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


V4 Extrusion Tubes and Profiles

Tim Slab timslab

V4 Extrusion Tubes and Profiles

25/04/2019 at 11:19

Hello hello!
Time has been ticking on and we’re still cranking hard at the Precious Plastic HQ with the development of v4.. The focus for the first part of the year has been on exploring and developing techniques for the Extrusion machine. This topic will be focused on the wide topic of Tubes and Profiles, with all the successes, failures and set backs along the way.

Stay tuned for further topics as the development continues..

-Machine Development 
-Beam Production
-Extrusion Moulds
-Tubes and Profiles  (You are here)

15 replies
3 subscribers
3 saved
sort on most likes
25/04/2019 at 12:22

Ok, so the idea of making a ‘simple’ tube from recycled plastic is fairly straight forward and seems achievable, right? We’ve got a strong extrusion machine and a workshop full of other machines.. but after really getting into it, making a plastic tube (using current methods) actually requires a LOT of knowledge and  infrastructure to achieve. In the images below you’ll see:
1 – a basic render of the concept behind making a tube (hey, we can do that!)
2 – a cross section of a die (hmm, lots going on)
3 – an exploded view of the different parts needed (that looks big)

To get an idea of the scale and infrastructure needed for the process of making even a small tube you can also watch this video (pretty amazing!). Note that this example is an industrial and highly ‘productive’ system but the steps in the process of making tubes would need to be similar. Such as the die, the cooling method and pulling machine to maintain dimensions/flow.
With all this considered I began to lower my expectations of making tubes in these methods due to our current machine setup and the knowledge needed to get a good result.
My thinking is that there’s already a world out there, a massive industry making perfect tubes with huge tools and big budgets.. Now I’m not saying its impossible or this technique shouldn’t be tried but when looking at our machines and this project as a whole, I feel that time could be better spent in other areas of the development.. for now

25/04/2019 at 16:34

Making a tube without the use of a ‘tube’ die would require a different take on the method used. The idea was to use a large, flat die/nozzle to wrap around an existing tube. Here are some results.

Test: Flat nozzle

Objective: Make a tube!

Process:For this test a thin metal rectangular tube has been clamped, hammered and shaped into a desired shape. Tabs were then welded onto to the shape so that it can be connected to an existing attachment. Time was taken to ensure a flush fit between the ‘nozzle’ and the flat plate. Results of initial tests were not so successful (image 1) as the plastic began to cool before leaving the nozzle. Another attempt was made after pre-heating the nozzle which provided a slightly better result. A big downside to this design is the amount of waste left in the nozzle and the extra energy needed to heat it up.

-This design for a nozzle is not so good. Plastic did leak and left overs are ‘waste’
-A flat shape can be obtained but the process is slow.
-The flow of plastic is not very uniform.
-Pre-heating the nozzle did help.


25/04/2019 at 18:21

Test: Flat nozzle v2

Objective: Make a tube!

Process: In an attempt to improve and simplify the flat nozzle, a thin cut was made into a piece of 4mm steel sheet. The nozzle was left to sit on the machine in order get warm. However, results were not so great. The flow of plastic was in faster in the centre than at the edges and produced a ball of plastic. Sheet was flipped around and a similar result occured.

-This design for a nozzle – also not so good.
-Flat profile was not achieved and plastic deformed into a ball.
-The flow of plastic was not uniform, faster in the centre than the outside.

30/04/2019 at 12:33

Noting how the plastic flows much faster in the middle than on the edge I wanted to try fix that by drilling two holes on the edge. The idea was to offer less resistance to the edges allowing the plastic to flow more uniformly.

The results were relatively successful but I think there is still room for improvement.

If you look closely at the plastic in image 2 you can see a small build-up at the edge of the nozzle. This may be due to the rough edges caused by grinding the slot.

01/05/2019 at 16:30

Test: Flat nozzle v3

Objective: Make a tube!

Process: After no fully successful results from tests with flat nozzles, I decided to reach out to @greendaddy from Zelenew. They’ve got their flat nozzle technique working wonders! (Check it out). I basically copied their nozzle exactly and it worked first try. A piece of 1″ tube thread, cut and then squeezed in a vice. Two pieces of 5mm sqr bar were then welded to the nozzle end, with aprx 1mm gap between the two.
I built a quick jig to support a tube to work as a mould to extrude around. I was able to make a tube by rotating a metal cylinder and wrapping the hot plastic in a spiral. Thus, a tube was achieved!

-Flat nozzle and profile extrusion success!
-The flow of plastic was uniform and thickness consistent.
-If done with care, the tube can even be airtight.
-Plastic shrinks around the mould and can be very tight. Wrapping teflon paper around the metal tube as a barrier between mould and plastic greatly helped with the demoulding process

01/05/2019 at 16:49

holy moly, these are great! if those can be a little flexible, we have a in-house solution for fume extraction, i can see @s2019 being very happy to see that too 🙂

01/05/2019 at 19:36

@timslab That is really great. It opens up a whole new area to experiment. Might push me to build some version of an extrusion machine. I can’t see why @pporg won’t have an NC controlled mandrel/leadscrew attachment built for the extrusion machine by Friday.
What material is this and what do you think is the largest cross section ribbon you could lay down? How hard was the machine working to do this?
I wonder if you could get rid of the overlap by having a trailing heated finger of sorts that closes the joint.
With a heated mandrel I would guess you could do some large diameters.

Great work.

I also like your equipment dolly

01/05/2019 at 19:48

@s2019, funny you, I have all the materials (incl. a 1KW servo motor) but it’s collecting dust only right now since we’re buried under production, well and terra bytes of tutorial footage, we’re off for 2 weeks; also for to figure ways to extrude PET into pellets or direct print, see you in a while 🙂

01/05/2019 at 19:53

@timslab, if you could upload a short video of this process please … eager to see that in action , thanks a million 🙂

02/05/2019 at 01:50

Nice try @timslab !

02/05/2019 at 18:02

@s2019, for now I’ve been using PS as it’s what we’re currently using for machine testing. Depending on your release agent/mould shape I think you could make large tubes. As long as the plastic you’re laying over is a bit warm/tacky the plastic will fuse nicely. The machine wasn’t running hard. I think maximum pressure reached at nozzle was about 5bar. Also running relatively slowly (80~120 rpm).. I like that dolly too haha

@pporg, we’ve also been talking of automating the process.. could get even better results.

cheers brother! thanks for the inspiration

02/05/2019 at 18:06

Some longer and thinner tubes. Airtight and lightweight – make for some good wind instruments!

02/05/2019 at 18:19

Looks great! Time to work on your first Chihuly piece. https://www.chihuly.com/work/installations

07/05/2019 at 18:16

How about a larger one of these https://bryanmkevan.bike/2018/12/16/carbon-framebuilding-das-faserwickelmaschine/ ? You will need to move the part instead of the nozzle.

08/05/2019 at 13:01

Oooh, great link @s2019. Interesting to think of larger rigs like that for plastic..  depending on size it might be necessary to have a heatgun preheating the plastic in order for the next lay to stick. I found even on relatively small tubes, if the plastic has cooled then the next lay over doesn’t bond so well.

Thanks for sharing!

Viewing 15 replies - 1 through 15 (of 15 total)

You must be logged in to reply to this topic.

Support our projects on Patreon so we can keep developing 💪