Since all of the windup stations are slaved to the main windup reel, why would the landing operation with multiple lines take any longer?
The idea of multi line kites is not new, but āthe use of diagonal stays to control automatic launch and landā can be new if there is no prior art.
Have you tested it?
Some simpler and more reliable ways can be studied, comprising a light mast (a little like SySailsā mast), or a drone, or a ground station as wide as the kite in order to expand it before launchingā¦
As the kite flies in a definite place due to the wind direction, the unrolled ropes will not be the same length for the four stations, which imposes a differential control. If such a control is not perfectly coordinated the angle of attack and the bank angle of the kite will change during landing (and also launching). So landing could be likely longer, not to mention the risks of malfunctions resulting from the complexity.
1 Like
My concept of a restrained kite is not the same as a free flying kite. It will stay in a fixed place determined by the length of the tether and diagonal stays. If the wind changes, then the kite will lose lifting power and probably the diagonal stays will become slack on one side. We might be able to move the unwind stations to compensate for this but it might be better to land the kite, relocate the diagonal stays and relaunch. In all cases during launch and land the length of the stays are slaved to the length of the main tether/cable drive. It does not matter if some of the stays become slack because they will be corrected in the low wind region and the kite will be forced to land precisely on the lifting frame ready for the next launch.
I donāt see the aim to make the kite static? What is the benefit of that? Bring the kite back at each wind turnā¦ Ouch that hurt. Why so much complexity? Having slack lines means getting stuck in the trees or whatever. I donāt think lines will behave has you say and all will be simple to operate. Test you own stuff and see by yourself if it is good or not.
1 Like
As a first try why not add an extra turbine coaxial with the existing turbine. The connection can be either rigid or flexible with universal joints or a torsion cable. You will require a larger lifter kite to compensate for the increased turbine weight and the additional tension in the cable drive.
Gordon,
The standard wind engineering comparison to adding more blades or rotors is simply to add a few inches to turbine diameter, by longer blades, for equivalent extra-power. This usually seen as the cheaper simpler option.
Kiwee is intending scaling up by adopting a larger rotor, consistent with wind engineering norms. Thatās going to work at the intended scale, but with a noticeable square-cube loss of max power-to-weight. Further scaling in this design space would eventually require multiple rotors.
1 Like
HORIZONTAL CABLE DRIVE.pdf (64.9 KB)
When turbines are oriented to face the wind it occurred to me that a pulley transfer to cable drive is possible without the need of bevel gears, universal joints, or right angle belt drives. The concept is shown in the attached drawing. The generator drive wheel on the ground is horizontal and we require idler pulleys close to the drive wheel since the sag of the cable would result in premature wear on the cable due to rubbing on the walls of the pulley, especially on the low tension side. Since cable drive is most efficient at high speed, we require a larsger pulley at the turbines and a smaller pulley at the generator. To keep the cable lines parallel we require an additional idler pulley. Unlike the Kitewinder system, which has a floating generator, the horizontal generator can be permanently fixed to the ground. Changes in wind direction can be accommodated by rotating the idler pulleys in a circle around the generator pulley.
Alexander Bolonkin proposed a KiweeTwo system in 2004. This is shown in Project 1 of the attached article:
The only difference is that he uses a horizontal wing or an LTA balloon instead of a lifter kite. All the other elements are the same including a single turbine oriented to face the wind, connected to a cable drive which powers a generator on the ground. It is interesting that he predicted an energy cost of 0.37 cents/KWH for a 20 MW system. Unfortunately he provided no method of launching and landing.
It is my opinion that the only feasible method of automatically launching and landing this system is by using a very large restrained lifter kite. This is the only way we can fully automate an enlarged Kitewinder system.
The required lifting force of the kite is approximately equal to the thrust of the turbine or turbine system. The actual value depends on the values of the ground tether angle and the lifter kite tether angle. The attached analysis shows results for lifter kite angles of 45 deg and 55 deg.
BOLONKIN.pdf (166.5 KB)
The analysis shows that the required kite lifting force is much less for higher lifter kite angles, but at these high angles the force that the kite can generate is limited. The analysis also shows that a smaller lifter kite is required for low tether/cable drive angles. If we want to capitalize on thigh altitude winds, we will require a long cable drive. I have used a semi-empirical method to calculate the lifting force of a kite as a function of wind speed and kite area. I assume that this would apply to kites flying at a tether angle of 45ā°. Larger tether angles would provide less lifting force. The results are shown below:
KITE LIFTING FORCE.pdf (119.6 KB)
I have calculated the lifter kite size required for a 600KW system (Makani size), at various altitudes and at 10ā° and 30ā° ground tether angles. The results are shown below:
600KW ANALYSIS.pdf (136.6 KB)
The analysis shows two important things. Firstly the size of the lifter kite required is much smaller at high altitude due to the stronger winds and smaller turbine required. The size of the lifter kite is also less if a lower tether angle is used. Secondly, the weight of the tether becomes unacceptably high at high altitudes and this is worse for low tether angles. In my force analysis I neglected to account for the tether weight and therefore the values obtained in the 5 ā 10 Kilometer range is not valid.
1 Like
Why 1/2 tether weight, surely the lifter must lift the whole tether?
The bottom of the tether/cable drive is supported by the ground. This support is half the weight.
