My boomerang

I build close to fifthy boomerangs over more than 20 years, till I got able to carve my first really perfect boomerang. Lots of items return to the thrower when they are tossed in the air. A stone thrown upwards falls back, a frisbee thrown high upwards in the wind, commercial boomerangs that return once in a while by mistake... To me a real boomerang is something you throw like a hatchet. It makes a long and complicated flight that seems to last an eternity. It rises in the air, changes direction, changes again... Several times you get convinced it will never return. Then it changes again direction. It passes behind you. It halts in the air, still rotating. Then it gently descends towards you, like a plane or a helicopter landing, rotating horizontally. You just have to extend your arms to catch it. The flight of a real boomerang is a mystical journey.

My best criterion to distinguish my favorite kind of boomerang is you have to throw it holding it vertically like a hatchet or nearly. At most at 10° to the right from the vertical. You throw it towards the horizon, with a pinch to make it rotate (just like a hatchet).

best boomerang throw angle

Other kinds of boomerangs exist. One kind flies a neat circular path and is thrown at an angle around 30° up to 60° from the vertical. That's the kind used in boomerang jugglery. Yet another kind is thrown nearly flat, up to 90° from the vertical. It makes a chandelle then returns at fast speed towards the thrower (very impressive, hard to catch and sometimes painful). That's the kind used for long distance boomerangs.

Some answers to common questions:
Below is a 50 DPI scan image of my final boomerang. The wing chord is 40 mm and the thickness is a little less than 4 mm. The leading wing is the vertical one in the picture. The trailing wing is the horizontal one. When throwing I hold the boomerang in my right hand, by the end of the leading wing. The trailing wing points in the direction I throw. It's really just like throwing a hatchet. If thrown, the boomerang displayed below would turn counterclockwise and travel towards the left of the page. Click on the picture to get a 300 DPI scan (1 MB size):

Test boomerang with weights and turbulators

The boomerang displayed above is rather dirty because it endured several days of experiments. I don't varnish my boomerangs because that makes the wood feel dead. Besides the polyurethane varnish I have makes the wood more fragile to local impacts.

You can see on the picture I glued two copper coins inside the wing ends. This is because while I tried out and modified the boomerang, the elbow got heavier and the ends got lighter. So I compensated by hollowing the ends, insert those two coins and glue them tight with epoxy glue. They are very lightweight so my boomerang cannot be called a "weighted" boomerang I think.

You need no special skills nor force to launch it. Just a little training. It doesn't fly very fast and it doesn't rotate at high speed. It is not much sensible to the wind during the flight. It won't make much damage if it hits somebody (some pain anyway, maybe even broken spectacles).

Let's first talk about what doesn't matter much.

The shape

As you can see there is nothing special with the shape. The angle between the wings is almost exactly 90°. For my next boomerang I choose a little more than 90°, because at 90° a boomerangs looks like a halve painting framework.

The boomerang elbow is part of a circle. The center of this circle is roughly situated at the center of gravity of the boomerang. I choose this so the elbow won't brake the rotation of the boomerang. I believe this is not very important, anyway I like the principle of it. The picture below shows the boomerang, its center of gravity and three circles. I wrote a little computer software to design the boomerang to get the center of gravity at the right place. Once the boomerang was built I used the standard method you learn at school to verify the center of gravity: hold the boomerang vertically by an end and let it rotate towards the ground, use a plumbline, draw the verticals on a paper taped to the boomerang (the white disc behind), intersect the verticals...

center of mass of a boomerang and efficient arrays

The three cyan, yellow and magenta circles are important. The inner cyan circle determines the range of the boomerang. I'd rather say the ratio between the external magenta circle and the inner cyan circle. The closer the cyan inner circle is to the outer magenta circle, the farther the boomerang flies. Little boomerangs with a cyan inner circle close to the outer magenta circle, fly very far away. While large boomerangs but with a comparatively little inner cyan circle, may fly less than twenty meters away. The middle yellow circle shows where the wings begin (and the main aerodynamic lift forces) (red little circles).

