Published on March 3rd, 2014 | by Michael Barnard38
Airborne Wind Energy: It’s All Platypuses Instead Of Cheetahs
March 3rd, 2014 by Michael Barnard
Kite-based wind generation was first proposed in the 1940s, the seminal power potential paper was published in 1980 and it was first demonstrated in 1986. So why isn’t there a single production system or even an a quarter-scale production prototype in existence today?
Each of the combinations of design choices inherent in trying to capture wind energy with a tethered vehicle involves a different set of compromises and a different set of technical challenges. Each of these paths has been explored one or more times, and the results are not encouraging. Technical, safety, siting and regulatory challenges abound, and it’s unclear if they can all be solved.
This material has been garnered from publicly available documentation of the products, from the 600-page Springer airborne wind energy book published recently, the seminal Lloyd white paper from 1980, extensive interactions with airborne wind energy experts around earlier iterations of this material and material from organizations like the NREL. Very intelligent people have made these choices and are facing these challenges today.
So what are the choices?
- Soft kite, hard wing or lighter-than-air
- Generation on the ground or generation in the air
- Crosswind flying or (relatively) static flying
- Single tether vs multiple tethers
- High-altitude or low-altitude
Rotorcraft approaches are included in the hard wing option above. The purely blue-sky, free-flying, untethered generation concepts are excluded entirely from this structure; they have radically different weaknesses and few apparent strengths. Finally, some of these choices are on a continuum, not either-or choices, so a subset of compromises have larger impacts. As one example TwingTec uses an inflatable rigid beam to give structure to a rigid wing comprised of very soft materials. As another, many of the pure soft wings have inflatable spars adapted from kitesurfing, or directly use kitesurfing kites. (Note that these semi-rigid, inflatable technologies have their own compromises as they will require pressure sensors, pumps and in-airpower to maintain the correct pressures.)
Ten representative technologies and their choices are included, giving a reasonable cross-section of the space. None of these technologies are viable today when compared to conventional modern ground-based wind turbines or other forms of utility-scale generation, and many of these combinations may never be viable.
For more on these technologies, here are links to detailed assessments, or the company website if a detailed assessment hasn’t been published yet: Makani, Sky Windpower, Windlift, Altaeros, University of Delft, Kitegen’s Stem, Ampyx, Magenn (now defunct), TwingTec, Skysails.
1. Soft kite, hard wing or lighter-than-air?
Soft kites have advantages of being relatively cheap, being easy to prototype, higher crash survival rates and lower crash potential liabilities.
- Fabric is cheaper to transport and build with than solid materials.
- It’s fast to stitch together another kite, use kitesurfing kites or have a kitesurfing company build specific kites for you as UDelft does.
- When fabric hits the ground it crumples into fabric as opposed to stray shards of shattered wing. Combined with the slower speeds, this increases the likelihood that a fabric wing that crashes can be used again without problem. That said, inflated spars pop, wings rip and battens break or bend.
- When fabric hits something it crumples. Combined with lower speeds, the likelihood of causing significant damage is reduced substantially.
However, soft kites don’t fly as fast, don’t fly in as wide a range of conditions, don’t last as long before replacement and are hard — perhaps impossible — to reliably auto-launch.
- Lower speed reduces the potential power output for crosswind generation.
- Reduced viable flight conditions due to less structurally robust wings reduces the capacity factors annually because of weather-related grounding.
- Reduced life span reduces capacity factors annually due to maintenance-related grounding and increases costs due to regular replacement (balanced by lower cost of the fabric kites). As a data point, the highest quality paraglider wings must be replaced within 400 hours of flight time due to fabric degradation.
- The last one, auto-launch, is a real problem for soft kites. To launch, a soft kite has to be furled correctly and there has to be sufficient wind to loft it. Given turbulence near the ground and lower wind speeds near the ground, this is a non-trivial problem that no one has solved to date. The three most complete systems — Skysails, Wind Lift and Kitegen — all require manual intervention and stronger winds near the ground for launches. Kitegen has two or three ideas for auto-launching in lower ground winds, none of which appear to be viable.
Hard wings have the advantage of higher speed in the air and with that higher speed comes greater energy potential.
- Higher air speed comes with airfoil rigidity and reduced drag.
- Energy potential increases for onboard generation with turbine blades with the cube of velocity of air, so rigid wings such as Makani’s approach have significant potential advantages.
