I've worked as an engineer / cad tech / cnc programmer for several years and all the techniques used here are no laughing matter. You can instantly see this guy has worked in the industry for a minute and understands probing / tool holding to a very advanced degree.
Iron worker to fab shop to Mech engineer, watching a variety of these videos for clever inspiration to hopefully curtail chasing any dead ends in my design. I though the same thing. I've seen quite a few designs that were mcguyver ratcheted together but this one tickles my fancy😂 well done
Good design and machining Andy. Well done. FYI - back in the earliest days of Taiwan's meteoric rise to a manufacturing powerhouse their first commercial product was- buttons! I'm sure you'll get a lot of use out of that injection moulding machine!
Yes, they often have one in each corner, particularly on larger machines, though sometimes only 2, usually in the horizontal plane and I've even seen 2 in diagonally opposite corners. There are advantages to each, I used 2 for simplicity and I also like that there is better access to the mould area.
Looks good. Compressing the unmelted pellets, during injection, makes you lose some "umph". For optimum performance the piston should bottom out just before it reaches the torpedo. Then you should dose just enough pellets so that the piston almost bottoms out. Then you have a minimum of unmelted plastic to compress. Commercial piston machines used to have a system that used the stroke of the piston to dose the next shot. A shorter stroke resulted in less pellets dosed for the next shot and a longer stroke dosed more pellets. This made it self adjusting.
Yes, I'm still learning details like this as I go. Shortening the barrel definitely helped, the piston still stops some way short of the torpedo though. I read somewhere that one shot should be 20-80% of barrel capacity. I still have some spare travel on the ram, but the piston rod is too short. Perhaps what I should have done was to cut 75mm out of the barrel only, leave the machine frame the same height and extend the piston rod. I'm planning a system to meter both the travel of the piston and how much plastic is added, though I'm mostly going to be using flakes from my shredder rather than pellets, flakes are harder to work with especially when they are not a uniform size.
What wonderful imagination, design, engineering and construction abilities you have, and how great it is to see you combine them to create such amazing equipment usually only found in the industrial and manufacturing sectors.
Excellent video, very interesting how you tackle and resolve each issue. In hydraulics the ‘buffer’ tank would be called an accumulator that are, as you have positioned yours, close to the actuator to allow for pressure drops within the system. This project must have taken ages with the many iterations to get to a working solution.
Thats awesome. I'm a mold maker at an injection molding company. Your machine basically has everything a large industrial one would have. People don't realize how much of an art it is to get perfect parts even with cutting edge equipment. So many variables. We dry our plastic pellets before we shoot parts. You may want to try that. Maybe add an auger to your heater tube for more consistent feeding. Idk but you definitely did a good job.
Thanks! Yes, I know there are a lot of variables I haven't addressed yet. When I come to actually make the parts I want to make on this machine I will probably have to work backwards from there and troubleshoot each issue one by one.
Super job and good result. Of course I´m more into wood working, I like to learn more about metal work. Your video´s are showing what's possible. Thanks for sharing.
Whao, very smart and impressive work, i think, when you finish the automating it add some flasy UI, bring it on kickstarter, I will be a major backer...
nice project. i plan on building one at some point. i got a lathe recently and working on getting a bridgeport mill, and surface grinder soon. i use to machine injection molds and now i work in the plastics manufacturing side. I got a pretty good idea how it all works. its cool to see how others are miniaturizing these machines.
There's not much you can't build with a lathe, mill and a welder. I wish I had a surface grinder but sadly I don't have space and would probably only rarely use it, though for some jobs it's the best tool you can have.
Nice job. You might want to consider heating your mould’s. I can’t remember what kind of temp but that’s generally what happens commercially. Excellent work.
Yes, at 5:32 the bottom PID is actually heating the mould while the top one is heating the barrel. I didn't show it in the video but there is a cartridge heater in the side of the mould base. I've been experimenting with temperatures from ambient up to 80°C and it certainly helps
@@AndysMachines This is awesome, by the way. There is going to be a delay in ramp up on the temperature inside the mold from the outside where the thermocouple is, because of the thermal mass, and coupling between the inner and outer molds. *maybe* a tiny bit of heat conduction compound between those parts. You can possibly fine tune the heat output by expecting to toss the first few parts on any cold run. The hot plastic, if you cycle quickly, should help keep the inner part warm. (I have zero experience injecting plastic, but do have some experience casting bullets. Keeping a steady rythm is important.)
Where I work we typically hear our moulds to 100-140° F but also a mould temperature controller isn't just to heat it it also keeps it from over heating as well because after so many shots it will get too hot to control.
I've worked with the professional version of theses machines. good job very interesting. Perhaps you could get into building table top machines for people to enjoy.
