. Or maybe have the buoys be submerged instead so the ships can go both inside and outside of them.




If the loop that went through the carriages were a rigid ring instead, or if the carriages were mounted on a rigid ring, it wouldn’t necessarily be, you could let the ring lift the carriages. The carriages now being connected to the towers with extending and retracting lines Battle Angel Alita:

If the loop is fully tensile you’d have to find a way so that the carriages stay in relatively the same position no matter if there is currently a kite train overhead or not.

You’d perhaps do something inspired by space nets. You’d still need the two inner loops as mounting points for the kite perch, but now between those and the outer loop that goes through the carriages you’d perhaps add (additional loops) with bridling between them so that the loop going through the carriages has more bridle lines going to it. The problem with that is that if the bridling doesn’t change in length it would be slack during rest, everything would droop and fall to the ground during rest when not supported by the kites, like a tent without its tent poles. You’d also have the extra drag from the extra (loops and) bridle lines.

Or maybe only the loops would be slack. The bridle lines and the kite tether would be suspended between the ground station and the loop going through the carriages. That would be a long span though. I don’t know what that would look like and if it would be better to have the loops or not.



In an offshore system you could move the central buoy away from the center of the circle. I an onshore system you could put the ground station on a circular/elliptical track, inside and concentric with the towers, the radius optimized for the desired elevation angle. You’d move the ground station based on the wind direction.

The issue I was trying to address with the “central” buoy is sharing a tension reference to it as an anchor so that we don’t get bunched (carriages/yachts/ground terminals) bunching on the upwind leg of a course

It smooths course flow for multiple elements around the course

Linking the enhanced pull downwind to pull the upwind moving body faster than it would travel alone

Overcoming the potential line drag in water is hard though

1 Like