- Higher airspeed increases the aerodynamic lift vector of force substantially, which provides value for ground-based generation.
However, hard wings require onboard aileron controls, are more likely to break if they hit something and more likely to break the things that they hit.
- While soft wings can be controlled by dual tethers — although several are not –, hard wings require onboard power and intelligence to fly. This increases complexity and weight in the air. Complexity increases the chances of failure. Weight reduces efficiency.
- Hard wings, if they hit something such as the ground or tethers of other hard wings, are much more likely to suffer catastrophic failure. Failures which soft wings shrug off turn hard wings into splinters. This requires hard wings to have much more built-in crash avoidance, which increases initial engineering effort and costs.
- Hard wings, if they hit something such as people or buildings, are much more likely to cause significant damage. This increases liability risks, and hence insurance.
Lighter-than-air options lift turbines into high-wind zones. They have advantages of low-wind launch and higher safety.
- When the wind isn’t sufficient to loft a kite, a lighter-than-air device simply floats up into the air.
- When the wind drops, lighter-than-air devices just sit there. And when they suffer major failures, they typically drift down out of the sky like balloons, as opposed to crashing like planes or helicopters.
However, lighter-than-air devices are dependent on helium, and increase the likelihood of dragging tethers downwind for long distances.
- Helium is an increasingly expensive commodity, and lighter-than-air devices depend on it. Federal reserves are less dependable, and futures are higher. Prices have gone up significantly in the past few years, and while helium can be extracted or manufactured, there is little reason to believe it will be as cheap in the future. Hydrogen, while abundant, has the Hindenburg problem.
- While many airborne kite generators have passive features that will cause them to fall out of the sky quickly if the tether or wing fails, lighter-than-air craft will tend to float downwind dragging broken tethers unless features are designed into them to prevent this.
2. Generation on the ground or generation in the air?
Generation on the ground typically uses a regenerative winch on the ground that produces electricity when the tether is stripped off of it, and uses electricity to wind the airborne device back in. It has advantages of relative simplicity in the air, simpler engineering and generally lower liability.
- If all of the generation mechanics and electronics are on the ground, the in-air device can be much less complex. This reduces weight and failure points of the flying device.
- If all of the generation mechanics and electronics are on the ground, it’s much simpler to design the overall system for robustness.
- In general, if the generation is on the ground, then the tether can be a much lighter-weight, thinner and more common high-strength cable. This improves cost, weight, and liability concerns.
- A lighter airborne device with no rotor blades or heavy generation components is less likely to cause significant damage if it falls out of the sky, reducing liability concerns.
However, if the generation is on the ground, it has the disadvantage of high variability between the generation cycle pulling downwind and the retrieve cycle feathering upwind, launching and landing automation, reduced avoidance control options and added complexity of powering in-air systems. Solving for this problem creates significant complexity by itself.
- In order to generate electricity on the ground with an airborne device, the only proven method is to strip a tether off of a drum that has an electric motor. Reversing the motor in this manner generates electricity. The process of doing this requires a two-phase cycle. The first phase strips tether off of the drum as the kite flies downwind. The second phase retrieves the kite and tether in in a couple of low-drag ways. The generation phase produces much more electricity than the low-drag retrieve phase. The entire cycle can be one to five minutes in duration depending on device, but during every cycle power will go from peak generation to negative draw. That has to be smoothed as it’s not useful with that degree of short term variability in any utility-scale application.
- The options to smooth the generated electricity are storage, which adds a lot of cost as well as complexity, or a lot of airborne wind generators with managed generation cycles, a challenge to achieve outside of simulations given the complexity of winds aloft at different altitudes.
- As a result, one of these devices won’t generate electricity economically, but dozens of them are a challenging investment without substantial and mostly unproven gains.
- Launching and landing devices which depend on generation on the ground appears to be an unsolved problem as well. There are additional challenges with soft wings, but unpowered hard wings such as Ampyx have no proven or even well-prototyped automated launching and landing technologies equivalent to Makani’s VTOL approach. It’s relatively easy to have a human or team of humans position a hard or soft wing downwind at maximum tether distance and have it launched in an ‘automated’ fashion by the winch. It’s also moderately easy to have a hard wing land with minimal damage on soft fields in an ‘automated’ fashion. Having the wing automatically repositioned for launch has so far proven to be an intractable problem. Ampyx’ available documentation points out that powering the device into the air using a propellor is an option, but that this breaks the boundary of their weight requirements.
