Building a wax extruder using Lost wax casting
I used my 3D printer and downloaded the extrusion screw that Siemen Cuypers designed and posted to Thingiverse. I then edited the file so it had a hollow center (better for metal casting) and a different base where I could machine in a keyed shaft into the base. I then printed it out at 5% infill in PLA with 2 layer walls. I made the screw 80 mm in diameter and almost 2 feet long to account for shrinkage and machining. Then glued the 3 parts of the screw together for the final size.
I coated the print in ceramic shell and via the lost wax method I poured and cast a bronze copy of the screw which is what you can see in the photos. For some reason I can only add 3 pictures in a post?
Anyway, the next step is to machine the sides of the screw so it will fit inside a 3 inch bored hydraulic cylinder (like the ones that power back hoe digger machine arms). Then I’ll connect it up to a 1/3 hp motor with a 20:1 gear reducer. I will have a variable speed controller on the machine and I have two PID temperature controllers to vary the temperature of the tube in two different regions. I bought band heaters to warm up the wax as it passes through the cylinder. I will cut different shapes into steel end caps. These will become dies for the wax to come out of. One thing I wonder if I will need is a breaker plate? Do you think a breaker plate is necessary or helpful in extruders? I see them on commercial units but I am not really sure what their purpose is or how necessary they really are at the hobby level.
Nothing new yet.
I’m out of the country until April 11th. Hopefully in the week or two after I get back I’ll be able to make some progress on it and post updates with pics.
The 1/3 motor is connected to a 20:1 gear reducer to knock the speed down from 1700 to 110 rpm. That should make the motor effectively 20x stronger because of the gear reduction. So it is as if I am using a 6 & 2/3 hp motor to turn the screw. Hopefully enough power?
So 0.3 mm on each side for a total of 0.6 mm if you include both gaps? That’s quite interesting. Larger than I had thought. Does the gap size change with the size of the screw or material being extruded? Or is it pretty constant across different setups?
True about welding pulling the barrel out of spec. The cylinder I’m starting with has thick walls, at least 1/4″ thick steel. Maybe a bit thicker. I’ll try keep welding to a minimum and use a mig to reduce heat input when welding the hopper on. I’ll be adding a thrust bearing at the back between the motor and the gear box to reduce backward pressure on the gears. The type of box I bought isn’t designed for massive amounts of back pressure.
Any idea about the necessity of a breaker plate? The commercial units all have them, but hobby DIY ones don’t seem to.
Do you think you could point me in the direction of the technical document threads? I’m finding it difficult to navigate this forum compared with other ones. It is set up in a strange way.
hi, impressive work. a few thoughts :
– you can’t use 1/3 Hp motor, even 800 W won’t be enough. it’s rather 3Kw in your case. You have to take into account the volume of the plastic but also the frictions caused by the screw along the barrel. apart from that, over time the extrusion screws build up stress in some areas, pushing even more against the barrel.
– leaving 0.3 mm play between screw and barrel was our best fit over time
– you may encounter some trouble turning this on size on the lathe. even with the high diameter the bronze screw will easily bend on your lathe. best option is a custom ‘follow rest’
– btw. I think there is a reason why traditional extrusions are done with chromoly based metals. such screws also receive various hardening and finishing treatments 🙂
– be very careful with making the barrel. welds will easily bend it and you are easily out of spec.
– there are few technical and scientifically documents about ‘extrusion screws’ here in the forum
keep us posted, btw. turning such a large screw on the lathe is around 12 hours and requires just a bit extra tooling (follow rest, limit switches)
More pics of the process
Here are up close pictures of the final bronze casting. There are some small pits and defects which I could tig weld and repair if I wanted to, but I don’t believe the holes extend very far from the surface and so I doubt if they will structurally weaken the screw. The walls of the casting are 1/2″ thick silicon bronze so I don’t think the defects will affect the function of the screw at all and will become plugged with wax over time.
Would a larger gap between the wall and screw allow air to get pushed backwards and come out the hopper? Or is a vent the only way to get air out?
No premelt. I am hoping the wax will get gummy inside the extruder and not liquid at all.
I have found (using a rather large steel horse medicine syringe) that molten wax is very difficult to squirt out consistently into strings unless it is a specific temperature when the wax is gummy and slightly thicker than honey consistency.
I wonder if I’ll need to cool the wax as it comes out the nozzle? Once the machine is all welded together and functional playing with temperature and fine tuning the feed rate with a variable controller is going to take a while to figure out the right conditions to extrude properly.
Just curious, do you plan to pre-melt the wax before it enters the screw?
Well isn’t that interesting! So the vent allows air out so it isn’t pushed into the extrudate?
I’m guessing it will probably need to be tiny for wax.
I notice that the vent is right above the midsection where the tines of the screw get wide again. That should make this area lower pressure and the wax less likely to be forced up the vent. Unfortunately I don’t have this on my screw. I wonder where the best location would be to place the vent?
Hmm interesting. I was wondering about backflow. I think I’ll try go for the closest tolerance I can get initially, test it out, and if it doesn’t work then I can always cut it down more.
I think I’ll do the tubing haha. The entire setup is going to be barely movable due to weight. The bronze screw is 30-40 lbs, the gear box and motor another 30-40 lbs, the cylinder is easily 70+ lbs and then all the framing I’ll need to make is going to weigh another 10 lbs or so. All in all we are talking close to 200 lbs! Not easily movable.
