We've just launched our map. Add yourself by clicking here!

close

Recorder instrument made out of recycled straws

Tagged: , , ,

This topic contains 21 replies, has 3 voices, and was last updated by  Siemen Cuypers 2 months ago.

3
Siemen Cuypers siemenc

Recorder instrument made out of recycled straws

08/04/2019 at 18:30

I was asked if I could create a recorder instrument (not being a native English speaker so I might call it flute every now and then) out of recycled straws for an advertisement campaign. And since I used a lot of knowledge which I gathered from the forum for my research, I figured it’s more than fair to also share back my experiences and the lessons I’ve learned during the process. So I’ll write stuff down in this topic along with some pictures.To give some more details on the project: McDonalds changed from plastic straws to paper straws here in Norway and to mark that, they wanted to do something with the remaining plastic straws they had in stock. The idea would be to turn the straws into a flute/recorder instrument and they would document the process to turn it into a video.

21 replies
1 subscribers
0 saved
3 likes
sort on most likes
dedicated
08/04/2019 at 18:35
2

Since there wasn’t much time before the release of the campaign I decided to get a bit more hands-on experience with melting plastic so I first made a quick mold out of aluminium using the cnc and melted some plastic in the oven (it’s an oven for non-food experiments we have at our coworkingspace). I squeezed the mold together manually and this worked way better than I expected it would and that gave me more confidence on the feasibility of this project.

Attachments:
dedicated
08/04/2019 at 18:41
1

Because of the time limit I decided it would be wise to look into 2 ways of producing the flute. One would be the manual approach, as I did with the test-mold I made. And the other would be using the injection machine that @martingautron has been building. So I started working on the mold and doing research on how to produce it. It’s actually quite interesting to learn about how a recorder is made but it also provided some extra challenges in producing it. So I decided to first make the body part. I modeled and produced a 3 part mold and started testing with that in the manual way.

Attachments:
dedicated
08/04/2019 at 18:46
1

I got a few mixed results by trying out different techniques: First I wanted to completely fill the two sides of the mold with molten plastic and try to push the core in while the plastic is hot. After 2 attempts with different melting techniques I concluded that this wasn’t going to work. The viscosity of the plastic was just too high and the plastic would harden whenever I tried pushing the core in. I also had the problem that I constantly had to take the mold out of the oven to fill it with more straws since they take so much space for so little material. On the picture below you can see a failed attempt with the mold not completely being filled when I pushed the core in.

Attachments:
dedicated
08/04/2019 at 18:50
2

However, what did end up working was filling both halves of the molds with molten plastic and then laying the core in between them and sandwiching that together. It’s quite time sensitive since I wanted to keep the core cold to prevent the plastic from sticking to it. But at the same time this makes sure the plastic starts hardening quickly. It was super satisfying to get this piece coming out the way it was intended. And the texture looks just awesome. One tricky thing was to get the core out. Even though it was tapered it required still a bit of fiddling before it would come out. The best way was to actually keep the plastic part in the mold and use the gaps I provided to put a flathead screwdriver in between to pry it open. That was a nice idea I took from @andyn ‘s adjustment to that peg mold.

Attachments:
dedicated
08/04/2019 at 18:54
2

For the front part of the flute I had to make a 4-part mold. This was quite a bit of a brainteaser. To make things not more complex I decided to cut the opening which creates the sound afterwards by hand. Then the sound could be tweaked manually and tuned (same reason why I left the holes out of the body part). The same goes for the block that gets inserted into the front piece and guides the airstream in the right direction. That part I just created separately by filling up a tube with molten plastic and then turning it to the correct diameter. I was a bit skeptical as to whether I would manage to make the flute actually make sound but after doing a lot of research I managed to get it too work. The tweaking of the holes and its diameters and the insert I left up to the guy that would be playing the flute, since he’s a flute maker and knew more about that than I do.

Attachments:
dedicated
08/04/2019 at 18:57
2

After the manual testing I handed over the mold to Martin so he could tweak it and try it out on the injection machine. There were a few challenges which had to be overcome (filling the injection machine with straws/hollow cylinders is cumbersome) but in the end he made it work. What we noticed was that the injection machine tended to blend the colors of the straws more together turning it into a pink-ish color in comparison to the nice colors of original straw with the manual technique. On the other hand, it seemed like the injection machine gave a nicer finish on the inside of the flute and it seemed like it was easier to get the core out. See the attached picture, the one on the left is made using the injection machine. The other two were made using the oven.

Attachments:
dedicated
08/04/2019 at 19:00
2

Since I ran out of time and had to deliver the flute to the fluteplayer/maker I didn’t manage to make proper pictures of the flute fully assembled which is a bit unfortunate. But perhaps I manage to make another one soon when I find the time. In the meantime here is the final video that came out of the project:
https://www.youtube.com/watch?v=YxmjmB40ljs

dedicated
08/04/2019 at 19:02
3

And some more visuals of the mold design of the recorder.

Attachments:
new
08/04/2019 at 20:15
0

insane awesome great ! thanks for sharing! Would be great if you can share the files 🙂 (working on a PP library on the side)

i just noticed your large CNC table, how is this holding up ? the finish don’t look that great, i hope it’s just speeds and feeds. I am asking because we’re also now facing a little disaster building machines and molds fast enough. I was thinking to build not only a plasma but also table at least 2m x 1.5m, in a way it can cut Aluminum, wood, plasma and weld, with tool changer, etc.. as well mark tubes for holes. Building this from Aluminium beams seems tempting but I guess I stick with steel 🙂

g

dedicated
08/04/2019 at 21:18
3

@pporg
I think I can share the files. However, I’d like to emphasize that these can greatly be improved for being used in a proper production setup. So I don’t encourage people from blindly spending money or effort on having the molds produced before they realize what they up to. Perhaps it’s more useful if I provide some kind of image on the mold to explain design decisions? What do you think?