- Unpowered launch approaches also have to deal with winds aloft often being from different directions than winds on the ground, so neighbouring device tethers may easily intrude into the launch and initial flight path until devices get to the appropriate altitude. This potential for mid-air collisions due to varying wind directions does not appear to be addressed, as all solutions proposed focus on single devices, not arrays of devices. This is true for all devices, but more problematic for unpowered launch approaches especially with cross-wind generation. This is an unexplored problem in the current literature.
- If these devices were truly able to stay in the air for months at a time, this might not be a deal-breaker in terms of viability, but issues related to weather grounding and maintenance would appear to preclude that.
- The final compromise of ground-based generation is that all flying devices will need power in the air for at minimum aviation safety lights. Many solutions also require power for in-air control and communications mechanisms. Some require power for sensors and pumps for air-spar technologies. Batteries by themselves would introduce too many maintenance issues for grounding and replacement, so either small in-air generation — solar panels or micro-turbines — or conductive tethers will be required. Any solution will add weight and drag, reducing efficiencies, not to mention in-air complexity and maintenance requirements of their own.
Generation in the air typically involves lofting wind turbines on the wing or with a lighter than air device. This has advantages of putting a well-understood generation component into stronger winds.
- The only potentially viable in-air generation technology today is a small wind turbine. Makani puts eight of them on the leading edge of its hard wing as a primary example of the approach, while Altaeros puts one in the middle of toroidal lighter-than-air device. This trades off the squaring of potential energy due to swept area for the cubing of potential energy from higher wind velocity. The seminal whitepaper
However, if the generation is in the air, it increases in air complexity and potential liability substantially, likely increases maintenance downtime concerns and introduces the challenges of a conductive tether.
- Putting generation in the air of any significant size requires wind turbines and their associated electric generators, significant power management technology, more cabling and more onboard instrumentation and typically intelligence. With increased complexity and componentry comes increased weight which reduces the window of operation, along with significant cost increases to use lighter and stronger materials and miniaturized components.
- In any tethered generation system, the wing is almost always in the most stressful environment. Putting more components on the wing increases the likelihood of that component failing, so increases both the cost of engineering for failure and the maintenance downtime.
- Generation in the air, of course, requires a conductive tether to bring the generated electricity to ground. Conductive tethers are much heavier, thicker and more expensive than non-conductive tethers. Makani‘s specifications call for pultruded carbon fibre with an aluminum core, at a weight of 3,660 kilograms per 1060 meters. Expense is increased as is aerodynamic drag of the tether and tether noise.
3. Crosswind flying or (relatively) static flying?
Crosswind flying controls the wing so that it flies faster in figure eights or circles, and has the advantage of generating more power through greater aerodynamic lift via speed of the kite through the air.
- Potential energy of air is a function of the velocity of air cubed when turbines are on the device. This means that the faster the air is flowing over any device, the more energy that can be gained from it. When generation on the ground is used, aerodynamic lift increases generation potential there as well.
- A dominant strategy for increasing airspeed is flying kite devices through the air in figure eights or circular patterns, using the forward motion of the kite to increase aerodynamic lift. Whether generation is in the air or on the ground, this increases the total amount of energy to be harvested.
Crosswind flying has the disadvantage of requiring more sophisticated control mechanisms, dealing with tether tangling, tether drag, human injury risk and potential avian mortality impacts.
- Flying a tethered wing in circles or figure eights requires intelligence and devices to allow this to occur without failure and with relatively high efficiency. While experienced human kiters can fly figure eights and circles blindfolded, automation needs to be built to allow this, and more importantly to deal with failure conditions.
- Multiple-tether kites will end up with crossed tethers. Figure eights substantially minimize the impacts of this, but edge conditions can result in degraded performance or crashes.
- Tether drag is linear with surface area of the tether — the longer and thicker the tether, the greater the drag –, but squared with velocity of the tether. As kites scale up in size, speed and tether length (to harvest higher energy potential in high winds), tether drag quickly becomes a limiting factor. This is multiplied by multiple-tether systems obviously.
- Tether drag is also multiplied by basic air traffic safety requirements, which may be the largest issue for crosswind systems in most jurisdictions. Tethered devices typically require lighting or reflective marking of the tethers so that pilots can see them. Lighting devices and marking devices increase tether weight and drag substantially, making many cross wind devices non-viable. Makani’s draft submission to the US Federal Aviation Authority requests relief from this requirement and clearly states that their device won’t work if tether lighting and marking are required; it’s quite possible that this request will be refused.
