The Tethered Uni-Rotor Network: Eternal Flight UAV | Hacker News Reader

The Tethered Uni-Rotor Network: Eternal Flight UAV

turnuav.com

119 pointsbennylope70 comments

Another interesting idea in this domain is to use wind power rather than solar power. Wind gradients can be harnessed through dynamic soaring although this can be a bit limiting. One way we might get around this is by flying two airplanes connected by a long tether.[0][1] One airplane flying at a different altitude where the wind is significantly different can be used like a kite sail to pull the other airplane. Although it has been shown to be a pretty nasty control problem at least for a human[2].

[0]https://www.nasa.gov/feature/flight-demonstration-of-novel-a... [1]https://www.nasa.gov/sites/default/files/atoms/files/14-2015... [2]https://arc.aiaa.org/doi/10.2514/6.2018-1492

I'm the originator of the TURN concept, and owner of the company. I can answer any questions you guys have about the vehicle.

Does it scale down? Could I use RC-scale technology, build my own, and use it to provide services to my own local network?

Sure, we all have the eye on the industrial horizon here.

But what would a "the streets have their own uses for things" application of the TURN concept look like?

For instance, I build RC planes for fun, and am a huge fan of Kline-Fogleman airfoils, for their simplicity if not economy .. and I can imagine a low-cost TURN implementation based on KF airfoils which might be more viable at the LoRa/WAN levels of scale. Something easily deployable, cheap/disposable/replaceable, on a much smaller scale of industry - yet widely available.

The video on the linked website mentions that smaller scales are potentially useful but implies they are not capable of indefinite flight.

(It mentions a unit with a five pound payload flying for around four hours, comparing it to a conventional platform with a similar on-board battery which was able to remain aloft for around an hour.)

How does it take off?

Can you generate lift from the tether. Maybe using the magnus effect?

How would it respond to gusts?

How strictly does it need to maintain angular velocity as it goes around?

What would the maximum speed in still air be? Some fraction of the rotor tip speed?

Does the tilt-wing configuration (presumably for landing and perhaps takeoff) set the constraint on payload mass fraction, as that will require more structure than the tension-stabilized rotary mode? Or is the payload mass fraction relatively insignificant compared to the battery and solar cell mass?

Safety: is there a way for it to gracefully fail? Transition from 4 blades to 3, and let the failed one dangle down as you descend? Would this be even remotely possible?

In steady wind, it needs to spin faster to maintain its wind robustness, which does consume more power... but that's common to all aircraft. Generally speaking, the it is robust to wind gust velocity about 25-30% of the wing speed.

It's more important that each wing maintain equal spacing between one another. But the angular rate of the entire system can be adjusted to accommodate the current operating condition.

I've evaluated models that can travel at 68 knots. Not terribly fast, but the intended application is to stay in one spot for as long as possible.

The tilt-wing configuration is actually dated. The L/D ratios of the wing would require the motors to be excessively oversized for takeoff and landing only. The design has been revised to have a single motor on the outboard section, and the system begins its rotation while on the ground prior to takeoff.

With three rotors it is still possible to stabilize the central hub, assuming the three can support the additional weight of the lost rotor arm. And with my controls background, I can't wait until I have more time to investigate this failure mode mitigation option!

Cool concept!

1. Any preliminary numbers for total weight and/or disk loading?

1a. How much weight budget is left for payload while still being past the break-even point, considering you'll need enough battery to weather nights and clouds?

2. What kind of numbers are you assuming for flux on the wings? In my experience these numbers are always supremely disappointing compared to theoretical maximum once you take into account angle of irradiance due to time of day, season and latitude, plus the unpredictable effects of cloud cover.

3. Is there any particular benefit to having each of the wing sections decoupled from the structure and attached by tether vs. just extending the airfoil and meeting at a hinge or universal joint?

4. Controllability during takeoff looks like it's going to be a shit show :) Any proof-of-concept results there? Does the h-stab span the entire wing section, or only just behind the props? If it's the former, does that mean the solar cells have to flex?

5. I don't know if the illustration right above the Flight Operation heading is representative, but if it is, I expect a pretty big drag hit from those tethers. (Brings me back to q3 - why not just make that more airfoil?)

