Designing, Building and Testing a Darivonius VAWT

[Ungrounded Lightning Rod]

The maximum Cp is realised if the wind speed in the rotor plane is reduced to 2/3 of the undisturbed wind speed V. This can be realised by a few blades or by many blades and for a high tip speed ratio or for a low tip speed ratio.

I think you typoed, Adrian.  Shouldn't that be "reduced to 1/3" or "reduced by 2/3"?

The derivation of the Betz limit is given in chapter 4.2 of my public report KD 35. The wind speed far before the rotor is called V. The wake around the rotor expands and the maximum power is generated if the wake expands such that the wind speed in the rotor plane is 2/3 V and that the wind speed far behind the rotor is 1/3 V. The expansion of the wake is very fluent, so there is no sudden jump at the rotor plane. The theoretical maximum Cp is 16/27 = 0.59 but the practical Cp is much lower because of four effects which are explained in chapter 4.3 of KD 35. The derivation of the Betz limit is valid for any device which extracs energy from the wind so for a HAWT and for a VAWT. For before and far behind the rotor means theoretically at infinitive distance but in practice you can take a distance of about three rotor diameters.

Oh, right.  Betz assumes an infinitely thin rotor and all the speed delta is in the spreading upstream and downstream.  Half on each side, so it's 2/3 V at the rotor.  A Darrieus is thick enough that some of the spreading happens between the upwind and downwind transits - but far from enough to treat it as two separate machines.

So it's arcsin((1 / ((2/3)TSR)) (a little higher on the upwind transit, lower on the downwind, but ignore that).  TSR 6 would swing about +-14.5 degrees.

So the angle alpha for a tip speed ratio of 6 is found from tg alpha = 2/3 / 6 which results in alpha = 6.3 degrees for a tip speed ratio of 6.

Yeah, I did it that way at first, too.  But when Wolfram Alpha said the swing angle got smaller with lower TSR I said "What?" and found the error.  The 2/3 goes below the slash, not above it.

How does tg alpha = 14.48 (or somwhat more on the downwind transit), rather than 6.379, affect your discussion of NACA airfoil selection?

[Adriaan Kragten]

In my public report KD 601 I give the formulas to determine the geometry for a H-Darrieus rotor. The formulas are derived from the formulas for a HAWT as given in KD 35. I have found that the optimum tip speed ratio is about 4.2 because the angle of attack alpha is 9 degrees at the front and at the back side. Formula 2 gives the angle phi and so also the angle alpha as the angle beta is zero. Formula 4 gives the chord c for a certain rotor radius R, a certain angle phi, a certain number of blades B and a certain lift coefficient Cl. I have calculated that c = 0.203 m rounded to 0.2 m = 200 mm for the chosen parameters. Formula 5 gives the Reynolds value and I have calculated that Re = 2.84 * 10^5 for a wind speed  V = 5 m/s. This is just high enough to prevent stalling.

Next assume that you don't agree with my conclusion that lambda = 4.2 is the optimum tip speed ratio and that you want to use lamba = 6. So you have have to play with the formulas and see what changes you get. Substitution of lambda = 6 in formula 2 gives that phi = alpha = 6.3402 degrees. In figure 1 you can see that this belongs about to Cl = 0.62. Substitution of R = 1 m, phi = 6.3402 degrees, B = 3 m and Cl = 0.62 in formula 4 gives that c = 0.138 m = 138 mm. Substitution of V = 5 m/s, c = 0.138 m and lambda = 6 in formula 5 gives that Re = 2.78 * 10^5. This is only a little lower than for the original design so the calculated geometry seems O.K.

However, if the chord is reduced from 200 mm up to 138 mm, so with a factor 0.69, the bending strength of the blade is reduced by a factor 0.69^3 = 0.3285 and the bending stiffness is reduced by a factor 0.60^4 = 0.2267. The areodynamic forces acting on a blade will be the same, so the blade will bend out about a factor 4.4 more and this might result in large vibration problems as the direction of the lift force with respect to the blade changes 180 degrees during one revolution. Another problem is that the rotational speed increases by a factor 1.45 and the centrifugal force increases with n^2 so with a factor 2.1 if the mass would be the same. The reduction of the blade mass will compensate this but still I believe that increase of the tip speed ratio is dangerous, especially I no proper braking system is used.

