WinDIY - HAWT Wind turbine

Mostly 3D printed HAWT windturbine incl. 3D printed disk-generator and brake system.

Similar projects worth following
WinDIY is a design of a (mostly) 3D printed wind turbine. The idea is to build a horizontal turbine with three blades. With a diameter of approx. 1.2m, it is also important to be able to brake or regulate the turbine properly in strong wind conditions.

The final goal is to develop a wind turbine that generates energy from wind power and adapts as autonomously as possible to the current wind conditions.

That is why an active pitch adjustment of the wings is planned. In addition, a mechanical brake and an electronic brake via the disc generator will be included.

The disk generator also comes from the 3D printer. I also built a small machine to be able to wind the coils automatically and with a much better quality compared to manually winding them.

More information on all these topics in the log. :)

Brief overview of what this is about:

  • Wind turbine build from 3D printed parts
  • HAWT design
  • Rotor diameters from 0.5 to 1.2m possible
  • 3D printed wings
  • Uses a 3D printed disk generator for energy generation
  • Safety functions through active pitch adjustment of the wings, mechanical brake and electronic brake function via the disk generator
  • Can be printed using any "normal" (20x20cm bed size) FDM Printer

WinDIY on its first test run:

Side view:

(Wind vane will be added soon)

WinDIYs pitch actuator:

More infos: and here


  • Test the heat sink bracket for the load resistors: pending
  • Design controller circuit: pending
  • Create assembly instructions for hub: pending
  • Create assembly instructions for wing: pending
  • Create assembly instructions for the wind vane: pending
  • Create assembly instructions for turret: pending
  • Create assembly instructions for main axis: pending
  • Design adapters for other standpipe diameters and also square bars (including clamping jaws): pending

Finished features:

1. Why (in general):

I actually started this project as a follow-up project from the Nerdiskerator (infos here: This is a disk generator that I initially built as a kind of experiment. When I had successfully built it up I thought "Just putting it on the shelf is a shame". (I like to build things that have a practical use even after they have been set up.) So the Nerdiskerator should also get a practical use and maybe even relieve my electricity bill a bit. :) 

Since I have no flowing water in the area, a water turbine is unfortunately not an option to drive the generator. Fortunately there is wind here where I live. Not much, but certainly enough to be able to harvest some free watts from Mother Earth here and there.

When researching for available and easy replicable wind turbine designs, I noticed that there is a lot of scattered information about building a wind generator yourself. Unfortunately, many of them were old and none of them seemed to me to be easy to replicate. I was hoping somehow that someone out there had already developed a design for a wind turbine that could easily be "reprinted" with a 3D printer.

I found a couple of designs, but either they weren't suitable or consisted of parts that could only be printed with very large 3D printers. Unfortunately, there was not a 3D printable design for a HAWT wind turbine as I imagined. I was looking for something easy scalable, easy adaptable and of course it should be safe. 

Since I couldn't find something like this I decided to develop an own design of an easy replicable HAWT wind turbine. :)

2. Why a HAWT-Design:

First, because of the many technical arguments:

HAWT wind turbines are more efficient because all blades work in the wind at the same time. In addition, none of the wings prevent the effect of the wind like it happens with a VAWT-design wind turbine.

Second, as a personal challenge:

I like the various technical problems that have to be solved here. Additionally, at least until now there are not many DIY HAWT designs for wind turbines available.

3. Why build yourself when you could simply buy it?

Of course you could also buy such a wind turbine....

Read more »

  • Brake actuator for the mechanical brake

    Fab08/10/2020 at 15:12 0 comments

    I posted some information about the development and the build of the mechanical brake in the project log of the Nerdiskerator. Because it's also one of the main security features of winDIY i think its worth mentioning here. (I do not want to post the same content in two project logs) You can find the full log here:

    Here is some (teaser) content: :)

  • PitchAktor Mark2, Mark3 and Mark4

    Fab08/06/2020 at 17:54 0 comments

    It has been a little quieter here lately. But that wasn't because I was idling. :D In the meantime, the pitch actuator has gone through three iterations. But more on that later.

    The last few weeks I've mainly worked on the pitch actuator but also on the wind vane. I have finished a draft for the wind vane and have already partially assembled it. More about this in a separate log. :)

    In addition to the (hopefully final) design for the pitch actuator, I also worked on the brake actuator for the Nerdiskerator. I will post more on the nerdiskerator's project log soon. (Link:

    The first design for the Pitch Actuator worked, but it took far too long to adjust the blades. The stepper motor needed almost ten minutes for a complete journey from end position to end position.

    Clearly too slow to be able to react quickly to a rotation-speed change.

