Well a completely new control architecture seems to have definitely helped today.
Most significant seems to have been the TTR (Tension Torque Ratio) and TSR live analysis and setpoints. Low wind performance seems to have gone way up. and a lot smoother. Will confirm after a lengthy delve into the data… I did manage to save a heap despite loosing loads of controller running stats arrrgggghhhh swears swears … and relax… was a good test
Well a completely new control architecture seems to have definitely helped today.
So if I never get any recognition, accolade or further successes…
At least I can say I was out-standing in my field today.
Actually you can see a kink in the main ring here. One of the tubes has popped out of the next tube cuff. The ring sleeve needs to be tightened to ensure better compression of the tubes… doh
Indeed doubling the diameter will allow increasing the swept area as the blades become further away from the centre and travel more space, that by keeping their initial length. Increasing their number is a cheaper mean to fill the higher swept area since there are already 3 blades, perhaps more. Thus the linear (but not angular) speed could be kept in a similar way as that of current wind turbines as they grow, leading (for current wind turbines) to a gearbox with more gearing as the angular speed decreases, or (for Daisy) to a sort of rim transmission.
By contrast increasing the chord or the length requires new and more expensive blades, and with respectively a lesser aspect ratio or a lesser relative efficiency (the root being far slower than the tip in comparison with initial blades).
Doubling the diameter is only a possibility for Daisy to scale up, knowing that high altitude winds are several hundreds meters above the ground. Daisy or other torque transfer systems like SuperTurbine ™ reach 10 or 20 m altitude. Going far higher without increasing the diameter looks to be a not trivial task. But is it really possible?
It is possible, just not very economic. Wubbo taught we can choose the AWES we prefer. I would like to see at least one high-altitude torque AWES. Yes it would take the greatest kite pros to make it work, if someone is willing to pay them (take over AWEu).
Just add enough pilot lift and guy lines, operate from bunkers, to make anything fly. Big Daisies and STs are worth testing, if not as cheap and handy as rope driven TRL9 COTS power kites.
All going well, I will test with another rotor on top today.
Have you considered testing a triangle structure as top ring? I know you’ve tested a triangle torsion transfer tube. No need to replace the tube, but just the ring with the propellers. Needs new 3D printed parts though. (Can do quickly in Fusion 360, if Rhino is making trouble) Bridle the triangle corners to three or so that when in operation all tethers are under tension. Not equal tension though due to trigonometry, except if you make a load distribution system.
Daisy progress with rigid blades seemed to go a step backwards on the practical side today.
We did however get the data from high wind today… Sorely lacking from a previous try.
@Ollie has managed to populate new areas on his charts, extending neat curves.
Plenty carbon rod and tube parts progressed into fragments today.
The video below shows the lift from the rotors. When they stop spinning they descend to the ground, while the lifting kite continues to fly.
Beside this I wonder if an accordion style technique could allow to bring the rotors closer for takeoff and landing then to expand them during the operation.
The rotors descend in this video because I pull the back line down.
So I was wrong. Thanks for the precision.
Like an autogyro parachute, the turbine should provide some lift? Have you calculated its theoretical lift or have you tried to measure it?
Today’s progress is printing out new fuselage prototypes to fit hexagonal tubes instead of bending them into a circle.
I will get around to some new force and operation explanation material as posted on twitter soon. Along with 1000000 other competing urgent jobs.
It’s not an easy thing bringing a new tech to market when cashless.
I’d put understanding that at in the top 5. It allows you to quickly evaluate different ideas and optimize things. Your system relies on tension after all. You and Oliver (and Christof) spent a little time on this already, surely you’ve put out some content on this already? Or if you haven’t you might be able to give some ballpark estimates?
Thanks for quoting me correctly Tallak. I should clarify: in reasonable, workable windspeeds, especially just for experimentation where you are there, you can get away with no overspeed protection. But if the winds pick up higher than expected, you are in trouble! The problem is, the cubic relation between windspeed and power output. A few more MPH changes from productive to destructive. This means you are balancing on a knife-edge between the most productive winds and destructive winds. So you mostly experiment in light or medium winds.
When you want to release a model that should be expected to run unattended and last for years, overspeed protection of some type is way more important than peak power output, since a machine rendered nonfunctional delivers zero power. Even 10 Watts is better than zero. For example this downsized (due to a crash) Sky Serpent
machine has no automatic overspeed protection, since it’s just experimental, and I can choose not to run it in very strong winds, but it does have a brake. Nice but not foolproof. Often such a brake will be overcome by a truly strong wind. The next step is an exploding machine, a burned-out generator, etc.
Still have to see which component weak spot overspeed will kill first on the current Daisy.
Was 5o’clock trailing edge bending the tube in the first rigid model… fixed with inner tip to centre alignment line in the next.
The fact it goes into even more tension is interesting but … to be prepared for bigger systems needing overspeed protection…
Yaw control using lift kite steering or backline side recovery for a TRPT / Daisy seems valid so far
But it will be a cumbersome operation to land a rotor this way in high wind and dependant on the lifter keeping flying well.
Hinging the blade pitch around TRPT connection will likely be fragile. Could maybe be based on polygon tension but looks heavy.
Active bridle reconfiguration solutions are all sorts of bother. But work in kite sports.
A compliant span folding wing hmmmm maybe.
A local weather warning activated turbine dropper … recover to base down the lift line… maybe.
Or just make a system so insanely cheap it doesn’t matter… but don’t litter.
I think the bigger you scale, the less you can rely on a lifter kite to supply the tension, and the less you can rely on centrifugal force. If you solely rely only on a lifter kite for tension, I think you’d need rigid elements out to your blades to resist the tension from your torque tube (and there’s limits on how big you can make those), otherwise there would be nothing to resist that. You’d still have the centrifugal force, but you’d like to lower weight and speed, and increase radius, which all reduce centrifugal force. The lifter kite can help with elevation angle.
I think you’re already relying on centrifugal force and lift from the blades to supply tension but you haven’t calculated/measured them?
I’m beginning to get a little more interested in this idea. If I were to try this though, I’d try to eliminate all rigid elements except for the blades and, as a start, have a single blade rotate around a lifter line. Something like that was posted before, but that could have been better executed.
Maybe I’m saying obvious or wrong things.
Yes you’re right. Daisy will have to rely less on lift kites above the turbine in larger models.
Once flying the lift kite is mostly used for aligning the turbine downwind and adds a bit of tension to the torque lines. Most of the tension in the torque lines comes from the rotary blades.
The turbine requires very little lift as it acts like an autogyro. I launch when there’s over 50N tension in the lift line… That’s tiny.
So as long as other AWES developers keep developing and scaling single line lift type kites (soft or rigid)… There’s a long way to scale.
I’m not planning to rely on centrifugal force for scaling. That could make launch very hard…
The turbine blades have a slight bank angle already (outer tip closer to the ground station)
This promotes aerodynamic expansion. Too much can however lead to heavy drag on the lower blades.
Eliminating rigid apart from the blades also makes launch hard… However, there will be ways to significantly reduce the mass to swept area once flying by expanding a given rig in the air.