In you don't agree with my formula 2 for the angle phi, then look in figure 2 of KD 601 for the postion 1. There you can see that W can be determined as tg phi = 2/3 / lambda.

Next assume that you don't agree with my conclusion that lambda = 4.2 is the optimum tip speed ratio and that you want to use lamba = 3. Substitution of lambda = 3 in formula 2 gives that phi = alpha = 12.5288 degrees. Figure 1 gives the Cl-alpha curves for different values of Reynolds. Assume Re = 3.31 * 10^5. In figure 1 you can see that this belongs about to Cl = 1. Substitution of R = 1 m, phi = 12.5288 degrees, B = 3 m and Cl = 1 in formula 4 gives that c = 0.332 m = 332 mm. Substitution of V = 5 m/s, c = 0.332 m and lambda = 3 in formula 5 gives that Re = 3.40 * 10^5. This is a little higher than for the original design and also a little higher than the value Re = 3.31 * 10^5 which was used to calculate the Cl value, so the calculated geometry seems O.K.

However, a chord of 332 mm gives a very big blade and if you look in the Cl-alpha curve for Re = 3.31 you see that a value Cl = 1 at alpha = 12.5288 degrees belongs just to the peak of the Cl-alpha curve. So the airfoil is just at the point of stalling. This means that it will really stall if the wind speed is only a little lower than 5 m/s. Once it is stalling, it may take a long time to cancell the stalling so stalling may continue even if the angle of attack becomes smaller at other positions of the blade. So this indicates that lambda = 3 is too low. So the real optimum lambda will lie somewhere in between 6 and 3 and my choice of 4.2 is about the optimum for a rotor diameter of 2 m

[MagnetJuice]

Very interesting and helpful discussions in the last few posts. Thank you ULR for jumping in, very much appreciated.

The Report mentioned by Adriaan, R-443-D, Catalogue of Aerodynamic Characteristics of Airfoils by A Hageman can be downloaded from Damon's website here:

http://www.earth.org.uk/catalogue-of-aerodynamic-characteristics-of-airfoils.html

I have spent the last few days researching parts for the physical construction of the VAWT. As much as possible, I have tried to stay within the stated goals of the parts being easy to obtain and inexpensive.

For the center column, initially I wanted to use a solid shaft, but because of the high cost and the weight of the solid shaft, I am now thinking that a heavy wall steel pipe should be strong enough. The center column doesn't have to be super strong because I plan to have a support frame around the turbine and the column will be supported by bearings at the bottom and at the top. The diameter of the pipe is 1.33 inches (34 mm). For those that want to build a turbine without a support frame, the solid shaft could be a better choice.

For the bottom bearing, I plan to use a trailer hub that has tapered roller bearings. The top bearing could be any bearing because all it does is hold the center column in place.

The radius of the rotor will be about 24.5 inches (622 mm). The height of the turbine could be anywhere between 70 and 84 inches (1.78 to 2.1 m). The exact height can be determined if there are advantages of having a certain dimension.

For the blades for the center Savonius-like device, I can use either thin sheets of Polypropylene or Gray PVC. Both of those thermoplastics are easy to bend into a curve and are strong and inexpensive. The PVC is better than the Polypropylene for very low temperatures, -22c vs. 0c.

My original plan was to use the center Savonius to control the high speed limit of the turbine, but it appears that it will not be possible to use the Savonius for speed control. The only function of the Savonius will be to help start the turbine in low winds. By the way, the TSR of this Savonius could be around 1.3, so consider this when doing calculations.

It is OK if I cannot use the Savonius for speed control, because I have been working on the design of a simple magnetic brake that can slow down the turbine at a predetermined speed. This brake can be set to activate at any high speed, for example, when the wind speed reaches 30 or 35 MPH (13 to 16 m/s). I will do extensive testing of this braking system and if it is effective, I will not get a patent, I will post the operational and construction details on this website for the benefit of everyone.

I am very pleased with the contributions that I have received from everyone so far, I am confident that this VAWT will be a success.