- Crosswind flight drags very strong, very thin, very long tethers at high speeds. Mast concepts expect low attachment points to avoid unnecessary forces acting on the masts. They claim this as an advantage. However, this means that the high-speed tethers are moving low to the ground near the attachment point, and failure conditions would cause the tether to sweep across the ground. This creates a high-risk situation for people working in wind farms for whatever reason. It’s unlikely that this would create conditions amenable to secondary uses of the ground between wind generators. And it’s unlikely that this would be acceptable to workers’ safety and insurance organizations, requiring temporary shutdowns of numbers of the devices in order to service a single one of them. As collateral, tether length is typically the minimum safe spacing for generation devices, as Makani recognizes, but Kitegen does not in their publicly available literature.
- Avian mortality is an important and poorly considered issue with cross-wind generation schemes. They propose to put thin, strong, fast-moving tethers of hundreds to thousands of meters in the air. These tethers will fly through conical or bi-conical sections of the sky (conical for circular flight such as Makani, bi-conical for figure eight flight such as Kitegen) of hundreds of meters across at the widest and decreasing width and speed as the ground station is approached. This covers a much greater volume of space than traditional wind turbines with a much harder to see and avoid tether. Once you project this out to dozens of devices in a wind farm, it’s fairly easy to see how the potential for avian mortality increases substantially over ground based turbines. This is a serious problem with this approach that hasn’t begun to be assessed or addressed because they just don’t have any production units.
4. Single tether vs multiple tethers?
Single tethers are advantageous from a tether-simplicity perspective and from a tether drag perspective.
- One tether requires only a single winch on the ground, and eliminates tether tangle for single wings.
- Two tethers doubles tether drag. Three triples tether drag. As crosswind solutions are constrained by tether drag, this reduces a physical limit on speed and hence generation.
Single tether devices, however, require in-air control mechanisms and have reduced inherent safety.
- Control devices, whether bridle-sited line winches or aileron controls, must be near or on wings with single tethers. These devices require power, intelligence and communication with the ground station to operate. Power requires generation and power transformation or batteries in the air, adding weight, complexity and maintenance. Control intelligence requires fast-processing devices in power-efficient configurations doing sophisticated processing in their air. Communications with the ground station requires wireless or wired communication subsystems and power. All of these add complexity, cost and risk to the in-air device.
- Tether failure is a constant concern of kite generation devices. If the tether fails near the base, it can cause situations where it is dragged downwind for kilometres. Very strong tethers wrapped around vehicle axles would cause catastrophic crashes. Tethers, especially if conductive, that draped across power lines would cause problems of multiple types. Single tether devices have to rely on the strength of the tether and the passive failure modes of the tethered device to prevent these risks.
Multiple tether systems typically put all controls on the ground and have higher inherent safety.
- Dual tether crosswind systems typically put all of the controls on the ground in twin winches which makes the in-air device dumb as a box of hammers (and that’s a good thing in this case). The overall complexity of the solution is reduced substantially.
- The device is inherently safer as the likelihood of both very strong tethers failing simultaneously is much lower than just one failing, so the probability of the device flying downwind dragging the tether is lowered substantially.
However, multiple tether systems multiply tether drag and ground complexity.
- For crosswind systems, dual or triple tethers multiply tether drag by two or three times, limiting speed and hence generation to a greater degree. High-altitude systems are typically posited as single tether solutions.
- Coordinating two or more regenerative winches increases ground station complexity and cost.
5. High-altitude or low-altitude?
High-altitude solutions – four to nine kilometres — promise very high wind speed and hence energy.
- Potential energy increases with the cube of wind velocity for in-air generation approaches, and higher-altitude winds flow at very high speeds consistently. This makes napkin math high-altitude wind upsides very appealing.
High-altitude solutions require in-air generation and extremely long, heavy, conductive tethers (or other non-viable alternatives), and require flight exclusion zones up to passenger jet altitudes. And the jet stream varies.
- There is no projected solution that appears even close to viable that does not require in-air generation and its compromises.
- In air generation requires that the electricity get to the ground. The only mechanism which isn’t too lossy and dangerous to be viable is a conductive tether. However, calculations suggest that any conductive tether is too heavy to be lifted by any high-altitude device.
