Author Topic: My First Vawt Project (Read 45770 times)

tjm2000 · « **on:** July 27, 2020, 08:20:54 AM »

My journey with darrieus vawts started with a small model 1ftx1ft that was made out of some thin strips of wood and some plastic sheet curved into a crude symmetrical airfoil. It ran really well. it would start in very little wind and wouldn't stall at higher winds. Eventually a storm came through and ripped off one of the blades. It still ran pretty well though.

I then started planning my next project, A vawt with 2m tall blades and a 3m diameter.
About 12 months ago, I started buying parts as I could afford them.

For my first project, I thought building a generator was a bit ambitious, so I based my project around a 1kw 150rpm generator I brought on Alibaba. Communication and shipping was really good, and the generator appears to be well made. I did some testing on a friends lathe and found the output to be very very close to what was advertised.

Due to minimal experience and tools, I wanted this build to be as simple as possible.
I decided to use 50x25x5mm rhs aluminium as the arms and the spars of the blades, I chose this as it is light, strong, cheap and readily available.
I used 10mm mild steel to make the plates for the hub. I had a friend of mine build and extension for the top of the generator to the 'upper' hub. It is all held together with m10 stainless hardware.
Attached is a pic of the completed (unpainted) hub assembly.

tjm2000 · « **Reply #1 on:** July 27, 2020, 08:47:09 AM »

The blades are cut out of sheets of polystyrene used for building insulation. I cut these in an airfoil shape by using a homemade hot wire cutter. I made 2 templates out of plywood and screwed then on each end of the polystyrene block. This worked very well and it didn't take long to get 3 good foam blades. There is a 50x25mm cutout where the Aluminium spar will be glued in.
Next I put 1 layer of fiberglass cloth (200gsm) in the cutout. This was to fill in the excess space as it was cut slightly oversize. I then drilled and glued the mounting bolts to the aluminium spar. I then glued the spar into the cutout with thickened epoxy.
Now it was time to fiberglass the blades. I used 2 layers of 200 gsm cloth, which is probably overkill but it will make a strong blade. This was my first time doing fiberglass so it was a big learning curve. It is not perfect but I am mostly happy with how they turned out. total weight of each blade is 4.5 kg. This was lighter than I was expecting and they are very strong.

tjm2000 · « **Reply #2 on:** July 27, 2020, 08:56:05 AM »

I did a test fit of the blades today to see if it would all go together as intended. It went pretty well. The blades are held on by bolts through aluminium 'L' brackets. I have not yet decided if they will be strong enough. So far i am very happy with how it is progressing. I hope to have a temporary stand finished by the end of the week.

MagnetJuice · « **Reply #3 on:** July 27, 2020, 05:27:45 PM »

Welcome to the Forum tj,

That VAWT looks really nice. That should be able to produce about 1Kw with 9 m/s wind.

Can you share how you decided on those dimensions? The blade chord looks like it is .33 meters. If that is the case, that is good solidity.

Ed

tjm2000 · « **Reply #4 on:** July 27, 2020, 06:16:25 PM »

There is nothing scientific about my dimensions, The blade chord is 0.35m because that was the width of the foam sheets I had and it looked about right.. I expect a tsr of 2.5-3, so I made the diameter 3m to achieve around 150 rpm in 10ms wind.
I will eventually be used to power an office/sleepout that will not be grid connected. The wind on site is a steady 8-10ms during the day, and dropping down to nothing during the night. Im hoping I wont need much in the way of storage.

MagnetJuice · « **Reply #5 on:** July 27, 2020, 10:06:52 PM »

Looks like a good design. My only concern is with the way the blades are supported. The upper and lower section of the blades could snap when is turning at full RPM.

A blade with a mass of 4.5 kg will have an effective mass 30 times that much when the turbine is spinning at 150 RPM.

If it was my turbine, I would feel more comfortable if the blade had better support at the upper and lower sections.
A stainless steel cable from the hub plate to the blade could help with that.

Ed

SparWeb · « **Reply #6 on:** July 28, 2020, 12:49:52 AM »

Looking good.
I'd second the suggestion to have a diagonal brace, though for compactness you could add the strut in between the two horizontals. Easier to attach at the existing bolt points. I'd avoid cables because the load alternates from tension to compression - a cable would snap tight, loosen, snap, loosen...

Have you done anything with the alternator connections? I expect your VAWT will turn <100 RPM.

MagnetJuice · « **Reply #7 on:** July 28, 2020, 02:09:59 AM »

SparWeb is right; the diagonal brace should be solid.