I have already figured out how to put together the complete turbine and how to connect the blades to the center column using a horizontal strut that will minimize drag. What remains now is the most critical part in my estimation. This is what lies ahead:

•   Determine the blade profile to be used, whether symmetrical or with a small camber.
•   Determine the blades aspect ratio.
•   Juggling the Reynolds number with the solidity of the turbine and the aspect ratio of the blade to obtain a TSR and speed that will extract the maximum power from the available wind.

Those are not easy tasks, but judging by the input that I have received so far and seeing the talent that is here, I am fully convinced that it is doable.

I am scheduled to have surgery tomorrow morning, so if I don't show up here again, it will be because the surgeon sneezed and cut the wrong thing. 

For those of you that believe in prayer, I will appreciate you keeping me in your prayers. For those of you that don't believe in prayer, a wish of good luck will be welcomed.

Thank you,

Ed

[MattM]

Good luck on your surgery.

Does your Savonius need to be connected at all times to your Darrieus?  I'm thinking of how pull starts often work on small engines.  If the Savonius only needs to start your Darrieus then could you use a simple slipper clutch so the Darrieus is able to get it's push start, but then spins unimpeded?

They also make something called one-way bearings.  The purpose built clutch using the same principle is called a sprag.  They can be quite inexpensive for what they do.

[MagnetJuice]

Thank you for your best wishes Matt.

If I set my mind to it, I could probably design a system that would disengage the Savonius once the Darrieus is able to turn by itself. Frankly, I don't see the usefulness of that, because once the Darrieus gets turning, if the wind is strong enough, the Darrieus takes over and the Savonius will just go for a ride with it.

A mechanism like that would add complexity to the design and would not be compatible with the design goals of being simple to build and inexpensive. What we have to make sure is that the size and shape of the Savonius will not interfere with the rotation of the Darrieus and keep it from reaching its maximum speed and optimal TSR.

Thanks and keep thinking outside the box.

Ed

[MagnetJuice]

The surgery went well. Now, for the next few days, I get to sit in front of the computer screen a little longer than normal. That's ok, I get more time to learn about Reynolds numbers and solidity.

I,m sipping a large glass of red wine to replace the lost blood. I didn't lose that much, just wanted the wine 

[topspeed]

Speedy recovery MJ.

I study too.

[Frank S]

Maybe this idea has already been discussed and debunked. But in looking at the overall concept of Savonius starting and Darrieus running is  intriguing. So why not carry it one step further for maximum wind speed stalling by mounting the Darrieus blades so they are hinged by spring loading. Think of the kitchen door of a diner it is allowed to swing both ways either by a double spring hinge or a gravity cam. This way as the winds reach speeds above the design speed factor the excess force would push the tips of the blades allowing spillage.
 If say your design RPM was 300 in  15 MPH wind it would still be turning at 300 RPM in 30 40 50 MPH gusts.
 Then since you would know your RPM would be near constant the gearing up of the generator to a known expected output could be accomplished. to achieve the desired wattage the size of the blades could be calculated.
 Like I said this may have already been thought of and tossed out. I've read many of the posts but not all of them

[Adriaan Kragten]

Maybe this idea has already been discussed and debunked. But in looking at the overall concept of Savonius starting and Darrieus running is  intriguing. So why not carry it one step further for maximum wind speed stalling by mounting the Darrieus blades so they are hinged by spring loading.

In KD 601, the idea is described of how a positive starting torque at low tip rotational speeds can be obtained if the blade can pitch in between + 10 degrees and - 10 degrees. This pitching is steered by the aerodynamic force acting on the blade. The centre of gravity of the blade is chosen such that it coincides with the aerodynamic centre around which the blade is pitching. However, at normal rotational speeds, the blade gets a blade angle of zero degrees and then the rotor works as a normal H-Darrieus rotor with fixed blades.

I have thought to use this system also to limit the rotational speed at high wind speeds but as both the aerodynamic force and the centrifugal force give no moment, this wasn't possible. Theoretically you can steer the pitch movement such that it helps the rotor to start at low rotational speeds and that it limits the maximum rotational speed but this requires a device which will be very complicated if it can even be designed. This is because the required pitch movement at low rprm is oscillating and the required pitch movement at high rpm must work in one direction to make the blade stall for any postion.