6. How fast are we expecting this thing to spin?

1. The payload likes to stay around 5-10% of the total vehicle weight (otherwise the tether sags to steeply). I've generated sizing models from 5 pound payloads on up to 250 pound payloads for the Air Force SBIR research.

2. I have a look up table that plots energy capture for various latitudes vs day-of-year. It accounts for azimuth angle of the sun and duration of sunlight, where the integral of the area under the curve accounts for reduced collection at sunrise and sunset. To put it in perspective, operating on the worst winter day at 55 deg latitude is 15x harder than flying at the equator.

3. That produces triangular span loading (common to helicopters) which is not nearly as optimal as an elliptical span load distribution (common to gliders). Inboard sections just add weight and drag, without generating that much lift. The tether also has drag, but it's only 6% of the total system drag.

4. Agree. Takeoff has been completely revised and the transformational component has been abandoned. Now there is only a single motor on the outboard tip, and the system spins prior to takeoff. So the control laws for the retracted state are nearly identical to the extended state, just with a different set of gain values.

5. To build upon Q3, the tethers do have a very high Cd value (circular cross section is about 1.2), but they are extremely thin (small frontal area), and because the system rotates, the average velocity is 40% of the wingtip (which makes a huge deal for the V^2 in the drag equation).

6. At the largest scale, it takes nearly 40 seconds to make a full revolution. This slow rotation really helps to reduce the overall power requirements (P=VD).

Great answers, thanks for the insight! Responses to the responses:

2. Definitely squares with what I remember of the problem. Seems like accommodating the most extreme operating conditions would compromise the design so strongly that it's better to just constrain the envelope it's intended to work inside. If you can reach continuous operation at "reasonable" latitudes during a large portion of the year, that could still be a strong value proposition.

3+5: Gotcha, so I just wasn't appreciating how small the tether cross-section really is.

4. Ah, makes sense - that sounds much cleaner. 4 flying wings trying to do a VTOL while attached to a weight seemed a little crazy.

1+ 6. Wow! So this is one big boi. Pretty cool. I would have pictured a much faster rate of revolution.

Not sure I understand the images. The cables seem to be connected to the end of the wing units. A wing can't normally generate force longitudinally. So how is the wing generating cable tension?

Added: OK, it is all generated by the centripetal force caused by the mass of the wing units. So fast spinning and lots of cable/parasitic drag to get the wings reasonably level. Seems like a lot of wing area for a small payload. How is this better than just a wing large enough to support the payload with cables to distribute the load across the wing length?

Added2: This scheme still has the problem that the downwash from each wing will affect the other wings, just like in the helicopter case.

The wing generates lift, the engines on the edge of the wing generate thrust. That thrust is at a tangent to the circle the wings travel. Essentially the wings are pushed forward but cabled at the side.

Seems that the four wings rotate around the central unit creating centrifugal force that will keep the cables tight.

Wouldnt a single wing being buffetted by a downdraft or sudden turbulence cause catastrophic loss of tension between all wing tethers and the whole thing collapse?

What if instead of teathers, you just had two large circles of composit, as a hub and a rim, that were stiff enough to keep all wing positions but stiffer than a flimsy string - and thus keeping formation?

I didn't watch the video (or read most of the page!) so it's possible I'm missing something, but the images show the wings having a propeller at each end, so I don't think a loss of cable tension would be catastrophic (at least not in the same way it would be if there was only one propeller per wing). The wing would presumably keep flying straight until the cable was tense again and it all started rotating again. Seems fairly self-correcting, with the second propeller.

I'm a paraglider pilot, not an aerospace engineer, so take this with a lot of salt... but:

Paragliders can have asymmetric collapses of the airfoil due to turbulence, but usually it's no big deal (other than being a little scary). Catastrophic loss of tension is avoided because there's still airflow over the rest of the glider, which remains pressurized. All the pilot needs to do is maintain safe directional control until the full wing re-pressurizes. You feel a sudden loss of pressure and then an abrupt surge of pressure, then things normalize.

My guess is that something if similar happens here, it'll happen asymmetrically, and the props might keep airflow over the other 3 wings, which are still flying. They maintain stability while the wing that went floppy drops, regains its tension, and then resumes normal flight.

Last: I'd wager that they plan on deploying these in stable air, at very high altitude.