It looked like your first problem could have been the mould temp too. It may have cooled the plastic so rapidly that it increased the pressure needed to push it all the way into the chamber. I think this is a great design though. I am working on a injection moulding design too and also decided on the piston system instead of a screw. I am trying to simplify it though and use the clamping mechanism to also push the piston so I only need one power source. I will be using a car scissor jack design made for an SUV because it gets 2.5 tons of lift and that should be plenty to close the clamp and push the piston.
This project is amazing I am very impressed! Why do you use a piston instead of an injection screw? It would eliminate the dead space issue. Melt the plastic more efficiently (you would need less heat. Because 80% of the heat to melt plastic comes from the friction. the thermocouples are used mostly to sustain the high temps.)
My previous machine used a screw which I made myself. It was a long and complex process and I never got the machine to work reliably. For this machine I am keeping everything simple and rugged. I know screws have many advantages and if I decide to build another machine in the future I might consider using one again.
Hard to control pressure without screw in barrel. No suck back either. Also your problem was coming from mould design(sprue bit to long for not heated one), but was resolved with forcing more plastic quicker.(you still had to use manufacturing grade PP resin) As home made IMM not bad, but it won’t be able to make any “normal” quality product. Good job, great explanation of basics in injection moulding process.
I can control the barrel pressure by adjusting the air pressure to the pneumatic cylinder (I have a regulator on the side of the machine). I've since made an additional smaller barrel which gives me greater control though a smaller shot size. Yes, the sprue in this video is quite long because the part is small and is centred in the mould frame, I had to install an extra ejector pin on the sprue itself. The black buttons were virgin PP pellets but all the others are recycled HDPE, the pellets were easier to add by hand to the machine (this was before I installed the hopper) but the HDPE actually works better.
Awesome series, enjoyed and learned a lot from your build. I'm an older welder/ mechanicals guy. The button mold was perfect and classic !!!! :) Possibly try a build of making your own recycled pellets machine with a vac / dryer. Looking forward to see your future builds. Subscribed !!!
Thanks! I'm attempting with this machine to be able to mould straight from shredded recycled plastic, without the pellet step in between, which is why the design of my machine is a bit different and also evolving as I build it.
Good job. i like your auto shut off nozzle and the material feeder tube. For mine one, the nozzle is constantly in contact with the mold. I manually rotate a lever that attached on nozzle to activate its internal mechanism to open and shut the nozzle valve hole (so i need to operate it when plunger move). I did one similar as yours to activate its the valve spring by vertical movement. But somehow I forgot why i gave it up, might be a serious leakage between the moving parts??!! For the feeder tube, i plan to add in my mine for a long time.
Yes, the shut-off nozzle valve is hard to get right, my first attempt didn't work well at all. A manual one is more reliable, but of course needs operating manually. If time between shots is short, you can get away with no valve and just a small nozzle aperture to limit how much leaks out.
As injection pressure increases, the molten plastic leaks between those sliding moving parts on the auto nozzle, of course also at the contact surface between nozzle and the mold. You might experience these when complicated mold and low flow plastic are working.
3D printers solve the "dead zone" problem in the barrel by constructing it in two parts: a hot end and a cold end, the hot end is where the heater block (in your case, the chamber with the bands on it) and the cold end is where plastic waiting to be melted is kept, well below Tg of the material, the two being connected by a polished tube that doesn't conduct heat very well, usually stainless, and the cold end is also actively cooled with a fan to ensure it never gets high enough for the material inside to soften and form a plug.
Yes, I did think about doing something like this, but I wouldn't be able to isolate the 'hot' and 'cold' ends from each other as well resulting in much more conduction and power consumption from the heater bands. Many manual vertical injection machines simply end the barrel directly above the heating zone and the piston rises completely out. There is a short cone on the top of the barrel where pellets are poured in, but this won't work well for my system of automatically feeding recycled flakes, this is why I'm trying to keep the barrel and feed system totally enclosed.
Dual pressure system, one that applies pressure into the system till it reaches n1-pascal then a second one that dumps full in behind it from a reservoir. Like slowly driving your truck forward to put tension on the tow-strap then gunning it to put mud on their windshield.
Not sure how I would do that with this machine but my previous machine (since abandoned) had a 2-stage injection system with a low pressure stage that melted and compressed the plastic and fed it into a high pressure stage for the actual injection.
Yes, you're right, I don't have a bandsaw (did have in the past). Probably quite a few other machines I don't have either, I'm running out of space. I do have a small power hacksaw, but I often prefer the old hand powered hacksaw.