[DamonHD]

Is it the type of complexity that a low-power Arduino or similar might be able to help with?  I appreciate that some servos may also need (to use some) mechanical energy, but having a partially software controlled mechanism might work.

Rgds

Damon

[Adriaan Kragten]

Yes, I think that a software controlled pitch mechanism might work technically. However, this requires a motor on each blade and measuring of the rotational speed, the wind speed and the wind direction. So each blade should be turned to the optimal blade angle at every position of the blade. This means oscillation at the correct position at low rotational speeds, fixed position at moderate wind speeds and positive or negative stalling at high wind speeds. The development costs and investment costs of these devices will certainly be too high for a small H-Darrieus rotor but for a big one it might be different. But a big one has to compete with a normal HAWT and I am afraid that it will loose the battle.

[DamonHD]

Well, lots of things can be done at least initially as a hobby or for free that aren't immediately commercially justifiable for volume use, eg for areas where solar is very poor and no grid.  Speaking as a firm believer in open source and many years into a project of my own that is finally looking like it has commercial legs too!  B^>

But yes, it certainly sounds complex and beyond my reach and skills.

Rgds

Damon

[electrondady1]

its strange . rather than develop an open source working design like Hugh Piggot  or the Dans  have with horizontal axis mils, anyone working  on verticals are always hot to get a patent and make a fortune by commercializing . that must be why its hard to get started with verticals, we all start with a blank page.
 

[Adriaan Kragten]

I have thought a little more about the idea to use a mechanism which increases the starting torque but which can also be used to limit the rotational speed at high wind speeds and I think that I have found a way how to do it. As I describe it here on this forum anybody can use it nobody can patent it.

Assume we have a 3-bladed H-Darrieus rotor with about the geometry as given in KD 601. Assume that each blade can pitch around the aerodynamic centre which lies at 1/4 of the chord from the nose. Assume that a lever connects each blade to an eccentre which is positioned close to the heart of the rotor. The lever is connected to a point on the airfoil which lies about at 1/2 c from the nose. A vane is used to make that the centre line through the rotor axis and the eccentre axis is in line with the wind direction. Up to now there is nothing new and these method has been used by the Gyromill and by several other brands.

Now I describe what is new (at least I think it's new but it might be described earlier by someone else). Assume that the axis of the eccentre can be moved by a small motor along the centre line. When the axis of the eccentre is behind the axis of the rotor, this results in a positive blade angle at the front side, a negative blade angle at the back side and a zero blade angle at the left and the right side. So this is the position for starting of the rotor.

If the rotor has started, the axis of the eccentre moves inwards until it coincides with the axis of the rotor. So this is the position for the normal tip speed ratio of the rotor of about 4.2. The blade isn't oscillating for this position and therefore nothing is wearing.

If the rotor speed at high wind speeds would become to high, the axis of the eccentre is moved even further so now it is in front of the axis of the rotor. This results in a negative blade angle at the front side, a positive blade angle at the back side and a zero blade angle at the left and the right side. So now the blade is stalling at the front and at the back side resulting in a strong increase of the drag coefficient and this makes that the extra drag prevents more increase of the rotational speed.

So this mechanisme works in the same direction for increasing the starting torque and for limitation of the rotational speed. This mechanism consumes only a little energy as the axis of the eccentric isn't oscillating but is only moved over a small distance. One only needs a rather simple computer program which measures if the displacement of the axis of the eccentre results in enough reduction of the rotational speed. So at least the rotational speed has to be measured but that isn't a big problem.

The force which is acting on the axis of the eccentre will fluctuate during a revolution, so the motor which drives the eccentre must have a kind of worm wheel gearing such that the variation of the force has no influence on the movement of the eccentre.   

[Adriaan Kragten]

its strange . rather than develop an open source working design like Hugh Piggot  or the Dans  have with horizontal axis mils, anyone working  on verticals are always hot to get a patent and make a fortune by commercializing . that must be why its hard to get started with verticals, we all start with a blank page.
 