I forgot about the wind hitting the other side on the way back.

Ed

tjm2000 · « **Reply #8 on:** July 28, 2020, 02:48:39 AM »

That is a valid concern, and I did think of it after I had wrapped the wing in fiberglass. I have plenty of extra foam cutouts so I will throw together another set of blades that have provisions for extra bracing if I have any issues.
Why do you think the rpm is likely to be less than 100? From what I have read, a loaded tsr of around 2.5-3 is achievable.
If I do not achieve 150rpm I was going to shorten the diameter.

MagnetJuice · « **Reply #9 on:** July 28, 2020, 03:34:29 AM »

These are some of my calculations.

Of course, I know that the results could vary; it depends of the equations and calculators that are used.

Ed

electrondady1 · « **Reply #10 on:** July 28, 2020, 07:30:47 AM »

very nice work tjm2000

SparWeb · « **Reply #11 on:** July 29, 2020, 12:25:12 AM »

This machine looks like it's going to work and it could work very well in fact. The more power the VAWT can collect, the more torque is applied to the arms and the column, and the more bending is applied to the tower. So you had better prepare yourself adequately for success.

It looks like it hasn't been spun yet. It's hard to appreciate the mechanical energy or the gyroscopic effects until you do.

I can think of a few more suggestions to make sure things go safely when you are ready to spin it for the first time.
1) Have room to walk under the swing of the blades (say 4 feet / 1.5 meter or more).
2) have a way to stop it. I suggest shorting the wire leads of the alternator together.
3) put a bicycle tachometer (cheap ones ~20 USD) on it to measure RPM exactly.
4) put a diagonal brace from upper inboard to lower outboard arm (or vice-versa)

I have built VAWT's like this and they vibrated in may ways. Having 3 blades is a good defense against this oscillation but it can still do it, and the cyclic bending is a killer on the struts. I'm pointing this out because I'm not sure my previous mention of the diagonal brace was clear.

Quote

That is a valid concern, and I did think of it after I had wrapped the wing in fiberglass. I have plenty of extra foam cutouts so I will throw together another set of blades that have provisions for extra bracing if I have any issues.

Not necessary. Attach the diagonal braces from the anchor points of the struts. Sharing fasteners if you prefer.

tjm2000 · « **Reply #12 on:** July 29, 2020, 02:35:14 AM »

I am putting the tower together in the next few days. I will hopefully have it spinning in the weekend.
The charge controller has a brake switch which shorts the output of the generator.

I am worried about vibration and the structural integrity of my machine. It feels very solid while it is stationary, But it is hard to imagine it spinning at 150 rpm. This is my first project like this, and I feel like I may have bitten off more than I can chew...

As for the diagonal brace, I assume you mean something like the pic below? If so, I will definitely make something up before it spins to give me some peace of mind.

SparWeb · « **Reply #13 on:** July 30, 2020, 12:42:46 AM »

Yup, that's what you need!

Right, my intent was not to make you feel daunted, but just to make sure that nothing will go wrong in the face of success.
My first VAWT (circa 2006) was not usually able to self-start, but it had a lot of torque when it was running. That meant that it could really get going up to speed, but not necessarily when I was ready for it to do so. One careless day I left it with no controls or brake set and when I came back a few hours later it was spinning very fast and I had no way to stop it. All I could do was stand back and watch. This is the kind of thing that was bound to happen because I didn't know what to expect. In a funny, ironic way, I wasn't expecting it to work.

A test that you can try out before it's deployed is to use the shorting brake you have and confirm that the blades are very hard to turn. If that's OK, then release the brake, give the blades a free spin up to a decent speed, and then switch the brake on again. Watch how firmly they slow down and/or stop. Feel the generator and the controller case for a build-up of heat. That should give you some more peace of mind that, if anything doesn't seem right, you can bring it to a halt.

How high is the tower/platform?

tjm2000 · « **Reply #14 on:** August 01, 2020, 01:31:11 AM »

I finished most of the welding for the tower today, and put it in position. It stands about 3 meters tall. I plan on bolting it down with threaded rod epoxied into the rock. The rock here is not very hard, and the layers separate pretty easily. To try combat this, I am going to drill several holes and hammer rebar into the rock to stop the layers shifting. There are hundreds of pieces of rebar in the rock around the foundations of my house.
I am still waiting on the generator mount to be machined, so it wont be spinning for another week or so.

MagnetJuice · « **Reply #15 on:** August 01, 2020, 02:53:32 AM »

Adding a diagonal brace as suggested by SparWeb can add strength to the blade structure and keep the blades from bouncing up and down.