You're so close to making a great machine... Where the difference is between a commercial machine is the plastic pellets are fed in via a screw in the main chamber. After injecting the screw starts, it pushes the ram back as it fills the chamber. What people don't realise is the screw mixes the pellets and also creates it's own heat in the churning process. The screw stops when the ram is at it's "full" position (usually a limit switch) Thus different size "loads" or "shots" can be programmed in. I've seem home units use a long 1/2 drill bit used (grind off some of the side flutes as it will cut away the main bore.) The drill can be joined to the ram with a coupling and have a toothed belt at the join to work the screw. That size chamber you should be able to make more than one button, more like 6 or 10. I hope this helps. ( I worked on and repaired plastic injection moulders for over 5 years)
Hi Andy, The first machine I tried to build (there's a previous video on it) was actually very similar to what you describe. I made my own 3-zone compression screw for it and it had a 2 stage injection with separate higher pressure injection cylinder. There were lots of problems with it though and I never got it to work reliably. My aim with the current machine is to go back to basics and build something more rugged and reliable, not necessarily as sophisticated.
@@AndysMachines Great machine though, I can't imagine a world without youtube to help. There's only so much a 3d printer can do, transparent lenses aren't one of them. Moulding indicator lenses is my goal so I might be trying to make my own at home as you've done. Good work.
Can we order this machine if you already perfected it,nice one man,..what you did is that you make it simplier machine compared to the existing ones that way more complicated yet same principle.
fantastic your machine! congratulations. does the control act repeatedly too, without having to press the start button every cycle? do you provide the source code you used on the Arduino for study?
In this video the control system is not finished yet. I am operating the machine manually. When it's finished it will run continuously by itself, in order to do this I need to add several sensors to detect eg. when the part falls from the mould, how much plastic is in the barrel etc. I have broken the control system across several microcontrollers that perform specific functions. I'm not using arduinos but the code should run on most arduinos (I'm using bare ATMega 328's programmed in assembly). For instance one chip controls the heating and cooling running 3 PID loops, another controls the clamp, another controls the user interface. It wouldn't be possible to run it all on one arduino due to the amount of I/O needed.
I'm beginning to realize the machine I want to build is going to be mammoth.. if it takes this much just to make a button.. What would it take to make a 6 inch mold with about 4 oz of plastic per shot x.x
This machine can make bigger things too, the absolute maximum shot size is probably around 2oz. The main thing that determines the size of the machine is the cross sectional area of the parts you want to make. A 6x6 inch part would require a clamp force of around 75 tons.
@@AndysMachines I went down this rabbit hole massively lastnight. I thought I was losing my mind until I realized I had accidentally doubled the weight of the part I want to make. The largest part is 1.8oz, but it has quite a bit of surface area. It is about 3x5 inch but thin walled. Long story short, My family used to make an item, but the dies were lost in a warehouse fire decades ago. After 2 years of pulling said item out of storage and selling online I have gained some momentum and am now finally running out of what was once a few dozen cases of said item, with people constantly mailing me asking for more. So I would need to get dies made up, and depending on the costs involved I may end up acquiring a bridgeport as well. Also I have a bunch of other large heavy dies for the same business that would make it more then beneficial (pay out in the long run) for me to acquire my own (presumably) 50 ton machine. I have a fairly decent understanding of fusion 360, and can always learn more. But my knowledge base mostly revolves around 3d printing, with a background in welding/fabrication/mechanical/hydraulics for 20 years and about 6 months working in a machine shop. Plus I have a smithy mill/lathe (not accurate enough for mold making) Anyway.. long story long this is fascinating =) and the only thing holding me back now is funding. (isn't that always the case)
The torpedo helps mix the plastic by narrowing the gap it has to flow through. The valve just stops the plastic 'drooling' when not pressed against the mould, it keeps the mess down but is not essential.
Well done! What happened to the flakes? I got the same challenge. I work with recycled hdpe and its flakes and at the moment I am still manually feeding them into the funnel, where I then have to stuff them into the cylinder. I nightmare but it works. Next week my setup is also completely automated apart from the flakes problem. I will propably use the screw from an old meat mincer to transport and push the flakes to the compression cylinder, but still have to do a proof of concept. How did you solve this problem?
I'm working on this at the moment and will cover it in the next video. I'm currently using a 25mm wood auger bit as a screw to feed the flakes, but it seems the flakes need to be within a certain size range. Too large and they clog in the hopper and don't get fed by the screw. Too many fine particles and they eventually jam in the screw and just rotate with it instead of being fed along. I think a bigger screw would work better, so your meat mincer screw might do better. Producing flakes of the correct size range is a pain, needing sifting and re-shredding. I'm almost thinking of building an extruder and pelletiser to first turn the flakes into pellets. The screw does work really well and consistently with pellets, which would allow me to precisely monitor how much plastic is added.
I've been wondering about this, I'm messing with the concept on paper with a view to building but can't afford a custom compression screw and was looking at masonry bits, specifically SDS bits. Have you tried this yet? Because if it worked with flakes then it should be possible to go flakes -> filament -> pellets all in one operation and use that for injecting.
