Hello everyone! I managed to do the post that I promised last time, that is the replica of a working mini crossbow completely printed in 3D. For the occasion the post will be divided into two parts, in the first I show how I arrived to have the working prototype, while in the second I will show the final version. So, if you are not interested in all the prototyping work, once you have read the introduction you can directly jump to page 2. Also the first part will be more conversational rather than divided into paragraphs as usual.
Unfortunately for the work I do not have the video modeling as it was a project developed as try & error. That said let’s see how I created the crossbow!


Before talking about my work I must admit that this is not really my original idea, in the sense that I had a little thought of “it would be nice to make a working crossbow”, but actually I did not develop it until on Instagram I did not saw 3dyazici51 who had made it a working little one.
When I saw it I tried to understand how it was made and I found some things that I did not like, things that I wanted to change in my project. Not to remove anything from his work (which I find very nice) I noticed that to keep the pieces together he used some screws and in a couple of cases the tips were coming out of the model. This thing has made me turn up my nose because the tips of protruding screws are always dangerous, especially for a model that you want to carry around to be seen by friends or worse to a general public. Moreover, unless you create special holes as a seat for the screws, there is no real reproducibility, as the screws are inserted “randomly”. This could also create problems if a piece breaks and needs to be replaced, as it would be difficult to re-insert a screw at the same point.
That said, my project had to work around these problems, so using plastic pins.
I want to tell this story because it shows the real work behind every project of this kind. I believe that there is no one who can achieve such a thing at the first blow, unless you have already seen it and simply copy it. With this first part I want to convey the message of not pretending that a project comes out immediately, some attempt is always necessary. And as I said, if you feel stuck in solving a problem, talk to someone.

The prototype

After studying the model on Instagram, I realized that the crossbow should have been composed of:

  • The bow that stretches the rope
  • The central body, on which the dart is loaded
  • A trigger to release the rope once it is taut
  • The handle to hold it comfortably
  • The handle to hold it comfortably

Since I do not know, nor can I measure at home the elastic properties of the arc (which depend on many factors including shape, material and I would also say the percentage of filling), the first thing I had to do was study the bow, how much I could make it big, how to attach it to the rest of the crossbow and how much I could stretch the rope.
For this, taking as a reference the work of 3dyazici51 I modeled two separate arms and the front part to which they would be attached:

Figure 1 – The first version of the bow and its support

We observe the image: the central hole is created as interlocking with the rest of the body. Each arm has two holes and you can already see that in one of these there is a piece that replaces the screws.
As soon as the front and the arm were printed, I saw the holes and I was not so sure that the glued plastic pins would hold up much, I was afraid they could come off. I therefore considered the idea of using screws and bolts. At the bottom the holes were present and being metal they would hold up.
From the image you can not see, but each hole has a diameter of about 2.5mm. Having said that, it is immediately clear that the edge of the arm in which the holes are there is really small (about 1mm). This is a serious problem since that point should hold up all the tension when the crossbow is fired. To test if it would hold up, I inserted a screwdriver into one of the holes and applied tension to the arm, which broke immediately.

So I made two changes: the first was to print the arms as a single piece rather than as two separate, the second to make a single hole on the side instead of two.

Figure 2 – The second version of the bow and its support

Even here, however, I found some critical issues at the structural level. To do the test I inserted the bow in a vice and I started to pry up my arms. As before, this time the bow was immediately broken, obviously at the height of the holes.
At this point I stopped for a moment. If making holes in the arch breaks in any case, does that mean the only method is really to put the screws without making holes before? While I was trying to figure out how to do it, I talked to my dad who told me “why do not you do it interlocking? You make a piece of it and the bow fits it inside “.
Actually I had focused too much on the pins that I had not thought about this possibility, so I recreated the model. After this change, I also left the idea of using screws and bolts.

Figure 3 – The third version of the bow and its support

Until before printing ended my concern was that if the measures were not precise, the bow would have risked moving due to a possible non-homogeneous tension of the string. The thought, however, disappeared when I put extreme effort into inserting the arch in the seat: I had guessed (fortunately) the size, making it impossible to make any possible movement.
At this point, solved the problem of how to make the front, I had to understand how it was possible to stretch the bow and then decide how much to stretch the rope, deciding how long it should have been along the central body of the crossbow.
The shape of this arch, however, has created problems, in fact it is wrong. Unfortunately I did not notice immediately, but only after some attempts, when I noticed that it did not stretch as it should and above all it broke too easily. I tried varying thickness and filling but the results were always the same.
So I began to look closely at the arms of the references and I realized the mistakes made. The correct and functioning arm has this shape

Figure 4 – The comparison between the two different versions of the arcs.

In this way the bow tends more and does not present evident risks of breakage.
I found the appropriate shape for the bow, so I created the rest of the model.

Figure 5 – The definitive model of the crossbow prototype

This was the simplest work of everything and even almost immediate I would say, since I copied the operation from the true crossbow: a groove where the dart is inserted, at the end there is a recess to hook the rope into tension, and the hole to allow a pin to raise it and then unhook it.
For the pivot it is sufficient that it can move in a space between the beginning and the end of the recess of the rope, to allow respectively the engagement and the release of the latter.
You can see that the rod that makes up the trigger has two cylindrical parts that serve to keep it attached to the rest of the body. These tighten enough to keep everything together thanks to friction, without the need for other supports.

Figure 6 – The printed parts of the protype,
before being assembled

With this the work is finished. The printed pieces were only to be assembled and to prove that everything worked as it should, which he did.

One part that I left out in the prototype was the handle, as it was not necessary for operation. It’s not the best in comfort, but the mechanism works.

In these images there is not yet the rope, which we will see in the final model, but I’ll tell you right away that it is not held taut, but soft, for two reasons: the first to avoid that once armed the tension was too much, the second is that this is a toy, it does not have to hurt, so the shot does not have to be strong.
That said, how much is the range and power of the shot?
For the range I saw that it reaches about 2-3m, while the force of the blow is not excessive. Using a pencil as a dart, the tip was broken, but it did not create any kind of damage to the target (in one case the protective jacket of fencing and in the other of the books hit by mistake).

At this point you are wondering: how big is it?
Well, the bow has a length of about 25cm, while the body is about 12cm long (without a handle, let’s remember it).

Figure 7 – The prototype of the assembled crossbow


This is all for the prototype. Aware of the results, I was able to move on to the next phase, which is to create the complete crossbow.

Leave a Reply

Close Menu