However, the top and bottom sections of the blades must be supported. I wonder if adding braces as shown in this image could accomplish the same thing. The braces should be thin so they don't cause too much drag.

I also think that the addition of endplates to the blades is worth investigating.

The image on the left shows a blade with endplates. Research has shown that adding endplates to the tip of the blades improves performance. That is why all major manufactures of large airplanes have added winglets to the end of the airplane wings.

I found two interesting documents in PDF discussing the addition of endplates to straight-bladed VAWT's. They also discuss other design parameters. They can be downloaded here:

http://gooa.las.ac.cn/external/share/1322154

https://www.mdpi.com/2077-1312/8/3/225/htm

Ed

tjm2000 · « **Reply #16 on:** August 01, 2020, 04:46:05 AM »

Thanks for those pdfs, They are interesting reading.
I definitely plan on adding end plates, I'm just not sure what material I should use.
On a different note, has anyone ever used vinyl wrap instead of paint on blades? I tested out a scrap I had left over and it stuck very well. It holds up to uv pretty well (lasts about 5 years on a car) and its cheap and easy to apply.

re the bracing, I can understand a diagonal brace in between the struts to stop the bouncing effect, But I am not sure why I need to support the ends of the blades? The aluminium spar runs the length of the blades and they are pretty strong. 85kg me can comfortably stand on the centre of the blade when it is propped up at each end.
Depending on how testing goes, I might make up another set of blades and change the bracing a bit.

SparWeb · « **Reply #17 on:** August 01, 2020, 01:16:41 PM »

Hi TJM,
You might want to go to the hobby store and find "monokote" which is used to wrap model aircraft. It has a self-adhesive coating on one side which is heat-activated. It also has a slight "shrink-wrap" so as you heat it (hair dryer) it both bonds to the surface and tightens up.

End plates... meh... They do things for the lift distribution which can be good, and they do things to the induced and profile drag which are usually bad. At best, it's tuning for the last 10%. You can get started without them - many far more important things to take care of.

The tower looks good and sturdy. Any thoughts about pouring concrete?

MagnetJuice,
This isn't meant to be a slapdown but that suggestion doesn't do anything. For a visual way to assess any truss*, compare the amount of area enclosed by rectangles with area enclosed by triangles. What TJM illustrated July 29 was a complete understanding of what's needed. No need to offer alternatives when the ideal one has been found.

I'll also add: TJM has also correctly employed triangular members in the construction of the tower, so no complaints there, either.

* Every single truss ever built by mankind can be divided into 2 groups: those that use triangles and those that are not actually trusses. This distinction has surprised and humiliated many professional bridge engineers for centuries. Google: Miami pedestrian bridge

Adriaan Kragten · « **Reply #18 on:** August 02, 2020, 01:51:14 PM »

Quote from: tjm2000 on July 27, 2020, 06:16:25 PM

There is nothing scientific about my dimensions, The blade chord is 0.35m because that was the width of the foam sheets I had and it looked about right.. I expect a tsr of 2.5-3, so I made the diameter 3m to achieve around 150 rpm in 10ms wind.
I will eventually be used to power an office/sleepout that will not be grid connected. The wind on site is a steady 8-10ms during the day, and dropping down to nothing during the night. Im hoping I wont need much in the way of storage.

In my public report KD 601, I have derived a theory to design a H-Darrieus rotor. This report can be copied for free from my website: www.kdwindturbines.nl at the menu KD-reports. I have used this theory to design a 3-bladed H-Darrieus rotor with a diameter of 2 m and I found an optimum chord of 0.2 m for a design tip speed ratio of 4.2 and for the symmetrical airfoil NACA 0015. If you scale up the rotor diameter to 3 m, the chord becomes 0.3 m for the same design tip speed ratio and airfoil. Your chord is somewhat larger and I estimate that the optimum tip speed ratio will be about 4 which is still acceptable. The advantage of a larger chord is that the Reynolds number is higher and therefore the rotor will perform better at low wind speeds. So I think that your geometry is rather optimal. The question remains if the rotor will start at an enough low wind speed without the help of the generator used as motor and if the rotor is strong enough at very high wind speeds.

tjm2000 · « **Reply #19 on:** August 09, 2020, 01:18:24 AM »

It has been a weekend of highs and very low lows for me.
It started off great. I glued end plates on the blades and painted them. I also had the generator mount welded onto the tower and assembled the turbine on top of the tower for the first time. It was not easy lifting everything up there but with a couple of people we managed. The wind was very light at 1-2ms which made assembly very easy.
Finally it was spinning!