This is awesome. I am actually trying to do something similar. One thing I am unsure of is how to make sure I am getting the air out of the melted plastic. Can you explain in a little bit more detail how your nozzle is designed?
I haven't really had problems with air in the plastic. At the pressure it is injected all the air seems to be forced out, even with a simple nozzle with just a hole (no torpedo or valve). Some plastics absorb moisture and need to be dried before use. Moisture can turn to steam when heated and this could be confused with air being trapped.
I experienced this. He explained well the moisture issue. But, also, for his design, it has a side feeder tube, the plunger won't fully comes out of the heater tube hole, so the hot air and melted material won't burst out there, instead it goes to feed tube, but there is a long length to travel, it won't cause problem.
Wouldn't you get WAY better thermal conduction if you'd have removed the oxide layer off the pipe and applied some kind of thermal transfer compound? Something like the thermal paste used between heatsinks and CPUs?
I'm not sure how much it would help in this situation. The parts are clamped tightly together and there's a relatively large surface area in contact. I think thermal grease is for when you need to push as many watts as possible through a small contact area or where the parts have gaps in between. I could try it I guess.
@@AndysMachines Per these two pages it looks like Iron Oxide -- 0.58 W/(m*K) -- Per: www.engineeringtoolbox.com/thermal-conductivity-d_429.html Iron (Cast) -- 52 W/(m*K) -- Per: www.engineeringtoolbox.com/thermal-conductivity-metals-d_858.html So the iron itself is roughly a factor of 100x more thermal conductive than the oxide surrounding the pipe. Even grinding it down (until you see raw iron), cleaning it with something like isopropyl alcohol and then smearing the thermal compound would get you much better heat transfer. If you really wanna get nuts you could machine them to a super close tolerance and with enough clamping force they might cold weld. Not sure though.
My gut feeling is that the oxide layer is so thin compared to the surface area that it wouldn't make much difference, maybe not even measurable, but I don't know. I might experiment with this next time I strip the machine down. I've read through the installation instructions from several manufacturers of band heaters and none of them mention using thermal compound, just that the surfaces should be clean and free from contaminants. The tube I'm using for the barrel has a 'bright' finish, not a thick black 'mill scale' oxide layer. I believe this type of tube sometimes has an electroplated corrosion resistant finish, though when heated it produces oxide colours on the surface that look just like steel.
@@AndysMachines Yeah the numbers above were for iron oxide, as I just guessed on the material. But any oxide is chemically different from the metal itself, and so I'd guess (huuuge uneducated guess) regardless of the metal, it would be worse at heat-transfer. We are talking about the pipe @ 3:51 which fit inside the custom machined sleeve you made right? An easy test would be to simply test the temperature of the pipe containing the pellets. If its not as hot as your thermocouple says (bc the thermocouple is mounted to the machined sleeve, not the pipe) then that's conclusive evidence you're not transferring heat and some kind of boundary-layer is stopping the transfer.
You're right, and that's actually a good thing as it limits how much heat travels up the barrel above the heating zone. I don't want that area to get too hot or plastic might start to melt in the feed tube from the hopper. The part of the barrel that's heated is encased in the aluminium sleeves (much better heat conductor) which are much thicker than the walls of the tube and have more thermal mass. The lower part of the barrel tube itself could be made from aluminium, which would give even better heat conduction, but aluminium is a relatively soft metal and the inside of the barrel would wear much faster. Using a PTFE piston instead of a bronze one might reduce wear, but if the gap between barrel and piston gets too big then the plastic starts to bypass the piston (molten plastic flows around the piston and ends up above it).
I don't think a valve would help unless the buffer tank were backfeeding the airline, which it can't in this case as there is nothing upstream taking air. A one way valve between the buffer and what it's supplying would only restrict the flow. The buffer tank needs to refill after each shot and has plenty of time to do this. The tank is really only a quick fix anyway and I aim to improve things so I can do without it.
I usually have several projects on the go at once and I don't think I'll ever run out of things to do! The terminator is going to take a while and will probably never be 'finished' I'll just keep developing it. I'm already thinking of better ways I could have done it and maybe even starting a 2nd one!
@@AndysMachines well then at least I got my answer. By all means keep up the good work and if you do end up working on that T-800 more all I can say is have fun and I mean that 👍
Sorry, just seen your reply and not sure if my answer above was very clear. When I said the terminator will never be 'finished', I meant because I'll still be working on it and yes there will be more videos, I'm working on the next one now.
There is a spring loaded valve at the end of the nozzle which lets the plastic flow when it is pressed against the mould and stops it when the piston and barrel are raised. This valve still leaks a little as it doesn't close immediately due to the viscosity of the molten plastic. I've also found that just making the hole in the nozzle smaller and not having a valve at all is fairly effective.
dumb question but would it be possible to close the mould with an air piston and do the "final" pressure with a shorter screw, making the cycle time alot fast and the ejection of the part a bit more snappy. :) Close (piston) -> apply pressure (halfturn/fullturn of screw) -> open piston and retract screw in one go.