- Other non-viable alternatives involve EMF transmission of energy — frying anything that gets into the beam or anything on the ground if the beam moves, along with incredibly lossy transmission –, or laser transmission of energy – back to frying anything that gets into the beam. Both require heavy duty in-air technology of great weight and complexity.
- Nine kilometres is the altitude at which passenger jet aircraft fly. No-fly zones that high exist, but they are for dirigible-based radar systems of strategic national priority. It’s unlikely that no-fly zones would be approved for power generation when much simpler alternatives exist that don’t require exclusion zones.
- Solutions commonly reference the jet stream as a high-altitude, very energetic and consistent source of wind generation. However, the jet stream varies north, south, east and west by enormous margins over the course of the year and from year to year. No tethered device can guarantee access to the jet stream.
Low-altitude solutions have shorter and lighter tethers, even if they are conductive and in some seasons may access even stronger winds.
- Makani’s conductive tethers are merely thousands of kilograms in weight for 400 or 1000 meter tethers. This is supportable given the energy involved, although crosswind tether compromises apply.
- Night time low-level jet winds, which are already used by ground-based wind turbines in some places that are geographically favourable, are also potentially valuable resources more easily reached by airborne wind generators. In the US, night-time low-level jets are common in spring and summer on the Great Plains, which already have an excellent nearer-to-ground wind resource in most areas.
However, low-altitude solutions forego the power potential of high altitude winds and are not substantially advantageous compared to modern wind turbines.
- Virtually all airborne systems are now targeting under 2000 feet or 650 meters in altitude, in large part due to aviation regulations, but also due to massively multiplying engineering challenges with increasing altitudes.
- Wind velocity increases as altitude increases. However, the current 150 meter height of mast-based wind turbines capture winds at 80-90% of most low-altitude airborne systems. This means that low-altitude airborne systems provide marginal energy increases over much simpler ground-based devices. This marginal energy increase is trumped by the other compromises.
- Areas where low-level jet winds exist in the USA are also areas typically covered by farm fields. Per the point below about secondary ground use, this would eliminate siting of airborne wind generation devices in most areas where low-level jets exist, at least in the USA.
Additional Viability Concerns
This analysis compares airborne wind energy systems mostly to one another. However, there are four additional major compromises with airborne wind energy systems which are important to draw out to provide a complete picture.
- Maintenance of airborne devices is much higher than for ground-based turbines.This is in part for safety and liability management, but it’s also in large part due to greater complexity and fragility of devices that are able to fly. Rotorcraft, as one extreme, typically require four to five hours of maintenance for a single hour of flight. Business jets have a typical ratio of two to five hours maintenance per hour of flight time. Small fixed wing planes rarely exceed 75% serviceable days, and this ignores maintenance during those days which enables the planes to fly. Gliders have service lives of 6,000 hours, less than the number of hours in a year. Soft wings tear and degrade in sunlight, requiring frequently replacement, with paragliders requiring replacement after 400 hours as one example. Hard wings have relatively delicate control mechanisms which must be inspected and maintained. All of these maintenance activities reduce the capacity factor of airborne devices substantially.There is no reason to believe that tethered flying wings, whether soft or hard, will vastly exceed the lifespans or flight to maintenance ratios of every other flying device.
- Interaction of multiple tethered generation devices introduces direct challenges.Ground-based HAWTs must be spaced out to avoid degradation of performance due to wake effects. This is more a downwind than a crosswind concern, with six to ten rotor diameters downwind and four rotor diameters crosswind being typical spacing.Airborne wind generation devices have much more direct potential interactions. They can strike downwind devices, their tethers can wrap around each other and occasional cross-wind direction must be treated exactly the same as regular wind directions. Further, winds aloft are often at different directions and usually at higher velocities than wind at the ground.The combination means that cross-wind devices must be placed at minimum tether lengths plus a small safety factor apart. The differing wind directions means that devices rising up in ground winds could run into tethers rising across the wind as winds aloft are at 90 degrees or more off of winds near the ground. High-altitude devices could have tethers at multiple angles, and the much longer tethers could easily wrap around one another or strike flying devices.Makani understands this at least in part requiring full-tether length spacing between devices in all directions. Sky Windpower‘s high-altitude generation approach would have a veritable forest of cables at different angles and directions, making launching or landing one of their devices very risky, likely requiring very large spacing between devices. Kitegen does not address this, suggests that their Stem approach could be placed closer than tether length apart without explanation and ignore this issue with their Carousel approach entirely in publicly available material.This will limit takeoff and launch safe conditions at least part of the time for virtually every airborne wind generation device, and result in loss of generation capacity in other conditions to maintain safe spacing in the air.At present, this is poorly articulated in the literature, as most attempts are still focused on getting single devices working. For example, in the 611 page, Springer book Airborne Wind Energy published in 2013, the term “wind farm” is used three times, and references are to smoothing power generation and multi-altitude kite systems which multiply, not reduce, problems, and no assessment of multiple unit tether interactions is presented at all.Interaction of generators is an understated requirement space in airborne wind generation that, if useful individual generators can be made viable, will rapidly rise to the foreground. Airborne wind companies and potential investors would be advised to give this matter more thought in the near term.