Today the wind was a steady 7-8ms, Turbine was sitting at around 120 rpm and producing 450-550w. These were very promising results and pretty close to what I had calculated.
Then everything came crashing down. I didn't see it when it happened but I certainly heard it.
It turns out I had forgotten to put the nuts on the threaded rod that holds the tower into the rock. The wind simply pushed it over. I cant believe I made such a stupid and obvious mistake, But ultimately I am the only one to blame and the only thing I can do is learn from it.
All 3 blades are damaged beyond repair and the aluminium arms are all twisted up. The hub assembly and tower seem to be mostly ok.

I don’t have enough money to buy materials to make it complete again, so sadly I will have to put this project on hold for a few months while I save.
Thanks to everyone on here who has given me help, suggestions and reading material, It was much appreciated.

DamonHD · « **Reply #20 on:** August 09, 2020, 02:13:37 AM »

Very sorry to hear of the mishap.

Maybe for next time prepare a checklist of the things to remember, well in advance.

Checklists save a lot of serious errors during surgery for example, when there are many other things to remember.

Rgds

Damon

MagnetJuice · « **Reply #21 on:** August 09, 2020, 03:52:14 AM »

TJ, you have put so much time and effort into this project. I can feel your pain. One of the good things that I have learned in my life is this; you only get defeated when you give up, so I like it when you say, “the only thing I can do is learn from it”.

It was a loss, but not a big loss. The main parts of the system, the tower, hub and alternator are not damaged and can be reused.

If it were my project, I would limit the size of the blades to 1.6 meters high. That could still produce 700 watts with 9 m/s winds at 145 RPM. An advantage of a smaller size turbine is that it could keep your 1 kW alternator from burning up if the wind get stronger.

In addition, I would stay away from aluminum for the blade spar. Aluminum is very bad when it comes to metal fatigue. Thin tubular steel is much better.

Ribs like in this image covered with a sheet of High Density Polyethylene about 3 mm thick could work fine.

I wish you the best. You have a nice project going here. Please keep us informed.

Ed

hiker · « **Reply #22 on:** August 09, 2020, 11:32:46 AM »

Gen make it out Okay, ?,,well , at least you know your design Works ,,next build will be A Breeze,,,😜

Mary B · « **Reply #23 on:** August 09, 2020, 02:51:48 PM »

Being a ham radio operator I have 2 towers... standard operating procedure doing ANY work is start at the base and work my way up checking bolts as I go to make sure they are all tight, nothing has corroded badly etc. Only after a complete tower check will I start working on the antennas.

SparWeb · « **Reply #24 on:** August 09, 2020, 09:01:59 PM »

Sorry to see it TJM.
What a bummer to see it look so good and have so much promise but then suffer the crash.

I do look forward to seeing the TJM Mark 2 up in the air very soon.

makenzie71 · « **Reply #25 on:** August 09, 2020, 11:03:20 PM »

Man that really blows...usually with the stuff I work with blades are cheap so I'm okay seeing them go, but you got a lot of effort in those vanes. Crummy way to learn a lesson :/

I know others have already said it...but as someone who makes all kinds of stuff and has been doing it for a really long time...if you don't put a diagonal brace in those arms you're going to learn another less a crummy way

tjm2000 · « **Reply #26 on:** August 10, 2020, 08:24:26 PM »

Today I stood up the tower and started making a plan for the rebuild. Tower (now bolted down!!!), generator and hub all made it out fine. I cut off all the bent pieces of aluminium crossarms and the lengths are long enough to make a turbine with a 2.4 Meter diameter.
From my testing before it fell over, It was doing 120 rpm in a 8ms wind, giving me a loaded tsr of 2.4.
A 2.4M diameter turbine would be doing 150 rpm in the same conditions, with less torque. My generator is rated for 1kw at 150 rpm. I’m worried that if I make the diameter smaller, my turbine will be a poor match for the generator.

I cant decide on how to make the new set of blades. There are probably better ways than my foam/aluminium core fiberglass blades.
My latest idea was to make a female mold of half of a symmetrical airfoil, lay fiberglass and vacuum bag, the glue the 2 halfs together with a spar (probably wood reinforced with fiberglass or carbon fiber) and ribs.
I am going to stay away from aluminium for the spar this time. The blades will also be slightly shorter.

SparWeb · « **Reply #27 on:** August 12, 2020, 01:17:35 AM »

Myself I didn't have objections to the original blades, but this is a simpler way. With a female mould the finished outer surface of the part will be very nice. You can make the mould surface with aluminum sheet if you can shape it. Have you vac-bagged before? There are many tricks...