Yes, you could do something like that. Whatever you use to apply the higher clamping pressure would have to 'move out of the way' of the air piston, or perhaps be on the other side of the mould with some sort of ratchet or locking pawl on the piston side. However reducing the cycle time is not always needed, the plastic needs time to melt in the barrel after each shot, for larger parts I had to introduce an extra delay for this to happen. Actually the motor driven screw I'm using could go a lot faster too, I'm driving it with something like a 25% duty PWM, but I don't actually need it to go faster.
@@AndysMachines Ah ok, that explains alot then.. Yeah i was looking at commercial offerings and their cycle time is "ka-thunk", but i guess they have some active cooling in their dies as well then.
Andy, awesome video. I’ve been wanting to do this myself. Did you create any plans based off your design or follow any plans? I’ve been looking around and have not come across anything as good as your setup.
This was so awesome, I genuinely felt I should comment just to say congrats! Also if you ever decide to make your videos a bit more detailed and/or narrated noone would complain I reckon. Anyways you're doing a great job. Keep it up!
Hi i'm from Colombia and i want to ask you because i'm creating a plastic proyect, can you sell your machine ? can you do a big machine ? I will wait your answer Nice creating
With the air set at maximum I can get around 200bar barrel pressure, but I usually don't need that much, 100-150bar seems enough. As these parts are small relative to the volume of the plastic in the barrel I didn't have to wait at all for it to melt, the cycle time is long enough that it can keep up when running continuously, though bigger parts would probably need more melt time. (Note that the machine is not finished yet and I'm operating it manually).
Andy, You have done some amazing engineering work. However you need to mix the polymer to achieve uniform heat in the barrel. You need to insulate the mould against heat loss. The injection phase needs to be speed controlled, with an additional control to manage the packing. Contact me via linkedin and I can help your project at no cost.
Hi Geoff, Thanks for the advice, but with this machine I'm trying to keep everything as simple as possible. In my previous machine I tried to implement some of the things you mention, but I could never get it to work properly. With a barrel this small (the smallest barrel I have for this machine is only 20mm ID) I don't really need to mix in the barrel, a torpedo at the end is enough, plus I like the marbled look from using different colours together. Yes, there is some heat loss from the mould, but very little and insulating it would be a lot of extra work. Though I do have a heater in the mould I find I can run it at very low temperatures and with the heat from each shot the heater barely ever kicks in. As for controlling the injection phase, I find that controlling the pressure and time is enough. I have a sensor that can detect when the injection is finished and the packing begins (added since this video).
On the aluminium block, a little way from the bandheater, though it is directly above it. The aluminium sleeve has quite a lot of thermal mass and is about 2x the diameter of the barrel (and of course is a very good heat conductor). Would you suggest a better location for the thermocouple? I will eventually have one on each block, and maybe another right at the nozzle. I was surprised though, I could go over 300°C without any noticeable burning of plastic (HDPE). I've had discolouration at over 260° on my old (horizontal) machine.
i dunno if it were me i might just wrap an induction heating coil around the steel pipe and temp sensor right at the nozzle - since that's the business end? seems like you're getting lots of loss between heater and plastic and also the sensor and the plastic. the plastic pellets are remote across thermal bridges from both heater and sensor
i think this is over enigineered for such a small mould, I have seen some machines so much smaller and faster than this one, have you ever tried using electric motors instead?
The machine can take much larger moulds, up to 10cm x 10cm (even bigger without the ejector rods). This small button mould was really just a test. I am using an electric motor for the clamp, but pneumatic for the injection. I have seen smaller machines use an electric driven screw for the injection which is a nice feature as it combines feeding the pellets and injection in one part, but for a machine this size I think pneumatic is better, giving more pressure and faster injection.
I have 2 barrels for this machine, one is 25mm ID and one is 20mm ID for when I need higher pressure but less volume. The stroke of the pneumatic piston is 300mm but I don't use all of that, the effective working length inside the barrel is around 100mm.
Felicitaciones ! muy buen material . Debería poner subtitulos en Español y Los videos con titulos en Español GANARIAS MUCHOS SUSCRIPTORES A TU CANAL , ya que no hay muchos videos con la calidad de tu enseñanza y conocimientos ! Tienes un nivel de conocimientos , metodo de trabajo , MUY AVANZADO ! FELICITACIONES MUY BUENA CALIDAD DE VIDEO ! GRACIAS
a beautiful creative video. the solution to India's unemployment lies in creativity in manufacturing . more and more of ideas would start pouring in . our engineering education needs more of the project work like that of Germany . most of the teaching staff in our engineering institutions have never been to a manufacturing facility . the engineering classes have become like a history class. simplified diagrams , old notes , imp topics ,....what else could they do . education , health sector, ....have become an industry , degrees and certificates are on sale ,....and now passed....w/o exam , assignments , practicals, .....in Corona ?