The wing ends are simply rounded. This is not mandatory. I made good boomerangs with square ends (just two rectangles of wood glued together). Anyway rounded ends are best, for several reasons:

Wing profile

The wing profile is roughly this, then rounded. The trailing edge is left, the leading edge is right.

simple boomerang wing profile

The profile in the middle of the elbow is roughly this; two janus leading edges:

All other profiles are smooth transitions between these two profiles. Also the shape on the rounded ends is a smooth transition between the leading edge and the trailing edge. I make sure there are no irregularities and the wood is polished with very fine sandpaper.

I carve the rough shape of the profile with an Opinel knife. Then I use very coarse sandpaper (grain 40) to get a precise squary shape. Then I round that shape out with softer sandpaper (grain 200). Then I polish it with even softer sandpaper (grain 600). I end with the smoothest sandpaper available (grain 1200). Don't use low-cost sandpaper for wood. Use high quality expensive sandpaper, for example the kind that can be made wet (but don't wet it).

The profiles don't matter much. I never got significant enhancements by changing a boomerang profiles. Differences in the profiles do change the boomerang behavior if they modify the wing incidence or create turbulences. Yet the profiles themselves change nothing. So just carve out a good profile, adequate for slow speeds, and everything will be fine. If you need to change the incidence or create turbulences, maybe better twist the wings or use dedicated turbulators. Your choice.

In order to carve good profiles you must understand a few facts:

A very important parameter for wings flying at low speed is their thickness. Look at the wings of the first airplanes or at ventilator blades. They are all quite thin. Don't exceed a thickness of 1/10 the wing chord. 1/10 is already very thick.

This is roughly the best shape for a wing profile. The leading edge on the right is drawn quite steep. This is adequate for model gliders flying at very low speeds and strong angles of attack. It is a bit overdone for a boomerang I think:

wing profile

Wide curve at the trailing edge, steep curve at the leading edge. The drawing above shows an infinitely thin wing (which is best for slow speeds). A real wing profile is such an ideal curve yet with matter around. To find the curve inside a wing profile you draw circles to find its middle line. (While drawing the picture below I realized the cuts I drew above for my boomerang profile are not the most adequate.)

wing profile

As you can see the profile I use very roughly matches the ideal curve. I experienced no need for a better match.

The material

A good boomerang needs to be both quite flat and very rigid. You can use whatever material to carve a boomerang out of it provided it matches these two conditions.

You may be intrigued why I glued two branches together although my boomerang is carved out of a thin plywood plate (Polish multipine). The reason is when you bend that plywood in one direction it is very rigid, yet in the other direction it bends a little. I couldn't afford for aeronautic quality plywood. So I cut the two wings in the same rigid direction then glued them together with Araldite epoxy glue. (My advice: heat the boomerang while the glue hardens. At 60 C° the glue becomes ways stronger.)

The heavier the material, the better, though I'm not sure if a density of much more than 1 would yield a good result. A density of 0.5 seems poor to me. My actual boomerang has 0.8. My next boomerang, in oak wood, will have a little more than 1.

Now let's talk about the three key parameters: turbulators, wings twist and wings dihedral.

The turbulators

The turbulators are situated on the upper side of the leading edges of the wings. They are those gummy flat knobs on the wings middle part and the carvings on the wings ends. The carvings are displayed on the right of the picture below. They aren't much visible and quite narrow, anyway they are mandatory:

turbulators on leading edge of a boomerang wing

This is a schematic drawing of these two kinds of turbulators. Please note I do not pretend the positions and sizes of these turbulators are the best ones. I simply don't know:

boomerang turbulators

When a wing passes trough the air at a high speed, the air follows a neat and reliable path around the wing (the less chord, the more speed needed, due to the Reynolds number). The problem with a boomerang wing is it travels at quite a slow speed through the air (especially the parts on the second (yellow) circle). When a wing travels too slowly through the air it doesn't lift at all, because the way the air passes around the wing engenders no lift.