Patents given on VAWT's give only a substantial income for consultancy offices which write the proposals and for patent offices which give the patent. I think that no manufacturing company has made a design which has a higher value than a design which is made without any patent. Patent costs are very high especially if the patent must be valid for a large area and for a long period. The commercial market for VAWT's is very small also because buyers have now learned that the yearly output is often strongly over estimated and that certain VAWT's simply don't work at all.

We don't start with a blank page. At least my public reports KD 215 and KD 601 gives certain essential information about Darrieus rotors but on the Internet more information can be found. However, if you have only little knowledge of aerodynamics or of economy of wind turbines, it is very difficult to determine which information is true and which is false.

[Bruce S]

Using an opensource to build something that is then copyrighted, Patented, etc seems to have become the norm these days.
Microsoft is doing the very same thing. It's new Browser (Edge) is built upon *nix based opensource. Seeing the writing on the wall Microsoft  even have coded into W10 with the ability to spin-up a Linux based virtual.

While trying to play catch up,I've found it interesting that  mbower did a full diary on his build  and he's been posting about it since 2017 that uses active mechanical pitch control instead of relying on coding to do the same. From the looks of it it's built to be robust enough to handle the higher winds too.

Moving to the mechanical side of my brain, tells me his active pitch control could be built to pitch the Darrieus blades .

Perhaps if he sees this post he could weigh in ? I would very much like to see his take on it! Especially since his fabrication skills are awesome to say the least!!

Cheers
Bruce S

[MagnetJuice]

Bruce, the one who had the active pitch control was midwoud1. What mbouwer did was continue to post other stuff in midwoud1 thread.

that must be why its hard to get started with verticals, we all start with a blank page.

That is the main reason that I started this Topic. I hope that we can continue to the finish line so that others don't have to start from scratch.

Since we are on the topic of speed control for a VAWT, this is what I am going to do.

Earlier I said that I was working on the design of a magnetic braking system. Actually, it will be more like a speed control than braking.

I don't want to use words to try to describe a mechanical system. That could tax the imagination.

I am going to spend some time drawing an image of the system that I have in mind and post it here. That way everyone can see it, criticize it, praise it, cut it to pieces, put it in the grinder and at the end, maybe an effective speed control system will come out of it.

Stay tuned.

Ed

[MattM]

Instead of getting fancy with pitch control, maybe utilize camber control?  Use a weight that is influenced by centrifugal force to transition from a drag-based shape to pure aerofoil.  It wouldn't take near as much sophistication if you added a hinged surface on the front of the blade. When unloaded the hinge opens to expose flaps or spoilers to drag on the airflow.  After it self starts the hinge closes and it's purely acting as a Darrieus design as centrifugal force keeps the hinge closed.

[MagnetJuice]

Matt, that sounds similar to the system that I was working on before I thought about the magnetic one. But I wanted to use it to control the speed at the high end to keep the turbine from over speeding.

The magnetic one also works with centrifugal force and works to control the speed at the high end. The Savonius in the center should be able to start the turbine spinning. I should have an image to post later. I am not a graphic artist, but I want to draw something that is clear and easy to see how it works.

[MagnetJuice]

Here are the images of the speed control system. These images are copyrighted. As an extra precaution, to keep bad people from stealing them, I embedded a dispositivo di calore within the images that will activate 17 seconds after being in any computer other than mine, and it will melt their hard drive.



The speed control system is supposed to work as follows:

The system is installed under the turbine. There are powerful magnets inside the four steel pipes. As the speed increases, centrifugal force will cause the four pipes to start to swing out towards the perimeter of the turbine base. As the speed continues to increase, the centrifugal force will overcome the pull of the springs holding back the pipes. The four pipes will fully extend until they hit the rubber stoppers (large black dots). At that time, the four magnets will be very close to the steel band attached to the base (blue) and the magnetic pull will start slowing down the turbine.

To keep the four pipes synchronized, a steel cable connects the four pipes. The steel cable is looped through small marine wire rope pulleys attached to a small spring to keep tension on the wire.