This is very interesting. I'm a little confused though. The page talks at length about eternal flight, and how the turn system is supposed to enable them. Then, in the "Smaller Scale Commercial Drone" section (which I imagine is a smaller scale version of TURN), it states that the endurance is only 4.5-7.5 hours. What does scale do, which enables eternal flight? Is it just a matter of having more power on board?

In addition, what are these eternal flight numbers based on in terms of latitude, and season? In the video, he states that they were looking at internal combustion @ 30 days. What about the solar approach? Would 18 hours a day of light be enough? 12? 6?

Can anyone here call out whether or not this looks like a viable thing?

They're only at the beginning of development, and they're starting at a small scale, so it's not clear whether the smaller scale drones can't achieve perpetual flight or whether that was simply not a necessary goal at this point to monetize the idea and get their business off the ground.

But, assuming that there's some point as you scale down at which perpetual flight becomes impossible, my first guess would be the surface area for solar cells at being a limiting factor.

You're right about the surface area. Altitude is the limiting factor. In the stratosphere, the air is thin enough to warrant large wings, which have lots of area for solar cells. I attempted to design a system for the Air Force at 10-15k feet, but the wings are just to small to collect any meaningful solar energy. That's why we shifted to an internal combustion engine design for that mission.

You're also right about monitizing the business. The smallest scale system is a necessary prototyping step towards eternal flight, but it also offers a compelling competitive advantage over conventional fixed-wing drones. VTOL plus 5x the flight endurance?... who wouldn't want that.

What size would the smaller versions competing with current fixed-wing drones be? You should post this to https://www.rcgroups.com/vtols-360/

The text mentions that the smaller version has no solar cells.

>Can anyone here call out whether or not this looks like a viable thing?

For the solar, truly eternal case (which isn't strictly possible because sooner or later you're going to lose a bearing or blow a capacitor or something), I think the limiting factor would be less on the aerodynamic side and more on the power-to-weight ratio you're able to achieve with solar.

A lot of that is going to depend on things like power system efficiency, latitude, and altitude which makes it hard to give a definitive yea or nay on whether this is feasible. I'm not really an expert in power systems but from an aerodynamic perspective it seems at least within the realm of possibility.

Like anything, the details of the implementation are going to make it or break it but there's not enough concrete info in the video.

You're right... there's no beating entropy. But the limiting case seems to be the battery charge/discharge cycles, which floats around the 500 range. So that's close to a year-and-a-half flight.

Other concepts are using tube-and-wing or flying wing embodiments. But long slender wings need more material to stiffen the slender structure. Just look at NASA/AeroVironment Helios, which ripped itself apart in a modest wind. Using centrifugal stiffening as a design element within the TURN concept eliminates the aerodynamic/structural tradeoff, and permits much better airfoils than standard practice will allow... nearly three times more efficient to be precise. By reducing that much structural material the TURN system carries much more battery mass. Most HALE aircraft cap out at about 20-25% battery mass, nearly 80% of the TURN vehicle weight is allocated for energy storage.

Are you involved with this project? From where are you sourcing your figures?

He said below that "I'm the originator of the TURN concept, and owner of the company. I can answer any questions you guys have about the vehicle."

Now outdated: Note for moderators (who I assume scan new comments): Some of this users comments have shown up dead, probably because he's a new user who made a bunch of comments quickly. It would be nice if you could come and revive them.

You can go to the comment permalink and click "vouch" to do that yourself.

I did... an hour ago... and they were still dead when I posted that. I figured a bit of manual vouching was in order.

Ah, then maybe me doing the same just now pushed the "score" over the threshold.

Also, emailing the mods directly using the Contact link in the footer is the most direct way to get their attention. They don't see all of the comments.

The research is using a spiral development process, where flight data from small scale prototypes are used to validate existing simulation and dynamic models, and then those models are used as design tools for the next largest embodiment. So right now, there are three different scale systems which serve different demographics.

The smallest scale system is purely battery power, but offers flight endurance well beyond what conventional fixed-wing can achieve. Traditional 10-ft wingspan drones carry 5-pounds for about 90 min. A comparable weight/payload TURN system can fly closer to 7 hours. This prototype is being used as a minimum viable product for an upcoming product launch.