    Pitch actuator Mark 2
    So I thought I would just replace the stepper motor with a small DC motor. On the search for a motor I oriented myself more towards what I had available at home than what was needed for this purpose. Spoiler: I should have looked at the data sheet.

    As you can see, the motor has enough speed for a quick adjustment. Unfortunately, a lot of torque was sacrificed to reach more speed. In addition, the motor draws almost 5A when it is blocked. Clearly too much. I have to admit that I chose the engine somewhat naively according to the motto "it will fit".

    Pitch actuator Mark 3

    So I went to the next draft. And with it to the next proof that before I start an additional printing process I should really check first that everything will fit. :D

    In this design I used the same gear motor that I use for the brake actuator of the Nerdiskerator. This already has a reduction of 1:48 integrated into the housing. The motor pulls 500mA when blocked. That fits perfectly with the I2C motor driver DRV8830 that I want to use for control.

    In the picture you can see that I did not notice that the vertically aligned motor would collide with the threaded rod when it was moved.

    Pitch actuator Mark 4

    For the Pitch Actuator Mark 4, I basically just turned the gear motor by 90 degrees. Photos and videos will follow. :)

  • A couple of mind games: What could you do with a "WinDIY hotspot"?

    Fab07/20/2020 at 13:48 0 comments

    Yesterday I started designing the board for the Nerdiskerator-controller (See my thoughts about that here: 

    During this I also provided a slot for a micro SD card, for example to be able to save log files on it. It occurred to me that it is somehow a shame to use this SD card only to stuff it with .csv files. Especially if the development of WinDIY is largely completed, this SD card slot would be largely useless.

    So I got the idea that WinDIY could not only be used to provide energy.
    With the help of the SD card slot, WinDIY could also be used to provide information locally and independently of other infrastructure.
    Since I plan to use an ESP32 as the heart of the controller, WiFi functionality is already available anyway.
    It would therefore be easy to install a web portal in the controller firmware that enables information to be uploaded and downloaded to the built-in SD card via WiFi.
    Because the wind turbine should be installed at a high location/pile, WiFi reception would be good.

    A couple of usecases I have in mind are the following:

    • If WinDIY is used in disaster regions where communication has broken down, information could be disseminated in this way. For example, rules of conduct, survival tips, repair instructions and more.
    • In isolated regions, information could be provided in this way that enables orientation. A WinDIY in Alaska could, for example, supply hikers who have strayed from the path with maps. Plus electricity for the possibly empty GPS or cell phone batteries. The information on the WinDIY hotspot could also include building instructions for a shelter or instructions for obtaining/treating drinking water. Also an overview of which local plants are edible (or not edible) would be helpful. Possibilities for making an emergency call could also be saved on it. If available, an emergency call via a Lora (relay) network could also be made in this way.
    • In developing countries, the "WinDIY hotspot" (this could be the name of the function) could contain teaching materials that make teaching possible/easier for the people living around it. In this way information could be distributed free of charge (and internet connection).
    • Of course, all information such as instructions, material lists and STL files would be included on the SD card in order to repair/maintain WinDIY or clone it.

    In order to make it easier to read all this information, a manual consisting of pictograms and / or QR codes would be included at the base of WinDIY, which uses pictures to explain which information can be found here and, above all, how to get there.

    What do you think about that? :)

  • The main shaft coupling

    Fab07/13/2020 at 12:41 0 comments

    While I'm working on the adjustments for the pitch actuator, I thought it would be time to present how I imagine the coupling between the main shaft and the generator.

    In principle, I didn't invent anything new here and was mostly inspired by the principle of well-known shaft couplings.

    You can see the well-known principle in the following picture:

    It consists of two parts that are firmly mounted on the respective shaft part and are inserted into each other. A rubber buffer inserted between them dampens any vibrations that may occur.

    And here is my version of this type of coupling. I used cut pieces of a rubber mat as a damper. Unfortunately, it is very stiff. Maybe I will exchange this for a softer material. Otherwise, so far this principle seems to work well. :)

  • The "pitch actuator" in action

    Fab07/05/2020 at 14:04 0 comments

    So, I successfully tested that the current design for the pitch actuator was a failure. :D

    The pitch actuator is the part that moves the threaded rod back and forth, which in turn adjusts the angle of attack of the wing.

    The construction itself works exactly as I hoped (see video). Unfortunately, the stepper motor used (the well known 5V stepper 28BYJ-48) is clearly too slow to allow a quick adjustment of the angle. At first I thought "Oh it's not that bad" because it wasn't really meant to have to quickly adjust the angle of attack anyway. After the first complete test run, however, I noticed that a complete trip from end position to end position takes almost ten minutes.