What is a vent? I have not read or heard about that? What does it do and where should I put it?
looks right to me, also have no idea how to get the breaker as insert like component but you’re getting there.
clearance, as it is wax, I can’t tell. i saw some materials even need a backflow, above 0.5 mm… i could imagine that the tolerances can be quite tight. there’s a table for sure for dealing with near liquids. hell you can even enjoy this certain blob sound, forever 🙂
you still may require a vent, for plastic its about a few micrometers.
you are sure about those tubing ? such a nice system deserves some solid plates 🙂 one underneath the hopper works well for me, at least 15cm x 8 mm and you should be fine
i am sure @andyn can help you better with that things
Yes wax not plastic. The extruder will be for making wax sprues & basic building material for my sculptures and castings. The wax should help lubricate the screw as well once heated. It is pretty slippery stuff.
So basically the tighter the distance between the wall and the screw the better? I’m pretty sure I can get 0.1 mm tolerances without too much difficulty. The cylinder I am starting with has been bored and honed inside so it should be almost perfectly cylindrical to start with (hydraulic cylinders need to be close to perfect to work!).
If the 1/3 HP motor doesn’t put out enough power I’ll upgrade to a larger one. The gear reducer I have takes motors up to 1.13 hp, so I have quite a lot of wiggle room if it comes to that.
I’ll have to have a think about the breaker plate then. I wonder what the best way of including it in my setup is. Maybe I’ll weld two end caps together, the first one can act as a breaker plate with many holes drilled into it and the outer end cap will be the one with the final shape.
I could tig weld it, but from what I’ve noticed tig welding seems to put more heat into the piece because it is a slower process. At least the way I tig weld it does! Ha ha ha. Definitely not a pro welder.
Here is the general plan I have for mounting all the parts and installing the thrust bearing.
here are some horsepower per diameter recommendations. 38 mm = 12 Hp but hey still depends on length and screw type (mixing, no compression, breaker mesh grid size)
well, all i can tell you about the motor is :
– once you have plastic in the barrel, it creates huge friction, no matter what.
– we ship extruders with standard 750 W and a 1:20 reducer, 3 Phase (inverter), it is just enough for less stressful applications
– it also depends on the quality you can archive, consider straighten out the barrel as well. our best tolerances we’ve managed to machine were about 0.1 mm
– the v4 team recently reported that 2 Kw is not enough for beams and so they added a 3Kw but couldn’t see much difference
– industrial screws with your size are used with 4 Kw + motors, i’ve visited a few factories and those screws operate on a pretty precise spec you just can’t do in the garage.
– you are not extruding high quality pellets! those flakes coming from the shredder are way harder to deal with.
– at the end you want twice the power you need 🙂
and yes, it’s 0.3 on each side, we tried 0.5 and also 0.4 but it turns more heavy to turn and create too much film along the barrel, obviously. I just saw some recommendations about (see docs soon) this but it didn’t really work out well. you can be luck to get the screw and the barrel with a 0.1mm tolerance. over time things seem to move so we went for 0.3 which was way more easy to deal with. going for less was even worst. i guess the specs/numbers for extrusions depend of course on the materials used.
about the documents, let me check, will post them here tomorrow.
about the breaker plate : it’s recommended yes, but also, depends on what you want to make though, it can work without too.
about welding: yeah, depends of course on your structure. we’ve opted for TIG, it’s way faster after all 🙂
see you in a bit! i am really eager to see this beast working 🙂
To melt out the PLA I had it in my burnout kiln which I had running with the propane on. I do not have a thermocouple on it yet but I believe it was around 1700 F. Ideal temperature for burning out is 2 hours at 900F but I haven’t set up the PID solenoid and thermocouple for the burners yet. I have the parts but just haven’t plugged it all in yet.
Why bronze? Because it is a good bearing material on steel so the lifespan will be long, it is easily poured, machined, repaired, has a similar expansion coefficient as steel so as it heats up it won’t bind in the tube, my foundry is a bronze foundry, and I had some extra bronze on hand. But it could have been made out of aluminum or cast iron as well. I just don’t usually cast those metals.
I was trying to figure out the tolerance it should be machined to. Basically what is the ideal gap between the screw and the wall? 1 thousandth? 2? 3?
wow very impressive work! Looking forward to see how it runs, why did you end up casting it in bronze?
Wow, thanks for posting, loved the video. Great use of 3D printing to build up a robust part. What temperature do you melt out the 3D print (PLA?)?
Here is the hydraulic cylinder with band heaters attached. The far right side of the cylinder has been cut and a 2″ black pipe welded on. The end caps are simple black pipe end caps from the hardware store that can be cut into different shapes and hopefully will act as dies.
And one more pic of the up close final bronze casting. There are some small pits and defects which I could tig weld and repair if I wanted to, but I don’t believe the holes extend very far from the surface and so I doubt if they will structurally weaken the screw. Basically I don’t think they will affect the function of the screw at all and will become plugged with wax over time.
Here is a video of the casting of the screw.
A few more pics of the process. I’m hoping to have it finished and running in the next 2 or 3 months.
You must be logged in to reply to this topic.