Around boomerang wings, the behavior of the air can be very unpredictable. You may have the ends lifting yet not the middle part. Or one wing lifting and not the other. It is completely unstable, oscillating, erratic... In one word: unpredictable. You throw a little different way, little gusts of wind occur during the flight, a little bit of dirt on a wing... any of these can cause strong differences on which parts of the boomerang lift. Hence the flight path of the boomerang changes.

The remedy are the turbulators. They engender little turbulences at the upside of the leading edge of the wings. These turbulences force the air to follow the shape of the wing, hence to lift. That's what makes a boomerang reliable like a Swiss clock. Many different kinds of turbulators exist. I didn't experiment much with them. I'm satisfied with the solution I found: little polyurethane flat domes for the middle parts of the wings and carvings for the wing ends. The effect of these turbulators is miraculous. They can make an excellent boomerang out of a mere piece of wood.

Turbulators allow to use thicker wing profiles.

I suppose ethnic drawings and carvings on genuine Aboriginal boomerangs act as turbulators.

These are three other common kinds of turbulators used on boomerangs. A steep angle on the upper side, holes through the wings and an angular shape:

boomerang turbulator shapes

I never tried to heavily weight a boomerang. Anyway this is reputed to better a boomerang. One reason may be the places where the weights are inserted become turbulators. Let's be serious, a good boomsmith smooths out the surface so the boomerang gets the exact some shape as before it was weighted. I think that if you weight a boomerang while keeping its center of gravity at the same place, there is at least one good reason this betters the flight: a heavier boomerang needs to be thrown harder and rotates faster. Hence the aerodynamic behavior of the air around the boomerang becomes more reliable. This is just a matter of personal choice but I don't like the idea of weighting a boomerang. First because I prefer slow and beautiful flights. Second because this makes the boomerang both more fragile and more dangerous.

Wing incidence

Second key is the respective incidence of the wings. The leading wing must lift more than the trailing wing. It may even be that the trailing wing actually forces downwards instead of lifting! Why this? If you already studied boomerang theory you heard a boomerang behaves like a gyroscope. It's like a rotating disc.

Let's proceed step by step to explain:

Imagine a boomerang is simply rotating. It has not been launched, it is just immobile in the air yet rotating fast. It acts like a simple propeller. Say the rotating boomerang forms a vertical disc. It blows air towards the rear, like a propeller. A lift force pushes the boomerang frontwards. Where is the resultant of that lift force located; its average position on the disc? It is situated right in the middle of the rotating disc, on the center of gravity of the boomerang. The picture left shows the rotating vertical disc. The picture right shows the disc from aside (that's what you see when the boomerang left your hand). The boomerang is rotating counterclockwise. The lift force is symbolized by the 3D red arrow. That average resultant point where the aerodynamic forces lift the rotating disc is of key importance:

rotating boomerang and lift force

Suppose the boomerang wings have no lifting profile. There is no aerodynamic force (no red arrow). Suppose you throw that boomerang like a hatchet. It will simply move straight ahead, like a hatchet or a stone. The picture below is a helicopter view from such a trajectory. We are high above the ground and look down at the person who will launch the boomerang. The person throws the boomerang to the North. The boomerang moves in a straight line, falling to the ground, like a stone thrown towards an object:

Lets reintroduce the aerodynamic lifting force (the red arrow). The rotating boomerang is a gyroscope, so its plane of rotation tends to be stable. In our helicopter view the aerodynamic force (the red arrow) tends to push the boomerang towards the West. The thrower throws the boomerang again straight towards the North. Yet the aerodynamic force makes it drift towards the West:

This flight path is wrong. Because of a key phenomenon we must now introduce. Look at the picture below, the left one. Suppose the rotating vertical disc moves towards the left of the page. When a wing passes in the upper part of the disc, it will get more wind than when it passes in the lower part of the disc. Hence the wings experience more lift force in the upper part than in the lower part. This shifts the aerodynamic resultant average force upwards. The red arrow moves upwards in our drawing (this is a key to helicopters.):

rotating boomerang and lift force

The average force is now situated above the center of gravity. What happens in such circumstances? At first thought you would expect the disc to turn towards the ground:

This is wrong. Indeed the rotating boomerang behaves like a gyroscope. It turns this way, very slowly:

rotating boomerang and gyroscopic precession force

That's the gyroscopic precession. What path flies such a boomerang if you launch it like a hatchet? It flies the circular path pictured below. Yet it hits the ground before it can close the circle and get back to the launcher:

Why does the boomerang hit the ground? Well because there is no upwards force to keep it in the air. Just like there is no upwards force to keep a stone in the air. The aerodynamic resultant force and the gyroscopic precession curve the flight path to the left. But no component of the force tends to keep the boomerang above the ground.

A first solution to keep the boomerang in the air and make it fly a full circle, is to launch it with an angle to the vertical, between 30° and 60° depending on the boomerang. Then a part of the aerodynamic lift force is directed upward and can keep the boomerang in the air:

That's the way some jugglery boomerangs and some toy boomerangs are launched. That's the way you launch simple triblade boomerangs:

simple triblade boomerang shape

What I want is a boomerang I launch holding it vertically and it comes back in a near horizontal position. That's the way you get poetic flights. I want the boomerang to fly a circular path and I also want it to go lay horizontally. How can this be achieved? Answer: the resultant lift force must be shifted frontwards. Indeed that way the boomerang/gyroscope will tend to rotate a way that gets it horizontal:

rotating boomerang and correct forces

How can we shift the resultant lift force forwards? My favorite solution relies on the fact the two wings don't have their maximum lift at the same place of the rotating disc. The leading wing has its maximum lift when it is vertical, which happens when it is forward (to the left on the pictures):

boomerang with wind on leading wing

While the trailing wing has its maximum lift when vertical and backward (to the right on the pictures):

boomerang with wind on trailing wing

So, simply privilege the leading wing lift and decrease the trailing wing lift. I do this by increasing the leading wing incidence and decreasing the trailing wing incidence. This needs to be tuned. If the boomerang doesn't close its flight path and lands left of me, I know I didn't privilege the leading wing enough. If the boomerang flies a path in the shape of an alpha letter and lands right of me, I know the leading wing is too much privileged and has to be decreased:

There exists a curious variant to the alpha shaped path. The boomerang obviously starts flying an alpha path yet it will bend its end and get back to the launcher anyway. What happens is the boomerang gets flying horizontal in the middle of the path (too soon, due to the leading wing lifting too much). The boomerang continues its slow rotation (gyroscopic precession) and gets turning vertically again. That's what bends the end of the path towards the launcher. The boomerang arrives at the launcher turning nearly vertical and flying quite fast, which is a bit hard to catch:

The boomerang pictured below is made out of cambara wood (a €5 slat bought in a store, cut in three sections, glued together then profiled with lots of grain 40 sandpaper) (the cut in the middle of the picture appears because two scanner pictures were pasted together). It makes a beautiful "bended alpha" flight. It needs to been thrown with a lot of force, makes its circle far away, slowly flying above trees like a bird of prey. It seems to have forgotten about me. Then it bends it path, comes down, accelerates and seems to be searching for me like a missile.

shaped wooden boomerang

Other methods exist to privilege the leading wing:

lift on bigger leading wing of a boomerang

increased lift on leading wing of a boomerang due to an excentered center of mas

triblade boomerang shape with bended wings

Put the wings closer together. That way the turbulences created by the leading wing will hamper the trailing wing and decrease its lift:

boomerang shape with a sharp angle

I change the angle of incidence of the wings by twisting the wood. I hold the middle of the wings above a camping gas fire (hence the little burnings you can see on the big scan). Once the wood is hot it becomes bendable. Ensure the inside of the wood is hot too, not only the outside. Be patient and don't heat too much. Force with your hands and keep the wood bended while it cools down.

boomerang wing tip incidence

I recommend you make all parts of the wings exert an aerodynamic force. Both the middle parts and the ends. I made boomerangs with for example a trailing wing build to be neutral; exert no lift force. The result is poor. No matter the aerodynamic force is upwards or downwards. What matters is there is a force. On my boomerang the leading wing was twisted about 7° upwards and the trailing wing 7° downwards.