By the way, I used Bolonkinās values for the tether weight which is for a 22 MW system. I should have used considerably smaller values for tether weight.
The way I look at it, if there is no upward vector component to the tension on the tether from the ground station, the ground station could not be supporting the tetherās weight through tension, but rather the tether suspended mostly horizontally between the ground station and upper station could only be supported in tension against gravity by the upper station, by a large amount of mostly-horizontal tension of many times the mere weight of the tether. I tried to plow through some of Bolonkinās brainstorming paper, and observed many physically-dominant factors seemingly unrecognized by Bolonkin. There is a general fantasy flavor of writing by someone unfamiliar with real wind energy systemsā¦ I even saw what appeared to be an erroneous citation of the Betz coefficient as 0.67. Did I miss something there? The low projected initial costs, and low projected maintenance costs, as well as 10-year longevity he repeatedly cites, seem to have no factual basis, just wishful thinking. A very creative and active mind, but also a lot of incomplete analysis and aspects that, in my opinion, would net even work at all, due to major yet dominant factors Bolonkin does not seem aware of.
How can the tether weight be supported from the ground if it is not stiff? I propose the whole tether weight is supported by the lifter. The split of weight will be where the tether is horizontal, if it is sagging
The 600KW ANALYSIS.pdf mentions a 48671 mĀ² lifter kite and a 47.89 m diameter turbine, at 1 km altitude, 9.13 m/s wind speed, 1.11 air density.
Now let us use a 48671 mĀ² crosswind kite with a glide number of 4, by the yo-yo method. By using the usual formula, taking account of 0.65 of the power after the cubed cosine loss with an elevation angle of 30 degrees, then divide by two because of reeling-in phase, the result would be about 15.8 MW, so more than 25 times 600 kW, and that by removing the turbine.
So I donāt think such a system is really scalable, and all the more so since it is very difficult to pile up units (turbines donāt like to meet), both horizontally and vertically.
I donāt think there is any other realistic way to scale something like this up but by doing it in numbers. I suggest making a study of why SpaceX isnāt using a giant rocket engine for their rockets but developed the smaller Merlin and Raptor engines instead.
Here you have almost the exact same issues with R&D and manufacturing and reliability for example.
A rocket has a limiting factor in the area of its base, I think here a similar limiting factor is the perpendicular distance to the main tether (achieved by a rigid beam). The smaller you make that distance, the lighter your system can be. You canāt have your rigid beams be too long because that makes them too heavy, brittle, and expensive in time and materials.
How would you string your beads on the string?
1 Like
These two videos explain a lot about rocket engine design. The first explains about ideal nozzle size - smaller in the lower atmosphere, bigger in space:
This one compares the Merlin, Raptor, and other engines, why they were chosen:
1 Like
The specific engine used is outside the scope of my analogy. My point is about the size of the engines. A quick search about why smaller is better gives this old thread, Iām sure there are better sources: https://www.reddit.com/r/spacex/comments/3562s9/number_of_engines_tradeoffs/
The most difficult part of a rocket are arguably the rocket engines, like arguably the most difficult part of a wind turbine are the wind turbine blades.
Itās easier and cheaper to build and test a smaller engine. They must have tested to destruction hundreds of engines. They can then put that (Merlin) engine on the grasshopper and experiment on a smaller, cheaper system. Then reuse that engine on the Falcon 9 lower and upper stage, continuously getting data on it and getting better at making it in volume and increasing its reliability, while being funded by paying customers. Now theyāre repeating that process with the Raptor engine.
So R&D and manufacturing is quicker and cheaper. Then because you have more engines your rocket also becomes more reliable because even if a number of your engines fail, your rocket is not lost, unlike if an engine fails if you only have two engines.
You could go much deeper than that, but I couldnāt quickly find a good comprehensive discussion of the topic.
This guy in the videos made such a study, if you watch his very interesting videos.
BTW Iāve been advocating multiple smaller rotors, instead one one big rotor, for decades. One avenue of many possible avenues. So many possibilities in wind energy - weāve barely scratched the surface.
I donāt believe you can use a large lifter kite in yo-yo configuration. The forces are much too great. Check my numbers: A kite generating 2 X 15.8 MW unwinding at 3.0 met/sec (1/3 wind speed), will require 1.05 Million Kg of force on the tether. If the fibers have a permissible stress of 200 Kg/mm2, then the cross-sectional area of the tether is 5250 mm2. Diameter of tether is 81.8 mm. If specific density of the tether is 1800Kg/m3, and the tether length is 2 Kilometers, (tether angle is 30ā°) then the weight of the tether is 18,900 Kg! How do you automatically launch and land a kite and tether of this size?
I took this example to show the huge difference in efficiency between a power kite used in crosswind flight and in yo-yo mode, and a static kite of the same surface lifting a turbine of 47.89 m in diameter as you present it, and whose the weight could be about 4 tons.
SkySails already launches 160 mĀ² kites. 48 000 mĀ² is 300 times more, while being 12 000 times a 4 mĀ² static kite lifting a turbine such like a current system. So automatically launch a lifter kite + a turbine for 600 kW looks to be even further than launching a crosswind yo-yo kite of same area, and for about 15.8 MW.
A detail point: if a kite pulls 1005 tons it can easily lift a tether of 18.9 tons (1/53 of the pull).