The company was awarded an SBIR research grant from the Air Force which considered an internal combustion engine TURN embodiment. While not eternal fight, it again offers significantly extended flight endurance. The best research aircraft can fly a 250 pound payload, drawing 2000 watts of power for about five days. My research shows that an IC TURN system could remain aloft for over 30 days.

Finally, the largest scale system is striving for eternal flight while operating within the stratosphere. At 65k feet, the system is above most weather and commercial airliner traffic, and the air is thin enough to warrant a large wing fitted with solar panels. By getting the power requirements low enough, the energy collected during the day is enough to remain aloft throughout the night, thereby eliminating the need to land and refuel.

This is a really neat project. I noted in the discussion that by rotating with end tip motive power you eliminate the need for stiffening the wing, however doesn't this increase the need for better tensile strength along the wing? (in order to prevent it from pulling itself apart, something "regular" wings don't get a lot of stress on).

Also, in typical rotor craft the lift is highest on the outer edge of the rotor and least in the center where the airfoil speed is slowest. Does that affect where you put the payload? Is it on the edges of the wing or still in the center? At the center, if the wing is supported by the centripetal force of the wingtip motor's angular momentum, there is a huge torque in the middle if you pull it down. (much like pulling down on a suspended cable). How much deflection before you have the same problem as the stiffened wings of current efforts?

Increasing tensile strength is super easy compared to improving bending and twist resistance, especially when the aerodynamics not only constrains the thickness of the wing but also rewards high aspect ratios (the tip-to-tip length of a wing compared to its front/back dimension).

Imagine a wing that's 20+ times longer than it is deep, and is only 5% as thick is it is deep (so, for example, a 20m wide wing, that's 1m deep front-to-back, and 5cm thick top-to-bottom) - that's really hard to get stiff in bending along it's long axis, and in twisting stiffness around that long axis. This is why - as the article mentions, a sailplanes long thin wings account for 40+% of the airframe weight, when a stubby-winged but less aerodynamically efficient wing (like, say a Cessna 172) the wing might only account for 10-20%$ of the total airframe weight.

This TURN design minimises bending/twisting forces by replacing the load bearing and alignment forces with mostly end-to-end tension ones - which are much easier to resist (just load the structure up with "axial" carbon fibre...).

Considering the system is essentially a constellation of fixed-wing drones with added drag from tethers, how can it be so much more efficient than a single fixed-wing drone?

Very exciting project! Love the idea, and can't wait to see some proof of concepts eventually. It obviously has the potential to be revolutionary in the aerospace industry, though time will tell if it's as effective as claimed of course.

They compare the TURN concept to satellites in the beginning, mentioning only positives by comparison - of which there are indeed many (cheaper, better comms). One thing I didn't see mentioned was the failure mode between the two. A satellite that ceases to function will stay in orbit, until this orbit decays enough for it to burns in the upper atmosphere. If one of the rotors on this device fails, it will just come crashing down to earth.

And also, in typical neo-luddite fashion, I'm fearful for the surveillance applications of this device. Even if staying airborne isn't eventually completely free (eternal flight) with this system, the ability to stay up for 9 hours at a time on a few Li-Ion cells opens up a world of possibilities, not all of which are great. If NOAA can afford to put a number of these out in the Atlantic to get better atmospheric readings, surely cities, states, city-states, and others can afford to put a number of these above their populace.

However, progress is inevitable, and the technology can't always be blamed for it's potentially negative applications. I think this is an amazing project - but it will require some self-control to not let this push us further into a surveillance state.

Regarding your second point: https://www.airbus.com/defence/uav/zephyr.html

"The first unmanned aircraft of its kind to fly in the stratosphere, Zephyr harnesses the sun's rays, running exclusively on solar power, above the weather and conventional air traffic. It is a HAPS: a High Altitude Pseudo Satellite, able to fly for months at a time, combining the persistence of a satellite with the flexibility of a UAV."

This isn't a new idea, at all. You have to go back to the late 1780's to find a time before the use of balloons as recon platforms: https://en.wikipedia.org/wiki/History_of_military_ballooning

You'd have to go back before humans to find a time before facial recognition was used to track individuals, yet people get upset about when facial recognition is used to automatically track and record all of your movements through ubiquitous cameras.