    That is a little too long for me. :)

    But I already have an idea for improvement. At the beginning I had planned to install only two limit switches for the respective end positions and to determine the position in between by counting the step motor steps relative to the end positions. (Similar to the system used on 3D printers) In the meantime, however, I had the idea of making it possible to query the current position of the threaded rod using a sliding resistor.

    This has the practical advantage that I can also use a normal DC motor (probably a "Reely R140") instead of the stepper motor. This creates a significantly higher speed (14000 rpm) and thus higher adjustment speeds. At the same time, the current position can be tracked using the sliding resistor. :)

    Now that I know that the design works in principle, I just have to change the motor bracket a bit. :)

    Please note that the video below is 20x faster than the original.

  • First design of the pitch control actuator

    Fab07/02/2020 at 15:07 0 comments

    In the last few days, I have completed a first design for the actuator, which will be used to control the pitch/angle of attack of the wings.

    Since I recently found some cool tools for designing gears ( I was able to use them here too.
    A small gear is mounted on the axis of the 28BYJ-48 stepper motor. This then drives the larger gear which is firmly seated on the M6 threaded rod.

    Since the threaded rod is mounted in a nut, which in turn is permanently installed in the holder of the stepper motor, a rotation of the threaded rod will hopefully result in a forward or backward movement of the threaded rod.
    Hopefully this should adjust the disc in the hub of the rotor and thus the pitch of the blades. :)

    So that the current position of the M6 rod can be measured, I have installed a sliding resistor so that it is taken along by the back and forth movement of the gear/rod. So you would have to be able to use an ADC later to measure the position of the M6 rod.

    Everything a little difficult to describe. So here are a few photos. As soon as I have a complete set up, I will also post a video. :)

  • Why (I think) a new main shaft mount is needed

    Fab06/24/2020 at 13:53 0 comments

    The last two weeks I have been working on redesigning the main axis of WinDIY. The old axis worked, but had two problems.

    The first thing, it was/is quite loud. I used angular ball bearings because they are actually perfect for absorbing forces in the axial and radial direction. Conveniently/unfortunately, the gap (between the bearing components) is adjustable with these. That means you can decide by the dimensions of the housing parts how close together the individual bearing parts are.

    Unfortunately I never managed to set these properly. (Maybe I also lack the experience with these kind of bearings.) Also in order to not make the structure even more complicated, I decided to replace the angular ball bearings with radial bearings. These can take  about 10% of the radial load in the axial direction. So if you install a radial bearing that is oversized accordingly, it should also fit with the forces in the axial direction. :)
    (And I am sure that I will find another use for the - now remaining - angular ball bearings.: D)

    The second problem is that the old connection between the shaft mount and the rest of WinDIY consists of 3D printed parts. I suspect that this will not last in the long run because this point is exposed to changing forces from the front (due to wind pressure) and the weight of the wings and hub.

    I tried to show what I suspect in the picture. The whole 3D printed structure will experience forces in the direction of the arrows. Plus vibrations from the components, which may not be perfectly balanced. All of these forces are ultimately dissipated into the rest of Windiy 's structure via the spot marked with the flash.

    Therefore the flash marks the point where I assume that this structure will not last very long.
    That's why I decided on a redesign, which you can see below. :)

    The new design

    The new design of the main shaft bracket actually consists of two parts. (this is also a relief compared to the six parts in the previous design). In addition to the screws, these two parts are connected to each other by two aluminum profiles on the side of the bracket. In addition, these profiles extend into the holder on the base of WinDIY. There, this can be connected again with a long screw with the aluminum profile in the base.

    The pictures below show the mount from different sides. To give you a better orientation here a picture of the mounted main shaft mount in WinDIY: (Note: The improved mount for the alumnium profile in the base of WinDIY is missing on the picture)

  • Turret bearing, new main shaft bearing mount and vane

    Fab06/19/2020 at 15:06 0 comments

    Here is just a small update about the actual state of WinDIY

    On the pictures you can see the ball bearing that i plan to use for the turret. (Banana for scale) It is actually a radial ball bearing and is therefore only suitable to a limited extent for absorbing axial forces. Since the ball bearing is mercilessly overdimensioned (compared to the low weight of WinDIY), this should work without problems. :)

    In addition to that, I have completely revised the mount of the main axis. I will go into this in an extra log entry. 

    Short story: I think that the old storage would not have lasted long. In the newly designed version, the main axis is held by two radial ball bearings. I call this technique (which is certainly not new) "Aluminum profile reinforced 3D print". :D Please also see the pictures in case of upcomming curiosity. :)

    That being said: The past few days I've been thinking more and more about the wind vane for WinDIY. This is the last big and important part of the wind turbine that I haven't worked on yet. A first (rough) sketch of the structure and a photo of the STL file for the first draft can be seen in the photos. 