The anti-drool one? I just came up with that myself, there are plenty of other designs. I think it would actually be better to put the springs on the outside though, so the plastic doesn't interfere with the action. It just looked neater hidden inside.
The finish is really more dependant on the finish of the mould itself, to get a glossy finish it needs to be like a mirror. And of course everything else needs to be dialled in, temperature, pressure etc...
Moulds of this type often don't have vents (for the air). At the pressure the plastic is injected the air has no problem escaping between the mould halves and through the ejector pin holes, you would need a really airtight seal to stop it leaking out, but the plastic is too viscous to fit through the tiny gaps. The only place you might need a vent is at the end of a blind hole that gets filled by the plastic last, otherwise the high pressure trapped air can cause the plastic to burn.
The barrel is floating when the piston is right at the top and does move slightly to the right of the picture. When the plunger descends it aligns the barrel with the piston rod and also the mould. You can also see the effect from 8:30, it's probably because I haven't got everything lined up perfectly, though you're right it does seem to jump at 6:00, probably because I cut several shots together as I only have one camera to film all of the angles.
@@AndysMachines Oh and that's why there is so big of a gap between the piston and the barrel ? Wouldn't this results in plastic squishing on the sides of the piston and so an unpredictable flow of plastic ? Maybe reallying on the pressure sensor is enough to know if the mold is "full", I know nothing about injection molding :')
Not sure what you mean, I didn't really show the clearance between the piston and barrel, but it is quite small, there needs to be some clearance to allow for expansion and prevent binding, it doesn't need to be airtight. The gap is probably less than 100 microns, though the piston rod has more clearance, maybe that's what you are referring to? I haven't had any plastic bypass the piston, even after shortening the barrel. I plan to measure the travel of the piston to detect when the mould is full, this isn't really necessary for a successful injection, you can just apply pressure for a set time, but it will help identify any problems such as short shots or running out of plastic and the barrel becoming empty.
Really nice ! I'm currently building a pick and place machine to assemble my prototypes, and the injection molding machine will be the next one. Thanks for posting your videos, really inspiring ! Do you plan to share the plan and part list ? Really nice build :)
Thanks! Good luck with the pick & place. Yes, I'll probably draw up plans when I get it all working properly. So far the clamp is the only part I'm really happy with. The rest a work in progress.
I've worked as an engineer / cad tech / cnc programmer for several years and all the techniques used here are no laughing matter. You can instantly see this guy has worked in the industry for a minute and understands probing / tool holding to a very advanced degree.
Iron worker to fab shop to Mech engineer, watching a variety of these videos for clever inspiration to hopefully curtail chasing any dead ends in my design. I though the same thing. I've seen quite a few designs that were mcguyver ratcheted together but this one tickles my fancy😂 well done
Good design and machining Andy. Well done. FYI - back in the earliest days of Taiwan's meteoric rise to a manufacturing powerhouse their first commercial product was- buttons! I'm sure you'll get a lot of use out of that injection moulding machine!
Great work! Injection moulders usually have 4 guide rods,probaly only needed on a large machine! Very educational.
Yes, they often have one in each corner, particularly on larger machines, though sometimes only 2, usually in the horizontal plane and I've even seen 2 in diagonally opposite corners. There are advantages to each, I used 2 for simplicity and I also like that there is better access to the mould area.
This is fantastic! Have heard before that what you're here is "impossible" at home. You've proven that wrong.
Few things are impossible if you put your mind to it!
Well, some things will always be impossible😖
Is this for sale. Drop me a email if so Getdemgz@gmail.com
Looks good. Compressing the unmelted pellets, during injection, makes you lose some "umph".
For optimum performance the piston should bottom out just before it reaches the torpedo.
Then you should dose just enough pellets so that the piston almost bottoms out.
Then you have a minimum of unmelted plastic to compress.
Commercial piston machines used to have a system that used the stroke of the piston to dose the next shot.
A shorter stroke resulted in less pellets dosed for the next shot and a longer stroke dosed more pellets.
This made it self adjusting.
Yes, I'm still learning details like this as I go. Shortening the barrel definitely helped, the piston still stops some way short of the torpedo though. I read somewhere that one shot should be 20-80% of barrel capacity. I still have some spare travel on the ram, but the piston rod is too short. Perhaps what I should have done was to cut 75mm out of the barrel only, leave the machine frame the same height and extend the piston rod.
I'm planning a system to meter both the travel of the piston and how much plastic is added, though I'm mostly going to be using flakes from my shredder rather than pellets, flakes are harder to work with especially when they are not a uniform size.