Two remarks about other kinds of boomerangs:
Please note the explanation I give why the boomerang goes flying flat is partially unsatisfactory. It doesn't explain why the boomerangs ends flying really flat above the ground. Actually I don't know what exactly happens. Anyway the explanations I give here are all I need to build and tune my boomerangs.

The dihedral

Third key is the dihedral. A quality boomerang yet with no dihedral tends to continuously rise in the air and come back down a little loosely, once the energy is consumed. You have to force the boomerang to stay close to the ground (yet at a few meters altitude). You do this by tuning a negative dihedral:

boomerang shape with a negative dihedral

Again, I bend the wood downwards using a little camping gas fire to heat the middle zone of the wings. Ideally the bending should be constant over the whole boomerang length. I just bend the middle of each wings.

You need very little bending. It's almost invisible. It is far overdone on the picture above. Yet it changes the quality of the flight dramatically. Again, it becomes precise and reliable like a Swiss clock.

The problem with cheap wood is it looses part of its bending after a few days. That's one more reason my next boomerang will be made out of oak wood.

Some authors recommend to carve the above side of the wing tips strongly downwards:

To me this is a way to achieve a beginning of a negative dihedral. I don't use this. First, because this start of a dihedral is not enough. I need to bend the wood anyway. Second, because it makes the boomerang ends more fragile and more dangerous.

I believe some tricks to better a boomerang only tend to brake it in flight. That way it doesn't fly too high. Its a concurrent method to the dihedral. I favor the dihedral because it preserves the boomerang energy and ensures a cleaner flight.

It might be that a downward angle of incidence of the wings avoids the need for a negative dihedral.

Commercial boomerangs

I bought several boomerangs, from different manufacturers and in different stores. All except one had boasting commercial arguments: designed by a world champion, made out of aeronautic plywood or carbon fiber, beautiful paintings, extensive tutorial, only €1 price... All are really very bad, they can't be named boomerangs, except the one with no commercial arguments. It is the model "Sirius" from the well-known model aviation brand "Günther", pictured below. It really is a fine and enjoyable boomerang. Its sole default is it has no negative dihedral. So it rises high in the air and doesn't come back to the launcher with a lot of precision. Anyway that can be tuned with a micro-wave oven and a little skill. (Actually, one little commercial boomerang I thought to be quite bad returns correctly when it is thrown by a child or a woman. Is suppose this is because it needs a high rotation speed and for this a little hand is better.)

Sirius boomerang from Gunther


As I mentioned in the text, once a boomerang is made the right way it is no more sensitive to wind turbulences or the like. Anyway if the wind is really strong you have to compensate for the drift it will cause on the boomerang flight path. The common advice is to turn yourself towards the wind then rotate 45° to the right and launch the boomerang in that direction. My experience is I often have to rotate 60° up to 90° to the right. I believe the 45° to the right is for boomerangs you launch nearly flat and that make a chandelle.

Also I know the direction of the wind is not the same at the height of my head than at the flight height of the boomerang. I often went to a terrain where I could see a wind vane on top of a little tower. There often was a difference of several tens of degrees, up to 90° between the wind direction I sensed and the direction shown by the wind vane.

The other thing you have to do when the wind is strong is launch the boomerang a little more horizontal. Hold it say 15° up to 20° from the vertical.

When there is wind you have to find the correct angles by experimenting. Once you are trained you will "feel" the correct angles quickly.

You may sometimes get the feeling the wind direction is unpredictable. Again, what you perceive close to the ground is not always reliable. Sometimes it is better to forget about the wind you perceive and try to find the correct angles by launching the boomerang. I was able to get a majority of neat returns in a quite strong and changing wind.