There's a big difference between a person going up in a balloons for a few hour to scout a military locations and a swarm of drones constantly monitoring a city in peacetime.

I was not making a judgement about whether or not there's a difference between a single balloon and a drone swarm (there is), but addressing the comment:

> And also, in typical neo-luddite fashion, I'm fearful for the surveillance applications of this device. Even if staying airborne isn't eventually completely free (eternal flight) with this system, the ability to stay up for 9 hours at a time on a few Li-Ion cells opens up a world of possibilities, not all of which are great.

It looks a lot like you were saying these are similar. I'm not sure what else your comment could mean. What was your intended meaning?

Perhaps I just didn't understand the web page, but this seems unnecessarily complicated.

The Department of Customs and Border Protection already operates a fleet of tethered airships. These seem to be a much simpler solution to "eternal flight."

https://en.wikipedia.org/wiki/Tethered_Aerostat_Radar_System

It's not tethered to the ground. It's a group of rotating "wings" tethered to each other in the centre, wherein the centrifugal force of the rotation give structural stability. That way they get away with using less structural material, which means more of the weight can be batteries or fuel.

It probably wasn't a good choice for them to use the word "tethered" in the name.

It might not be incorrect, but tethered means you are connected to something. In this case, that something is the aircraft itself. Self is not the first thing most people are going to think of when they hear the word tethered. Especially since, as you point out, there are already aircraft that are tethered to the ground.

These plan to operate much higher at 60k feet and won't be tethered to the ground. The additional height will give them better coverage than the radar aerostats you mention. I think removing the tether might be required for them to be used a lot because the tether seems like it'd a) weigh a lot requiring much larger stats and b) if they're dense enough the tethers could pose a danger to air traffic (at the very least you'd have to route planes to avoid them). Both of those get worse as the stat operates higher and the forces on the cable get higher. It may become an issue where the increased strength required increases the cable weight beyond the point where an cabled aerostat becomes viable.

What's the current longest endurance, COTS (commercial off the shelf) flight platform?

It's not at all COTS, but the Rutan Voyager [1] spent 9 days in the air on it's round-the-world flight.

[1] https://en.wikipedia.org/wiki/Rutan_Voyager

Similarly not-COTS, but a series of UAVs, a Zephyr "7" made a record 336 hour flight - 14 days, with an unofficial runs of three minutes (!!) short of 26 days aloft for a recent Zephyr S.

https://en.wikipedia.org/wiki/Airbus_Zephyr

The Zephyr is using Lithium-Sulfur batteries, which have a better energy-to-weight ratio ("Currently the best Li–S batteries offer specific energies on the order of 500 W·h/kg" according to Wikipedia [1]).

These aren't commercially available though. Twice the performance of currently available automotive batteries is a major advantage for aircraft applications.

Pretty cool.

[1] https://en.wikipedia.org/wiki/Lithium%E2%80%93sulfur_battery

I wonder if the development of thinner, flexible solar panels and batteries would make a lighter than air UAV a better choice for eternal flight. That way it only needs to power rotors for positioning instead of maintaining flight.

DARPA already looked at that through ISIS. The problem is the massive volume of gas within the airship. Even the lightest winds require a substantial amount of power to travel or station-keep. Google is using weather balloons, but they don't have any active controls, so they drift aimlessly away from the equator, and then need to be manually picked up after about a month.

I have to wonder what the maximum translational velocity would be for these hovering platforms. How well could they deal with high winds? Gusts? How quickly could they change their position?

To make a comms network you'd need them relatively evenly spaced out to maintain coverage over a wide area. You'd also need them to have beefy and power hungry broadcast/receiver systems. How well can they scale with the solar energy?

Overall a very cool idea though!

The use-cases discussed in TFA all seem pretty low-power, mostly recon-type stuff. That said, it'd be interesting to contemplate what the right proportions would be to actually run a hungrier payload.

At a certain point, "perpetual" also starts to run into problems of mechanical failure. Then again, if these things are cheap enough and/or able to limp home in the case of a snapped cable or failed prop motor, maybe it still works out.

Seems like the video talked about wanting to make it into an internet communication network, which just doesn't seam feasible to me.

Then again, I'm just a programmer. I recognize I have no grasp of the power requirements for that kind of project.