The speed at which the pipes will fully extend can be adjusted by changing the position of the springs that hold the pipes to give it more or less tension.

When the turbine slows down some, the force of the springs pulling on the pipes will overcome the pull from the magnets and the steel band and return the pipes to the original position. I believe that the speed that the magnets engage and disengage can be determined by a careful choice of spring tension. Ideally, it should engage close to 35 MPH (15 m/s) and disengage close to 20 MPH. (9 m/s)

Please analyze it and tell me if you see any problems with it.

Thank you

Ed

[Adriaan Kragten]

Please analyze it and tell me if you see any problems with it.

Thank you

Ed

So now you need two Savonious rotors to start the H-Darrieus rotor and a magnetic brake to prevent too high rotational speeds at high wind speeds. And you claim that this is a simple HAWT which can be built by anyone. I think that building a magnetic brake will be even more complicated than building of a big direct drive PM-generator or a small one with an accelerating transmission. The further you come with your design, the more you will be hit by the inherent disadvantages of a Darrieus rotor which I have mentioned in KD 215.

[MattM]

MagnetJuice-

Pretty good for a non-artist.  Much better than my paint scribbles.

I know the biggest problem with my design is overcoming torsion effects on the blades.  So do you have to duplicate the contraption high and low on the Darrieus, or should I ask what will overcome the torsion forces trying to twist the blades?  I like the pipe idea, because it gives you an option to add or remove ballast to each 'flyweight'.  It is simple and producible.  You also stick to a poured stator for the PMG, which makes it reminiscent of a nod at the Hugh Piggott design.

I'm not sure why Adrian sees the magnetic speed brakes being an issue.  They appear to be fixed into their location and devoid of any unnecessary complexity.

[Adriaan Kragten]

Instead of getting fancy with pitch control, maybe utilize camber control?  Use a weight that is influenced by centrifugal force to transition from a drag-based shape to pure aerofoil.  It wouldn't take near as much sophistication if you added a hinged surface on the front of the blade. When unloaded the hinge opens to expose flaps or spoilers to drag on the airflow.  After it self starts the hinge closes and it's purely acting as a Darrieus design as centrifugal force keeps the hinge closed. (Attachment Link)

In one of my earliest windmills I have used elastic air brakes on each of the blade tips of a 3-bladed HAWT with wooden blades and a design tip speed ratio of 6. A brake was made of 1 mm stainless spring steel and had no hinge but was bending outwards due to the centrifugal force. I have also tested a scale model in the wind tunnel and proved that the air brakes gave a strong reduction of the Cp of the rotor if the rotor turns fast enough. For the real rotor (with a diameter of 4 m), the rotational speed was also strongly reduced at high wind speeds. However, the air brakes made the rotor very noisy at high wind speeds. It sounds if a helicopter was flying over. So although the construction was very simple, this was the reason that I finally cancelled the idea.

Air brakes can work as has been demonstrated by the Windcharger but there they are mounted at separate arms which are much shorter than the blade length. If they are mounted at the blade tip you get an extremely high noise levels because there you have the maximum speed. A H-Darrieus rotor has a lower optimum tip speed ratio than 6 and using the blade itself may produce less noise then when a thin strip is used but certainly the noise level will rise if the system starts becoming active. This is because the increase of the drag coefficient is only gained when the blade is stalling.

I don't see the need to split the blade and how the inside part will go inwards. If one massive blade is used and if the hinge axis is lying at the tip of the blade, the centrifugal force will make that the whole blade is pushed outwards. The braking effect will be strongest at the front side of the rotor but braking will also be effective at the left and the right side. At the back side it will be minimal. But you need a spring and a stop to get the blade at the neutral position when the rotational speed isn't too high. I think that this system is cheaper than a magnetic brake.

[MagnetJuice]

Adriaan, thank you for reminding me to keep it simple and inexpensive.

That magnetic brake that I posted before doesn't cost that much to make but it would be a bit complex to put together.

I am working on another speed control system that is easier to build, uses fewer parts, and doesn't use magnets.

The way I see it, speed control for a VAWT is essential, just like speed control for a HAWT. Without speed control, both VAWT and HAWT could end up self-destroying during high winds. It makes no sense to build any wind power turbine that cannot survive high winds. I don't mean surviving a tornado or a strong hurricane, just winds below 90 MPH.