Adriaan Kragten · « **Reply #28 on:** August 15, 2020, 09:33:14 AM »

You are not the first one who has made such a mistake. I have seen a complete tower falling down on a green house because one of the guy wires unhooked at the foundation. So never use open hooks at the guy wires. When I worked at the UT-Eindhoven, we had a big water pumping windmill for which the tower could be lowered using an auxiliary tower, a long steel cable and a Tirefore winch. It appeared that we had two cables of different length. The tower was erected using the longest cable. Next someone lowered the tower using the shortest cable and halfway this procedure the cable run out of the Tirefore. The tower fell on a supporting structure and was destroyed completely.

Although it has run only for a short time, you have measured a certain power. Did you measure also the wind speed for that power? So do you have an idea of the Cp of the rotor? It has run so it must have been started. Has it started by its own or did you use the generator as motor or did you pushed the rotor by hand to start?

tjm2000 · « **Reply #29 on:** August 15, 2020, 04:55:03 PM »

I measured 450-550w at a 7-8ms wind speed. This gives me a Cp of around 0.28 I believe? It started by itself.

I am not too far off getting it running again. The new blades have the same NACA 0018 profile at 0.35 chord, but they will be slightly shorter at 1.8 meters.
Hopefully it will be running again by next weekend

Adriaan Kragten · « **Reply #30 on:** August 16, 2020, 03:46:18 AM »

I have checked your Cp and I found a somewhat higher value. I think that this is because you have neglected the generator efficiency. Assume that the electrical power Pel is 500 W and that the generator efficiency is 0.8. This gives a mechanical power P = 625 W. Assume the wind speed V is 7.5 m/s. Assume the air density rho is 1.2 kg/m^3. The original rotor had a diameter of 3 m and a height of 2 m so the swept rotor area A = 6 m^2. The Cp is given by: Cp = P / (0.5 * rho * V^3 * A). This gives Cp = 0.33. This is a realistic value for a H-Darrieus rotor.

I am surprised that the rotor has started by itself. A H-Darrieus rotor normally has a negative Cq value for low tip speed ratios and so it needs some help to reach the tip speed ratio range for which the Cq becomes positive. However, at very low tip speed ratios it works as a drag machine and a strong wind gust may cause a sufficient high rotational speed for which the Cq is positive if the wind speed is low directly after the wind gust. So I think that you have had very fluctuating wind speeds during starting. This effect is described at page 11 of report KD 601. The estimated Cq-lambda curve for a rotor with fixed blades is given in figure 4 of KD 601 and there you can see that Cq is positive for 0 < lambda < 0.3 and negative for 0.3 < lambda < 1.4.

SparWeb · « **Reply #31 on:** August 16, 2020, 08:02:18 PM »

My experience with VAWT's is that they can self-start in gusts given a little time and opportunity. Wind near the ground can vary quickly, and if you were to consider the travel of one molecule as it passes into the circle described by the blades and then out of it, it would take several seconds to cross all the way through. That coincides with the rise-time of some gusts.

What this means is that one blade can be stimulated by the force of a gust which moves the arms before the other blade is pushed opposite. As the gust weakens the second blade is not damped by the same force that struck the first blade. Enough angular momentum may remain to complete a rotation.

Adriaan Kragten · « **Reply #32 on:** August 17, 2020, 01:15:20 AM »

The point is that at low tip speed ratios, the blade is completely stalling when it is at the front and at the back side of the rotor because there the angle of attack is very large. A stalling symmetrical airfoil has a lot of drag and the tangential component of this drag is much higher than the positive tangential component which is generated for other blade positions for which the angle of attack is small or for positions for which the blade works as a drag machine because of the difference in drag coefficient depending if the wind faces the airfoil nose or the airfoil tail. So to my opinion a H-Darrieus rotor will never start if the wind speed changes slowly, even if there is some local turbulence.

But if there is a wind gust of for instance 15 m/s during a certain time, the rotor will reach a certain rotational speed because it works as a drag machine with a tip speed ratio of about 0.3. If then suddenly the wind speed is reduced to 3 m/s, so reduced by a factor 5, the tip speed ratio for the existing rotational speed will be increased by a factor 5 and so it becomes 1.5. This may bring the rotor in the range of the Cq-lamba curve for which the Cq value is positive when it works as a lift machine and then it will accelerate. This procedure will only work if the generator has a very little sticking torque but as an axial flux generator is used with probably no iron in the coils, this is the case.

News:

Author Topic: My First Vawt Project (Read 45770 times)