What wonderful imagination, design, engineering and construction abilities you have, and how great it is to see you combine them to create such amazing equipment usually only found in the industrial and manufacturing sectors.
I agree,sir, being an Indian,I didn't see or observed such a typical art of making machinery,thank you.
Would like to know the cost of replicating this wonderful effort…
Excellent video, very interesting how you tackle and resolve each issue. In hydraulics the ‘buffer’ tank would be called an accumulator that are, as you have positioned yours, close to the actuator to allow for pressure drops within the system. This project must have taken ages with the many iterations to get to a working solution.
Thats awesome. I'm a mold maker at an injection molding company. Your machine basically has everything a large industrial one would have. People don't realize how much of an art it is to get perfect parts even with cutting edge equipment. So many variables. We dry our plastic pellets before we shoot parts. You may want to try that. Maybe add an auger to your heater tube for more consistent feeding. Idk but you definitely did a good job.
Thanks! Yes, I know there are a lot of variables I haven't addressed yet. When I come to actually make the parts I want to make on this machine I will probably have to work backwards from there and troubleshoot each issue one by one.
That's some 'Wow' engineering right there. 😁
I learnt so much from just one video... Awesome work! thanks for sharing. Keep it going!
Super job and good result. Of course I´m more into wood working, I like to learn more about metal work. Your video´s are showing what's possible. Thanks for sharing.
Whao, very smart and impressive work, i think, when you finish the automating it add some flasy UI, bring it on kickstarter, I will be a major backer...
Wow man, really impressive. A project this awesome deserves more views!
nice project. i plan on building one at some point. i got a lathe recently and working on getting a bridgeport mill, and surface grinder soon.
i use to machine injection molds and now i work in the plastics manufacturing side. I got a pretty good idea how it all works.
its cool to see how others are miniaturizing these machines.
There's not much you can't build with a lathe, mill and a welder. I wish I had a surface grinder but sadly I don't have space and would probably only rarely use it, though for some jobs it's the best tool you can have.
Nice job. You might want to consider heating your mould’s. I can’t remember what kind of temp but that’s generally what happens commercially. Excellent work.
Yes, at 5:32 the bottom PID is actually heating the mould while the top one is heating the barrel. I didn't show it in the video but there is a cartridge heater in the side of the mould base. I've been experimenting with temperatures from ambient up to 80°C and it certainly helps
AndysMachines ah great. PlastikCity dot co dot uk has some charts on the matter.
@@AndysMachines This is awesome, by the way. There is going to be a delay in ramp up on the temperature inside the mold from the outside where the thermocouple is, because of the thermal mass, and coupling between the inner and outer molds. *maybe* a tiny bit of heat conduction compound between those parts. You can possibly fine tune the heat output by expecting to toss the first few parts on any cold run. The hot plastic, if you cycle quickly, should help keep the inner part warm. (I have zero experience injecting plastic, but do have some experience casting bullets. Keeping a steady rythm is important.)
Where I work we typically hear our moulds to 100-140° F but also a mould temperature controller isn't just to heat it it also keeps it from over heating as well because after so many shots it will get too hot to control.
I've worked with the professional version of theses machines. good job very interesting. Perhaps you could get into building table top machines for people to enjoy.
Great job! Can you share the design or a video showing how you made the spring nozzle?
Amazing man ,I start this business not so long ago it is hard and a lot to learn. I got JSW 85 Ton unit
It looked like your first problem could have been the mould temp too. It may have cooled the plastic so rapidly that it increased the pressure needed to push it all the way into the chamber. I think this is a great design though. I am working on a injection moulding design too and also decided on the piston system instead of a screw. I am trying to simplify it though and use the clamping mechanism to also push the piston so I only need one power source. I will be using a car scissor jack design made for an SUV because it gets 2.5 tons of lift and that should be plenty to close the clamp and push the piston.
Yes, one of the 'bottlenecks' is the gate where the plastic enters the mould cavity. Heating this and making it larger definitely helps.
This project is amazing I am very impressed! Why do you use a piston instead of an injection screw? It would eliminate the dead space issue. Melt the plastic more efficiently (you would need less heat. Because 80% of the heat to melt plastic comes from the friction. the thermocouples are used mostly to sustain the high temps.)
My previous machine used a screw which I made myself. It was a long and complex process and I never got the machine to work reliably. For this machine I am keeping everything simple and rugged. I know screws have many advantages and if I decide to build another machine in the future I might consider using one again.
Hard to control pressure without screw in barrel. No suck back either.
Also your problem was coming from mould design(sprue bit to long for not heated one), but was resolved with forcing more plastic quicker.(you still had to use manufacturing grade PP resin)
As home made IMM not bad, but it won’t be able to make any “normal” quality product.
Good job, great explanation of basics in injection moulding process.