- Weather groundings are a major compromise of all airborne wind generation systems.Airborne devices are much more fragile and usually will have to be landed in lightning storms, hail and high winds, then launched again. This process will be more frequent and take longer than equivalent occurrences for ground-based devices. And there is no evidence that airborne wind generation devices will even work in winter conditions, unlike ground-based turbines due to icing of the wings degrading ability to launch and fly. Despite this, proponents claim capacity factors of 85%-95%. Realism in calculating potential generated electricity appears low.
- Safety challenges eliminate secondary land uses which will drive up the cost of generation.Energy density of generation schemes is generally understood as an anti-renewables disinformation point, however it becomes much more relevant with airborne wind generation systems. Effectively, each airborne wind generation site is an airport with multiple devices in the air, high-strength tethers moving constantly and in many approaches at great speed close to the ground, and with devices landing and taking off regularly. The safety and operational requirements effectively prohibit other uses for the area used for generation, unlike for conventional ground based wind energy. For most approaches, much more of the ground-area around the devices must be cleared and smooth at least. This has direct economic impacts that are not factored into any attempted life cycle cost of electricity for these approaches that I’ve been able to identify.Ground-based wind generation requires a bit less than 1% of productive farmland including tracks, gates, transformers and wind turbine bases, and approaching 2% for ridge-based deployments, with very reasonably lease rates from landholders compared to the electricity generated. With greater safety setbacks and no secondary uses, airborne wind energy systems will likely have to lease all of the land that they are set upon, and in offshore configurations this eliminates all surface uses such as fishing, not just a small portion of those surface uses.
These factors, in addition to the significant design compromises outlined earlier, make airborne wind generation systems much less viable than ground-based solutions according to the very rough analysis matrix below. This analysis matrix is not deeply empirical, but my informed opinion on the comparisons between these devices and HAWTs. And of course the wind energy market is voting for ground-based, horizontal axis wind turbines with billions of dollars annually, while no airborne wind energy system has built and sold a single production generation unit.
Note: Sky Windpower is non-viable due to tether weight alone in its high-altitude incarnation; some people not directly involved with the company have said that they have abandoned that approach for the lower altitudes, but as of my full analysis their public documentation still claimed high-altitude. Magenn is non-viable because it’s now a defunct organization; the technology is not that much better or worse than most airborne systems.
The potential energy available in the wind flowing high above our heads is alluring, and harvesting it with tethered flying wings has great appeal, but as soon as you start engineering an airborne solution to harvest that energy, the compromises strip away the potential bit-by-bit until it just isn’t viable in any incarnation so far attempted. And it’s clear that many of the current organizations in the field were started at best with optimistic assessments issues regarding safety and aviation authority approvals.
Regarding the longer term, I’ve done a couple of thought exercises on materials changes and potential systems solutions. The technical solutions mapped out, and expressed in articles and on the AWE forum all appear to multiply complexity — lowering viability — or to increase one negative compromise as they decrease the impact somewhere else. Two rotating kite systems on a single tether is fraught with failure conditions. Lighter than air lift with a separate crosswind kite flying off of it reduces strength of tether force and multiplies complexities. Arch kite concepts are so complex and fragile that they don’t seem worth exploring further. Nano-material, ultra-thin tethers increase safety risks, increase winch engineering challenges and increase air safety regulation impacts in all likelihood.
It doesn’t mean that there isn’t a way out of the thicket to something viable. But all conceptual and prototyped solutions so far appear to multiply complexities and risks. There is no concinnity of design. It’s all platypuses instead of cheetahs.