I have most of the parts to put the turbine together and have all the parts for the alternator. The alternator will be an axial flux and since the turbine should be able to turn at 300 or 400 RPM, it will be direct drive. I expect to get an average of 200 watts from the turbine, but I will build the alternator to produce about 700 watts or a bit more.

One thing that I am not sure about is what kind of base I am going to build for the turbine. I will build it in the garage and do all the tuning and testing there. Then I will move it to the back yard. I don't want to dig holes in the back yard so I am thinking of getting four used car tires, (I can get those free from a tire shop) lay them in a square about 8 feet apart and fill them with rocks and dirt. After that, I can use landscaping timbers to tie the whole thing together. I hope that will be heavy enough to keep it from tipping over in moderately high winds. Does anybody has a better idea?

Ed

[MattM]

Post hole digger and bury some sleeves where the pivot point is below ground level and it shouldn't ever tip short of a hurricane.  If you're stuck on tires then use cement with wires running  through the outside of the tire to a sleeve and secure the sleeve to your pole with a long screw as a keeper.  That way it's still portable, but it's disposal is simpler.

Back to braking... you can always use centrifugal force for braking, too.  That helicopter sound Adriaan is talking about is pretty common with flutter.  Helicopters make that sound when the pitch of the blade begins to oscillate in a twist.  It also happens when the blades bend longitudinally in forward flight and the blade deforms in and out of the stresses.  That's why helicopters use blades with highly flexible properties.  Every variable-setting choice will have some flutter.

[Adriaan Kragten]

I don't think that the helicopter sound which my rotor was producing is caused by blade flutter. I have tested a scale model of the rotor in the wind tunnel without the air brakes and flutter wasn't happening for an unloaded rotor even at the maximum tunnel speed of 11 m/s. Even without an air brake, most sound of a rotor is produced by the blade tip because there a large vortex is created because of the over pressure at the front side and the under pressure at the backside. An air brake at the tip creates its own extra vortex and this extra vortex is the cause of the strong increase of the sound level. So you hear three big bangs per revolution if the rotor turns fast at high wind speeds.

If blade pitching is used for a H-Darrieus rotor, you don't create extra vortexes at the blade tips but only more drag along the whole blade length and I expect that the extra noise which is produced by the stalling airfoil, is acceptable, at least if the system starts working at a not very high rotational speed.

The blade angle of an helicopter blade varies during a revolution when the helicopter has a horizontal speed. Without variation of the blade angle, the lift created at the left side of the rotor would differ strongle from the lift at the right side of the rotor. To make that this variation of the blade angle at the blade root is felt at the blade tip, helicopter blades are very torsion stiff but they can be rather flexible for bending.

[SparWeb]

Warm up your slide rules, kids. 
I just found a NASA tech report from 1975.  5 meg download, PDF. 
"THEORETICAL PERFORMANCE OF CROSS-WIND AXIS TURBINES WITH RESULTS FOR A CATENARY VERTICAL AXIS CONFIGURATION"

https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19760003944.pdf

There's enough analysis to work out an analysis for a straight-bladed VAWT, plus a BONUS FREE FORTRAN PROGRAM! [Drool]

[Mary B]

Adriaan, thank you for reminding me to keep it simple and inexpensive.

That magnetic brake that I posted before doesn't cost that much to make but it would be a bit complex to put together.

I am working on another speed control system that is easier to build, uses fewer parts, and doesn't use magnets.

The way I see it, speed control for a VAWT is essential, just like speed control for a HAWT. Without speed control, both VAWT and HAWT could end up self-destroying during high winds. It makes no sense to build any wind power turbine that cannot survive high winds. I don't mean surviving a tornado or a strong hurricane, just winds below 90 MPH.

I have most of the parts to put the turbine together and have all the parts for the alternator. The alternator will be an axial flux and since the turbine should be able to turn at 300 or 400 RPM, it will be direct drive. I expect to get an average of 200 watts from the turbine, but I will build the alternator to produce about 700 watts or a bit more.