We have a 2440×1220 cnc-machine which is mostly being used for milling wood but I’ve been milling aluminium on it with a good finish. The image below has also been machined on the same cnc. The difference here was that I was milling with a ballnose bit and that seems to require some more experimentation for getting a superb finish.  I think it has to do with the fact that you are dealing with different speeds depending on which side of the ballnose you are milling when milling a 3D object. The closer you get to the center of the bit, the slower the speed because the diameter decreases.

Attachments:
warrior
09/04/2019 at 13:06
2

WOW! super nice project!

 

The moulds look awesome and having done similar things I can appreciate the difficulty involved and the amount of thought that goes into something like this.

 

Any reason you didn’t turn the tapered core on a lathe? It’s possible to achieve a much smoother finish, you can even polish it like a mirror and it should release more easily. Yes the ‘screwdriver slots’ are a useful little trick. I often also use jack screws and ejector screws to open mould halves and push parts out.

 

Coincidentally I’ve also been playing around with straws recently (not as impressive though):  https://youtu.be/L5VtpPch23I

dedicated
09/04/2019 at 13:51
2

Any reason you didn’t turn the tapered core on a lathe?

@andyn Good question! Yes, for 2 reasons:
A: I didn’t know we had a chuck for holding square stock for our lathe. Afterwards I found out we do have it.
B: I don’t have the skills to do this manually on the lathe to get the correct angle for tapering. (nor do I know if it is possible on our lathe)

I did end up polishing the cores to make it easier to take out. And with the injection machine this went quite ok but with the manual casting it seemed to stick to the core (even while using different anti-stick liquids). It could be related to the fact that I was cooling the mold down with water when using the injection machine, whereas the manual way I just let it sit and cool down. What do you think?

Attachments:
new
09/04/2019 at 13:54
0

@siemenc, no worry about the files 🙂 i think it’s clear that all this here is garage stuff 🙂 Anyways, by putting this into the new library system I hope people keep improving it due to a built-in version system :-). Making a zip of whatever you have is fully enough , best is to have the original CAD files (Fusion360,Solid Works). I do add remarks, links to here etc… In June or so the very library should be accessible to anyone, it’s basically just a file – system with a pretty interface, hosted on Github. Thanks for considering to share this. I think this kind of work makes easy for everybody to take off 🙂

new
09/04/2019 at 14:04
0

to make a taper on the lathe : you need a lathe with toolpost you can rotate. the angle gauge is usually not precise so you need to use a level indicator and some math to dial it in.

new
09/04/2019 at 14:12
0

@andyn, something you didn’t tell us, making ‘punisher’ logos from straws ?

dedicated
09/04/2019 at 14:57
2

best is to have the original CAD files (Fusion360,Solid Works).

Not sure if that’s the best in this case. I drew it in Rhino, which is a non-history CAD-software unlike Solidworks or Fusion. I find this the easiest/quickest to work in and allows for a more iterative 3D modelling process but it has also its obvious downsides. I’ll try preparing it as a STEP and a STL file, so it can be opened in most CAD software.

Coincidentally I’ve also been playing around with straws recently (not as impressive though):  https://youtu.be/L5VtpPch23I

What a coincidence! Yeah what I ended up doing was to just use scissors and cut 4 straws at the same time in smaller pieces for the injection machine. For the manual way I used an old bread-mold and melt the straws in there in the oven to one blob and then used a spatula to scoop the molten plastic into the molds. Then I would put the mold back into oven, to make sure the plastic wasn’t solidifying before I clamp the molds together.
Using just a standard oven is actually a very great and low-tech way to get started with recycling plastic without having to build any machines, I was pleasantly surprised by that.

Attachments:
dedicated
14/04/2019 at 10:35
2

@pporg
Here are the .STEP files of the molds. Let me know if you can open these.

Attachments:
  1. 11_Flutemold.zip
new
14/04/2019 at 11:17
0

@siemenc, thanks a lot for sharing ! I placed the files into the new library here.
I will update the molds for quick release bolts, all the way around, as soon I can.

thanks again,
g

dedicated
15/04/2019 at 08:29
2

One improvement I can already suggest for is for the mold of the body part. In the current design the core can still move sideways in one direction, since It’s not constraint. So the mold needs to get something that prevents it from doing this. During the molding I solved it by putting first a clamp there that covers the 3 parts of the mold, to make sure it doesn’t miss align.

new
15/04/2019 at 08:34
0

yeah, thanks, the Aluminum to reproduce this here will arrive sometime this week; so in 1-2 weeks I can start adding the CAM. We have somebody here who does this from wood for a living, takes him a full week to make one, let’s see what he can add 🙂

dedicated
15/04/2019 at 12:46
1

@pporg
In that case here’s also the 3D model of the flute after demoulding. Perhaps he can have a look at the shape of the flute as well. I made some adjustments along the way to the diameter and length of the recorder to fit my production process. I think, in comparison to the industry standard, the body should be a tiny bit longer and the diameter a bit bigger.

Viewing 21 replies - 1 through 21 (of 21 total)

You must be logged in to reply to this topic.