Some tricks

Keep your eyes on a point on the horizon line and throw the boomerang towards that point. It can be an object nearer than the horizon but it should be on the same height as the horizon. This allows you body to better coordinate its movements when launching. When training friends I sometimes place myself about 20 meters in front of them and propose they launch the boomerang towards me like if it was a hatchet intended to hit me.

When you get to a terrain to launch boomerangs, maybe first launch low-cost plastic boomerangs a few times. This can avoid hurts to your expensive or self-made valuable boomerangs.

If a boomerang falls in high grass or bushes, look closely in the direction it fell and memorize that direction and its decor. Then mark the place where you are (by depositing an object or by using always the same place). Then only walk straight in the direction the boomerang fell. If you don't find it, you get a second chance by coming back to the marked place and get the direction again.

Boomerangs with at least a strip of red/pink fluorescent paint all around are far easier to find back.

Tend to launch the boomerang always from the same optimal place, even if this means double the length to walk. If you continuously change place by launching the boomerang from the place it fell, you increase the risk to make bad throws, to hit somebody or to loose the boomerang.

Boomerangs were birds in another life. Or they contain the soul of a bird to be born, I don't know. One thing's for sure: they love to perch high in a tree. Avoid trees or bring with you a very long fishing rod. To get back my final boomerang I had to go buy two long wooden rods in a nearby store and duct tape them together. At that time it was still quite badly tuned and I really thought of leaving it in the tree. Getting back boomerangs perched 7 meters high in a tight mesh of branches, is a sub-art of the art of the boomerang. Especially when you forbid yourself to harm the tree. Obviously the boomerang is responsible for the situation, not the tree.

If you throw a boomerang towards an object it will never reach that object because it will change direction and rise in the air. Nevertheless don't throw a boomerang towards a person, even far away. Boomerang throwers are cool and reassuring persons. Think of the boomerangs throwers that will come after you. You should not make the indigenous population afraid of boomerang throwers.

Keep in mind some open spaces are used by all kind of persons: bikers, joggers, football players... My experience is they won't come to you and ask you to let them use the terrain for a while. Instead you have to notice them, redraw and let them use the terrain. Only go back throwing once they are finished and gone.

If you get pain in your arm after a few launches this means the boomerang is not adequate for you. It is either too light or too heavy. The principle is the boomerang should leave your hand carrying most of the energy your muscles developed. If it is either too light or too heavy, part of that energy will stay in your arm and echo in your muscles and bones, harming them.

When I go launching boomerangs I cannot expect a boomerang to return perfectly on the first launch. Sometimes I need a ten or so launches till the magic builds up. Then the boomerangs seems to love me, whatever way I launch it. This also means maybe you should not go launching with all your boomerangs. Often I just take only one good boomerang (and a plastic boomerang for the first few launches). I focus on that boomerang and I get far better results than by throwing loosely all kinds of boomerangs.

I tend to launch all my boomerangs the same way (holding them nearly vertical) and tune them so they get back neatly. If you are performing your first trials with a self-made boomerang, most probable is the boomerang will return far too much to the left of you. If you don't want or cannot tune it, simply launch it holding more horizontally. If you launch it too horizontally, at best it will arrive right of you. At worst it will smash into the ground, a ten or so meters in front of you.

The content of this text is about boomerangs for right-handed persons. If you are left-handed, revert the wing profiles (leading edge - trailing edge). Also you need to turn left from the wind.


I wish to thank my friend Pierre Backers who lent me his Dremel electric saw.

Science and modern technology helped me build a good boomerang but I would be nowhere if I hadn't spent hours studying photographs of Australian Aboriginal boomerangs and read through lengthy descriptions. The current situation of Australian Aborigines is quite good compared to other minorities in the World. If you would like to do something to please me take a look at www.survival-international.org and possibly angryindian.atspace.com

Eric Brasseur  -  November 24 2005  till  April 23 2008