    Little cliffhanger: I stopped printing the first draft after half an hour. More on this also in another log entry. :)

    With all of these parts, I largely assembled WinDIY for the first time last weekend. I think now you can slowly see how it should look. :)

    Turret bearing:

    New main shaft design:

    Rough look including wind vane:

  • More words about safety - Electronic load as brake

    Fab06/19/2020 at 14:16 0 comments

    As you may have noticed, I am also trying to equip WinDIY with a few (in my opinion) important security systems. I planned a mechanical brake and the possibility to adjust the angel of attack on the wings.

    In addition, I would always want to keep the generator in a certain load state via an artificial load.

    Because what naturally always brakes a wind turbine is the load on the connected generator. You already know this from a small bike generator: as soon as the bike light is switched on, you have to push the pedal harder. What you can also notice (admittedly with a little sensitivity in the legs) is that the resistance of the dynamo increases with higher speed.

    The principle is also clear for the wind turbine: the more load connected to it, the more the wind has to exert itself to accelerate the turbine. To a certain extent, when the wind is constant, you can also regulate the speed by regulating the load.

    This means that by controlling the energy drawn from the generator, you can also control the braking effect on the wind turbine and thus protect it from excessive speeds. But what to do if the load connected to the wind turbine is not large enough? 

    If, for example, a connected battery is already fully charged, there is no way to load the generator of the wind turbine. There is simply no way to get rid of the produced energy.

    In this case, the load on the generator would decrease, which in turn would reduce the braking effect. If strong winds occur now, this can lead to dangerous speeds of the wind turbine.

    In order to have a "reserve load" available in this case, I planned to mount load resistors on a heat sink and attach them to the boom of the wind turbine. This additional load can easily be switched on if required. The heat generated can then be released into the ambient air via the connected heat sink.

    Here a photo of the heat sink and the resistors i received and plan to integrate soon into winDIY.

  • Hub-hood, hooded-hub, hood for the hub... The hub gets a cover :)

    Fab06/14/2020 at 12:34 0 comments

    The WinDIY hub is literally at the forefront. In addition to the many forces that it has to endure, it also includes mechanics that are quite important for the function of WinDIY.

    In addition, most of the mechanical parts come from the 3D printer and (of course depending on the material) are sensitive to weather influences such as UV light.

    To protect the mechanics and the general structures of WinDIY at least somewhat from the weather, it is of course also provided with a hood. After assembling the hub, this is simply attached from the outside with three screws. :)

View all 21 project logs

Enjoy this project?



Josh Starnes wrote 09/19/2020 at 14:33 point

Hey guy , love your project! I saw you were looking for a solid way to test loads. You can use resistors 50 watt ones and just attach them to a bar of aluminum for the heatsink. Put alligator clips on the wires for each resistor and you can customize the load. Start low, then add say 25 watts at a time and monitor your systems performance and heat.

  Are you sure? yes | no

Fab wrote 09/20/2020 at 08:53 point

Hey josh,
thanks for the hint. :) I'm planning something similar. Using one really low resistor (0.1ohm) and then regulate the load with a pwm. :) You can also find more infos here:

  Are you sure? yes | no

Martin Lorenzo wrote 08/13/2020 at 12:22 point

Such an awesome project, congratulations! How much electricity do you plan to produce with one turbine, I wonder? It would be so cool if everyone could build their own wind turbines and use it in their houses in the future. I will try to build it for my garden in my real estate in Budva as soon as possible!

  Are you sure? yes | no

Fab wrote 08/14/2020 at 20:05 point

Thanks a lot :) Yeah that would be really awesome and this is also one of my goals: To build a wind turbine that can be self made and easily repaired. :)
The power output is not yet clear. I'm still working on the test stand for the disk generator that I plan to use. You can read details about that here:

  Are you sure? yes | no

Bram Peirs @ fWtW wrote 5 days ago point

Hi Martin, you might also check out our windturbines:

  Are you sure? yes | no

Chadd Van Komen wrote 07/13/2020 at 13:58 point

Hello Fab,  This design is great, I love how far you have taken it. Do you have any test data on the generators output (for wind or with belt testing)?

  Are you sure? yes | no

Fab wrote 07/15/2020 at 13:39 point

Hi Chadd, thanks :) The only tests i did with the generator were performed by driving it with the power drill. So with limited mechanical input power. But i finshed the second revision of my teststand recently ( but didnt had enough time recently to test the teststand. Hope to do it soon. Will let you know here. :)

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates