V4 Beam Production
We’ve been cranking hard at the Precious Plastic HQ with the development of v4..
The focus for the beginning of the year is on exploring and developing techniques for the Extrusion machine. This topic will be focused on Beam Production, with all the successes, failures and set backs along the way.
Stay tuned for further topics as the development continues..
Test: Polystyrene Beams
Objective: Create transparent beams and explore the control of colour with PS.
@tafnstuff and I tried various beam moulds attached to the extrusion machine. The machine is in good state and using it is very friendly. We screw on each mould, fill the hopper and turn the motor on. The machine was set to 240 C and motor speed varied between 210~250 rpm. Demoulding the beams can sometimes be very hard with PS as it doesn’t shrink as much as PP or PE.
-PS can remain transparent with the extrusion process.
-Quality of finish is high.
-Beams are very hard. (Sound like a peice of rock and take long to cut through)
-One beam developed a large cavity and was completely hollow. We think this could be from the thin gold flakes (coffee cup lids) shrinking rapidly and drawing to the outer layer of black.
Test: Polypropylene Beams
Objective: Create large, strong beams
Shred large flakes into smaller bits. The machine was set to 200 C and the motor speed was varied. For most tests, I used a wooden block attached to a longer piece of wood (plunger). I hung a 1kg bucket onto the end of this plunger to create some relatively consistent pressure to the extruding beam.
-Smoother surfaces can be obtained by using a weighted plunger.
-Smoother surfaces can also be obtained with a warm mould.
-Smaller granules work better than large.
-The end of a beam seems to have better consistency and surface. Possibly due to the heat of mould and surrounding plastic.
Test: PP Beam consistency and speed
Objective: Discover the optimal machine settings
There is an image of this beam above, and I went through the same process however did not use any plunger/resistance and the mould was cold (4 C). I set the machine to a constant temperature of 200 C and ramped up the rpm and changed colour every 10 minutes. Starting with [email protected], [email protected], [email protected], [email protected] and [email protected]
-Initial section of beam was very rough, did not fill mould cavity or fully melt.
-Consistency improved with length. Plastic has more time to melt inside the beam core.
-There was a hollow channel at the end of the beam. Possibly from letting machine run with no plastic.
-With a cold mould, it may be better to start off even slower and ramp up the speeds thereafter.
Test: Polyethylene Beams
Objective: Create stronger beams. I’ve found PE to be an elusive material for the hand powered machines due to it’s viscosity even in molten state. These experiments were to learn about ease of production and the differences between PP and PE beams.
Shred larger flakes into smaller bits. The machine was set to 220 C and speeds were adjusted according to the rising and falling barrel temperatures. The plunger was weighted with 1kg.
-PE can be used to make strong beam.
-After enough time, demoulding is easy.
-Surfaces feel more slippery/waxy than PP or PS.
Test: Large PE Beams
Objective: Make the biggest PE beam possible.
We had some large rectangle and square tubes in the workshop onto which we welded angle iron brackets for connecting to the extruder. It is helpful to clean the inside of the tube before use (image 3). The process of extrusion follows the same steps as before. Machine set to 220 C and speed varied according to the temperature of the barrel. A 1kg bucket was used for resistance on the plunger.
-The rectangular tube mould broke while extruding. This happened when the resistance of the plunger rose as the weight moved further away from the mould. The temperatures spiked to 270 C and the mould bulged (image 2). Closeup of the swollen area of beam also pictured in image 2.
-The square shape worked well too. You can see the shape of the nozzle and initial plastic section in image 3.
-Cooling takes far longer with such large shapes.
-There is a fine line between too much pressure and too little. Only some sections took the shape of the mould, others remained round/organic.
-Temperature on average was 30~40 C hotter than the machine was set.
I’ve had too many bad experiences with extruding pe and ps … that’s why I am trying to develop simple ways of making beams. .. but always looking for getting more information and ideas from other colleagues. What do you think ?https://www.youtube.com/watch?v=gsnsitKldyU
Yo @wolfgang, nice videos! Thanks for sharing your process. Great to see the creative use of “low tech” tools. Quite a lot of hands on work but the results are very surprising, at first I didn’t think they would be strong. A whole new take on making beams.. really interesting!
Looking forward to seeing how your tests go with the hot sand and use of solar power
Hello, any tips about PS demount?
Some times I couldn’t extract beam from metal tube, even with brute force.
Do you use any mould release or oil inside tube?
Hey @copypastestd, for releasing the PS beams we gave the mould at least 30mins to cool. The beams we did make were also not too long.
Circular / square shapes seem to cool more uniformly and are easier to release than the rectangular shapes. So far we’ve had no need for release agent in beam moulds
In description you mention that motor speed varied between 210~250 rpm – it is not clear for me. This speed of motor, or speed of screw after gearbox? What ratio of gearbox do you have in your machine?
Do you have heaters on the mold, or are all your heaters mounted on the side just on the extruder side?
In my case, at some point the plastic gets stuck in the pipe – and the only working way to push it forward is to use heatgun (which in turn is extremely inefficient).
I am satisfied with the quality of my plastic beams (they are strong and smooth) but the process if really difficult takes a long time.
And I’ve never been able to make a bar longer than 50 cm.
Hey @copypastestd, thanks for you questions.
The speed mentioned is the speed of the screw itself, I’ll have to find out what the exact reduction is.
The only heaters are those on the extruder itself. To get a hot mould, I’ve been extruding a beam, demoulding asap and then using the mould again while it’s still hot.
What material have you been using and what temperatures? Your beams look smooth and consistent.
Perhaps the issue of not being able to extrude longer than 50cm is that plastic is not warm enough to maintain the flow or your machine is unable to maintain the pressure
Here’s an inside look at how a beam is born 🙂
@timslab; you have any recommendation for the v4 screw parameters (pitch, diameter ,… ) ? I do the next days new footage on how to make extrusions screws as part of the ‘PP CE Academy v4.1 program’ (advanced level); also adding outstanding wood auger hacks to add compression in v3.1 CE. Happy to have some input. Currently I know only of 3-4 Kw 3phase at ~1:40 and a screw diameter of around 3-4 cm. target beams are 5×5 or 8×4 if possible.
thanks in advance
Hey @pex12, a PP academy sounds great.. master class 😎. We’ve made some adaptations and improvements (testing coming soon) to the screw for the v4 extruder.. I haven’t been too involved with the engineering design aspect so I’d ask @jovinc and @peter-bas about the exact specs for the screw
A few quick questions, if you don’t mind.
Any updates on board quality? I’m curious about the feasibility of making a chair pattern like this one, with no board larger than a meter (the back slats are the largest at almost exactly a meter in this design).
What materials are you recycling for your polystyrene beam? Those look the best in the thread, but I thought PS was notoriously toxic when recycled.
I would’ve expected the best results with HDPE, if only because that seems the material of choice for production of boards at scale, like these recycled chairs. Have you had any luck improving the rectangular mold since this update? Are the boards you are making reasonably strong?
Just curiousity: Why is there the circle of color in the middle of the PE beams? That would suggest the mix isn’t very homogenous, but what causes the separation?
Also wondered what kind of plastic Dave was using in his video for beam extrusion here: https://www.youtube.com/watch?v=zNGuuSKE1pY
His board around 5:35 (screenshot attached) looks extremely nice. What material is that using? His beams, especially that one, looks, at a glance, more uniform than the ones here, but it’s hard to see them all closely.
Is there an ETA for the v4 extrusion plans?
Thanks for all the work, and keep it up!
did somebody try to dry the plastic before extruding/injecting ? Normally the drying process can take up to 2 – 10 hours depending on the plastic. That makes sense because along the coasts or on islands I remember that we pulled out up to 5 liter from a 20 sqm room, every day. However, here is a DIY variant and this is how it’s done industrial wise. We’re about to work something out for v3.1 & v4. It’s not much tech per-se and those dryers vary from 200 – 400$ (but huge transport fees) and we’re about to source the parts (no idea what to use for the heat source).
