Fully parametric vertical axis wind turbine (VAWT)

by motoz Jan 23, 2019
Download All Files

Thing Apps Enabled

Please Login to Comment

Very nice work! This is not the historical Savonius cross section, which was 2 half cylinders. I suspect this will work a bit better than the historical design. If you want to find the original geometry and wind tunnel work, google "Savonius" "Blackwell report". It's from the 70's.

More interesting might be the report by Ian Ross, wind tunnel blockage factors, or something like that, which will help you understand why the Blackwell report claimed .2 Cp (captures 20% of energy in wind moving through it). Modern very large HAWT's get close to .55 (55%), modern small HAWT's are lucky to get .31, and the best testing I've been able to find shows that the wind side type turbines get between .05 and .1 (5% to 10%). Wind side claims .2, however in the 20 years they've been in business, they've never bothered to certify that claim, only that the alternator is safe to produce that much power. They came into the north american market for a few years as the GUS and other names. Mostly gone now because they didn't meet the expectation their sales propaganda lead people too.

You might want to take a look at my design.


I've got a bit update to the design, but the one posted here reached .31 Cp, in testing by consulting engineer. The new version may not be that much more efficient, but it will be much easier to build.

ART Turbine
by drewrt

Thanks for the info! I know there are better designs than the different savonius types, but my interest in this is mainly that of learning freecad, and building something for fun. The main difficulty when trying to build a small power generating wind turbine is anyway the generator, just look at the number of wind turbine projects here on thingiverse that end with a comment 'just need to get a generator' or if they already build one 'need to do some testing to see if it gets up to a useful voltage level'. Your design looks intriguing, but I can't seem to find any info about the working principle and what makes it better, do you have some links? Learning more is always fun!

In actuality the details around how VAWT's work, Savonius, and the many Darius varieties, are pretty murky. Surprising to say that some basic science hasn't been done in 2019, but it hasn't. If you google NASA and how does an airfoil work, you'll find that the reasons changed only within the last 5-6 years, as CFD made it clearer and clearer that Bernoulli was an effect, not a cause (hint, the low pressure vs high pressure is no longer the explanation in style). Various wind tunnel studies have been done on VAWTS, however it's been well understood, but unpublished, that most of these studies don't translate well to real world performance. The savonius is a great example, Savonius himself claimed 30%, the Blackwell report in the 70's said 21%, real world tests showed 11% or below. A great report that explains this is the Ian Ross 2010 paper on wind tunnel blockage. No real concencus, other than that in the real world the Savonius seems to have an efficiency too low to be worth putting a generator on. The wind side works, but no real world testing has shown the 21% they claim. I'm not an academic, so I don't have much motivation to publish. If you are looking for some of the work that inspired me, look up the Modi, Benesh and Fernando papers on VAWT's.

You are correct about the issue being alternators. The problem is twofold, small VAWTS don't turn fast enough for most alternators to make useful volts, and alternators are linear output, double the speed, double the watts. Wind turbines like these, as the wind speed doubles, the power available cubes, so you have 8x as much shaft power available, but only 2x the speed.

I'm in talks with a Chinese manufacturer for an alternator, but think I may just have to make my own again. I've had some good luck building axial flux version that are large, but I'm struggling to get one that's small enough to be 3D printable on a small bed.

After I've got it working I'll decide about patenting. You can see the difference between my published version and the savonius types in my patent.


I spent fortunes on the last patent, and I'm not as convinced as I used to be that it's worth while.

Have you considered using a stepper motor out of a 3D printer for a small/lowspeed turbine? I've some arriving today and plan to see if I can get them to work. I don't want to generate large amounts of power, just enough to do some interesting stuff with the windmill like possible drive an arduino to use as a controller or something. Do you've any insight into how I can learn about the aerodynamics of these sorts of turbines? I've no idea which curves to use, the twist that's best, etc... I can use half cylinders but is that really best? Is it imported to keep the central support open like this model? etc... etc...

I have considered and discarded the idea of stepper motors as alternators, as I have not been able to find one instance of someone successfully doing so in a way that would work. You are correct in your assessment, that finding an appropriate direct drive alternator is complicating DIY VAWT research. Something that is an additional complication in finding an appropriate load for a research turbine, which you may or not be aware of, is that most turbines have a relatively stable TSR, (Tip Speed Ratio). This is the relative speed between the turbine blade (usually outermost), and the flow. A turbine blade, loaded (turning against resistance) for it's maximum Watts output with it's tips moving at 10m/s while the flow is 10m/s, can be said to have a TSR of 1. This TSR is relatively consistent over most turbines operating flow speed range. The issue is that as the flow speed doubles, the shaft power available cubes, so 8 times as much energy in shaft power is available, but the speed has only doubled. Most alternators have relatively linear outputs, as their speed doubles, their wattage doubles, meaning that the turbine will speed above it's ideal load, generating less watts than are available, and overspending can have it's own problems. So the issue isn't simply matching an alternator, it's controlling the alternator so that it loads the turbine effectively. This is called MPPT tracking, and it generally uses a controller, like an arduino to measure the speed of the turbine, and then use a precacluated value based on the speed to set how often the alternator is turned on. It allows you to attach an alternator that's got enough capacity to match the turbine required load at very high winds, and pulse it on and off (MPPT + PWM), in way that loads the turbine for maximum power, and still get energy generated effectively at lower speeds. Sadly MPPT controllers are getting rare as the small wind market shrinks. The good news is that some RC motor controllers seem capable of this, via regenerative braking functions, especially Castle Creations. Rather than look to steppers, I'd suggest the lowest kV 3phase rc motors, or cordless drill motors may be the way to go. If you want to understand my thinking around high solidity VAWTS, I'll be releasing some big news around that in the next month or so, including my newest design, so easiest is if you just follow my profile, here or on linked in.