The inexpensive nature of them seems a major draw. I wonder if they can detach cables and have the wings fly home on their own while having the center parachute down?

Limping home would be the easy part. It would be really impressive if the system had the ability to attach the tether of a replacement "blade" in flight, extending/replenishing the formation on the go.

But we are hardly able to keep conventional solar glider UAVs in the air, I suspect that adding the drag of a tether moves the goalposts far out of reach.

This seems like a really cool idea. What I can't figure out is the bootstrap process. How do you launch/start the wings rotating?

There's an image. https://static1.squarespace.com/static/5901f9f12994ca40051ae...

Basically, the whole thing takes off like a helicopter, before turning the wings on their sides.

I think there is merit for the last idea in the article - some kind tethered drone array as a windmill using something like autorotation:

https://en.wikipedia.org/wiki/Autorotation

Similar ideas have been suggested using balloons:

http://latesttechnology-world.blogspot.com/2013/07/magenn-ai...

https://www.nsf.gov/discoveries/disc_summ.jsp?cntn_id=134023

https://en.wikipedia.org/wiki/Airborne_wind_turbine

An array of wind-powered drones could scale well beyond what's possible with a traditional wind turbine, since most of the torque gets applied at the wingtips. There might be advantages with inconsistent wind gusts too, since the rotor could wind up in strong wind and unwind in weak wind. Startup costs would be negligible since a few drones could start the rotor and more could be added as needed. It might even by safer for birds if a drone array turns slower than a fixed rotor.

Not really sure how those ideas are all that similar to the TURN here. Those are long flying and involve tethering, but the flight systems are quite different and they're tethered to the ground.

Utterly Brilliant!

Something like this is sorely needed for bringing broadband to other parts of the world. Best of luck to Justin in this endeavor. I hope he succeeds wildly!

Is the cable portion essential? I'd think a semi-rigid structure (like they talk about with helicopter blades or glider wings) when under load would survive longer.

it'll be really interesting to see how this performs in extreme weather. i wonder if the cables would result in a much higher tolerance for shear, etc.

There's not much weather up at the planned operating altitude.

Er, use the Magnus effect. (Cf. "Magenn Power, Inc." company, now apparently defunct.)

https://en.wikipedia.org/wiki/Magnus_effect

I've been playing with geodesic cellular kites, and "eternal" flight is certainly possible. If the ratio of surface area to mass is great enough it's actually hard to stay on the ground. Cellular kites scale. I would like to eventually make flying buildings, etc...

Lol I just asked the same question without seeing your response. I was thinking hollow canvas tubes radiating out from a central hub, slowly unspooled in flight. Some mechanism would use the mechanism of rotation to draw air outward along the tubes to keep them inflated, combined with rotation induced from the central hub. No idea if that would work...

Sorry for the delay. I don't know if that would work either, but it sounds good. ;-)

Check out "Tensairity" https://en.wikipedia.org/wiki/Tensairity

clever design, but that video was awful. It came across as either ill-informed or misleading - and the beige - oh my, the beige.

Sorry about that. It's all I had to work with. The TURN concept recently won a local business pitch competition, with a video posted on their website. Slightly (just slightly) better production quality.

https://757pitch.org

Why not use a balloon? or dirigible or something like that?

Balloons?

Is this "solar freaking roadways", but for UAVs ?

I don't think that building (maybe) better solar glider UAVs falls in that category.

They basically combine the existing principles of control-surface based helicopter blade pitch control (e.g. Kaman K-Max) and blade tip propulsion, then drive them to the UAV-exclusive extreme where the individual blades gain complexity until they are a coordinated team of autonomous planes while the "body" of the helicopter along with all the complicated mechanical linkages to the blades is reduced to simple tethers between the blades and a central connection point where a stationary payload could be attached.

I suspect that they are overestimating the structural savings relative to something like the helios glider (where mass was already distributed along the wing) and underestimate tether drag.

The big difference is that a new type of aircraft won't fly at all if it's as bad an idea as solar roadways, whereas the latter could be built mile after mile until the world runs out of stupid money, no matter how bad it really is. If nothing else, the idea of "tethered uni-rotors" deserves a place in the canon of cool sci-to technologies that could theoretically work. They would fit in quite nicely in the future segment of Seveneves for example.

Reply on news.ycombinator.com