One thing that I am not sure about is what kind of base I am going to build for the turbine. I will build it in the garage and do all the tuning and testing there. Then I will move it to the back yard. I don't want to dig holes in the back yard so I am thinking of getting four used car tires, (I can get those free from a tire shop) lay them in a square about 8 feet apart and fill them with rocks and dirt. After that, I can use landscaping timbers to tie the whole thing together. I hope that will be heavy enough to keep it from tipping over in moderately high winds. Does anybody has a better idea?

Ed

When we needed to demo 10" satellite dish antennas at county fairs back int he 70's we used 4 truck rims with a 4x6 framework(4x4 flexed to much) to hold the mounting plate and pole. Stood up to 90mph winds one year that shredded every other TV vendor there. Key is make some trays to stack concrete blocks in(cheap at the lumber yard or free at a house tear down) as far out on the outriggers as you can. We used an 8x8 foot square setup... run 4x6's in the perimeter bolted to the rims with threaded rod, then 2 more 4x6's running across the center spaced far enough apart for the mounting plate. 6 inch dimension is the vertical to hold the down forces...

[topspeed]

Warm up your slide rules, kids. 
I just found a NASA tech report from 1975.  5 meg download, PDF. 
"THEORETICAL PERFORMANCE OF CROSS-WIND AXIS TURBINES WITH RESULTS FOR A CATENARY VERTICAL AXIS CONFIGURATION"

https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19760003944.pdf

There's enough analysis to work out an analysis for a straight-bladed VAWT, plus a BONUS FREE FORTRAN PROGRAM! [Drool]

Cool..looks to be similar sized as the McDonnell Aircraft VAWT model that promissed .51 Cp....in the seventies.

Mine is a bit smaller.

Thank you very much...I study fluid dynamics...this is right there...with equations and assumptions.

[MagnetJuice]

Matt and Mary, thanks. Any of the ideas that you suggested is better than using tires filled with rocks and dirt.

The thing is that I plan to move within a year or two, so I want this thing to be portable and I don't want to do a lot of repair to the yard when I move.

I checked the prices of those pivots and WOW! Very expensive. I am thinking of using something similar to pivots. Get two 24 inch long galvanized stove pipes and cut them in two. That will give me four pieces 12 inches long. Then I use a posthole digger and dig four holes 16 inches deep. I'll put the stove pipe in the holes and I put the landscape timber into the pipes. Then pack a few stones around the post.

I think that should be strong enough because the whole turbine should weight about 100 Lbs. and winds over 80 MPH are extremely rare around here. When I move, it should be easy to remove the posts. The stove pipes will stay in the hole but they will be about four inches below ground. I can fill the holes and place a piece of sod on top. Easy fix.

I think that if I place the posts in the ground toe-out, it could be a stronger base. Something like this.



Maybe Frank knows something about that; it looks like he knows a lot about heavy-duty construction.

Ed

[CraigM]

MagnetJuice,

Have you considered screw in earth anchors? They do not disturb the soil much and come in many sizes.

I use them as a safety tie-down for a pre built shed. They're fairly easy to install and are removable.

Quick search showed 10,000 pound pull force on larger sizes.

They've have been discussed on several threads in this forum.

2 cents,
CM

[MagnetJuice]

Those screw-in anchors would work if I had the right soil. My soil is about four inches of sod and below that is very rocky-stony ground. It would be nearly impossible to screw one of those in this type of ground.

Thanks anyways,

Ed

[MagnetJuice]

I think I found a solution to anchor the turbine base to my yard. These 24 inch bolts have a hex head and can be screwed into the ground with a power drill and a ¾ inch socket.

https://www.amazon.com/GROUNDGRABBA-Stakes-HexHook-Secure-Warranty/dp/B07NPBQXPM/ref=pd_rhf_dp_p_img_1?_encoding=UTF8&psc=1&refRID=6WPYZAQ26R9Z8GGA2GKY

I can drill a hole through the timbers, then place a large washer on the bolt and screw it through the timber into the ground. They should be able to stay put in my stony ground.

Now I need to work on finding a solution to provide speed control to the H-Darrieus turbine.

Ed