Our extruder managed to do upto 2 meters nicely. We have tried with PP and HDPE. or even a mix of the 2. We mostly worked a 32mm square tube to make our beams, We have even tried 50mm sq tubing upto 2 or 3 meter lengths. We could 2 meter lengths nicely but nice the core is the hottest and the extruded plastic flows through the core to last point, it eventually freezes at arounf 2.5meters makes liking sharp-rounded noses towards the end on the steel mould.
In our conclusion is that the 50mm sq plastic beam had so much deflection that is it not suiable for making long plastic beam items unless it is support along . We prefer making a few short moulds and make the production work quicker by screwing on one mould after the other while the other ones cool down to slide out.
That’s great work.
Have you weighed some of them to get an effective density or sectioned them to see what the void content is?
Yeah, HDPE or even PP is probably not a good enough material for small cross section beams to be useful unless they are supported along the way.
I wonder if you capped a fixed length mold and then applied axial compression during the cooldown, whether you would get a smoother surface and fewer voids. I do something similar with injected molds, it seems to help.
@crutchfieldengineering, yeah. good work ! how is it in the inside, could you upload a picture ? Also, what are the specs for your machine. I am always interested to see speeds, feeds, etc…
I was wondering there could be a way using a die and cooling & rollers, just as it’s done in industrial beam production. Those beams come out endless and have a decent surface finish. I would be happy seeing PP producing similar results.
Hello, @ppboys and @s2019.
Thank you for your kind works,
Both of you give some intering thought,
I have weighed some beams
i got 750-800grams per 1 meter of length for a square section of around 28mm.
So a specific density of around 0.95-1.05 which sounds rightfor HDPE and PP. my beams seem to be solid throughout, for the most part, their may be some air bubbles or voids to some negative pressure.
I have a attached a picture where I had cut down the middle of the entire length.
Capping a mould would be a good experiment. I think their would a huge pressure build up however, I have had a few cases where the mould actual buldges becasue the maerial stopped feeding but with feeding the plastic had to go somewhere.
Our extruder runs on single phase (230Vac 50 herts with a squiral gearbox to reduce the rpm from 1400 to 97rpm. Screw configuration e.g. pitch , compression ratios and shear angle are also very important. some screw designs wouldn’t work on the same motor, where by the shearing action the plasticwas to high and would actually stall the motor. Their is fine blanace between shearing action and feed.speed.
With our injection we feed under pressure compression too which give us a nice solid part.
@ppboys, of coarse the post-production process is possible we are busy working on a few ideas to get continious profiles by pulling the material through a chilled die or rollers, This is possible and is done commercially . You will just need to matchthe feed rate of the extruder.
@crutchfieldengineering, you’re welcome, and thanks a lot for the details 🙂 Same here, I just started investigating how to avoid air but also bad surface finishes. In the picture below I am using ‘SEBS’, just to have a way analyzing different speeds and material conditions (dried or not) and air bubble development over time – at different temperatures and speeds. In the next experiments I will try building up pressure with a spring loaded plunger, pushing the plastic toward the nozzle. And another one – but needs time more time – is to ensure that the barrel will feed plastic with an auger based hopper; no idea this could create improvements but apparently it seems to affect the overall consistency and possibly surface finish. Quite some stuff in the equation, see you in a bit 🙂
I am somehow tempted since long to put per heatband one PID controller and control the whole thing via bezier curves on a touchscreen, fun stuff 🙂
Yeah, a spring loaded cap/plug is what I was thinking. I think having pressure during the cooldown reduces the sectional shrinkage by compensating with length.
I ran an interesting experiment in my injection machine. I made a mold where two of the sides were 6 mm aluminum and two were wood laminate covered with aluminum tape. The results are shown below for two parts. The one on the left shows the side that was against the aluminum and the one on the right the side against the tape/wood. The aluminum side has the typical surface finish that shows the shrinkage patterns. The wood side is smooth. I think optimizing the thermal mass and conductivity of the wall material (basically the thermal diffusivity) can get you to a smooth finish.
You must be logged in to reply to this topic.