I can control the barrel pressure by adjusting the air pressure to the pneumatic cylinder (I have a regulator on the side of the machine). I've since made an additional smaller barrel which gives me greater control though a smaller shot size.
Yes, the sprue in this video is quite long because the part is small and is centred in the mould frame, I had to install an extra ejector pin on the sprue itself. The black buttons were virgin PP pellets but all the others are recycled HDPE, the pellets were easier to add by hand to the machine (this was before I installed the hopper) but the HDPE actually works better.
Nice! Now you can start pumping out terminators on your injection moulding machine 👍
wow , that great, good idea for the nozzle vale close and open , i note it , very usefull . you have been a good idea , a genious ,
Awesome job!!! Any idea if coating the pellets in a glow powder would work, that would really be neat. Can't wait to see the final project.
Now there's an idea! 😉 But wow that stuff is expensive!
Awesome series, enjoyed and learned a lot from your build. I'm an older welder/ mechanicals guy. The button mold was perfect and classic !!!! :) Possibly try a build of making your own recycled pellets machine with a vac / dryer. Looking forward to see your future builds. Subscribed !!!
Thanks! I'm attempting with this machine to be able to mould straight from shredded recycled plastic, without the pellet step in between, which is why the design of my machine is a bit different and also evolving as I build it.
wow a lot of work really nice job!
This would be an awesome way to recycle 3D prints
No idea why I came across this video but I’m glad I did
Good job. i like your auto shut off nozzle and the material feeder tube. For mine one, the nozzle is constantly in contact with the mold. I manually rotate a lever that attached on nozzle to activate its internal mechanism to open and shut the nozzle valve hole (so i need to operate it when plunger move). I did one similar as yours to activate its the valve spring by vertical movement. But somehow I forgot why i gave it up, might be a serious leakage between the moving parts??!! For the feeder tube, i plan to add in my mine for a long time.
Yes, the shut-off nozzle valve is hard to get right, my first attempt didn't work well at all. A manual one is more reliable, but of course needs operating manually. If time between shots is short, you can get away with no valve and just a small nozzle aperture to limit how much leaks out.
As injection pressure increases, the molten plastic leaks between those sliding moving parts on the auto nozzle, of course also at the contact surface between nozzle and the mold. You might experience these when complicated mold and low flow plastic are working.
3D printers solve the "dead zone" problem in the barrel by constructing it in two parts: a hot end and a cold end, the hot end is where the heater block (in your case, the chamber with the bands on it) and the cold end is where plastic waiting to be melted is kept, well below Tg of the material, the two being connected by a polished tube that doesn't conduct heat very well, usually stainless, and the cold end is also actively cooled with a fan to ensure it never gets high enough for the material inside to soften and form a plug.
Yes, I did think about doing something like this, but I wouldn't be able to isolate the 'hot' and 'cold' ends from each other as well resulting in much more conduction and power consumption from the heater bands. Many manual vertical injection machines simply end the barrel directly above the heating zone and the piston rises completely out. There is a short cone on the top of the barrel where pellets are poured in, but this won't work well for my system of automatically feeding recycled flakes, this is why I'm trying to keep the barrel and feed system totally enclosed.
Dual pressure system, one that applies pressure into the system till it reaches n1-pascal then a second one that dumps full in behind it from a reservoir. Like slowly driving your truck forward to put tension on the tow-strap then gunning it to put mud on their windshield.
Not sure how I would do that with this machine but my previous machine (since abandoned) had a 2-stage injection system with a low pressure stage that melted and compressed the plastic and fed it into a high pressure stage for the actual injection.
I feel like you have every machine but a bandsaw.. always with the hack saw
Yes, you're right, I don't have a bandsaw (did have in the past). Probably quite a few other machines I don't have either, I'm running out of space. I do have a small power hacksaw, but I often prefer the old hand powered hacksaw.
You're so close to making a great machine... Where the difference is between a commercial machine is the plastic pellets are fed in via a screw in the main chamber. After injecting the screw starts, it pushes the ram back as it fills the chamber. What people don't realise is the screw mixes the pellets and also creates it's own heat in the churning process. The screw stops when the ram is at it's "full" position (usually a limit switch) Thus different size "loads" or "shots" can be programmed in. I've seem home units use a long 1/2 drill bit used (grind off some of the side flutes as it will cut away the main bore.) The drill can be joined to the ram with a coupling and have a toothed belt at the join to work the screw. That size chamber you should be able to make more than one button, more like 6 or 10. I hope this helps. ( I worked on and repaired plastic injection moulders for over 5 years)
Hi Andy, The first machine I tried to build (there's a previous video on it) was actually very similar to what you describe. I made my own 3-zone compression screw for it and it had a 2 stage injection with separate higher pressure injection cylinder. There were lots of problems with it though and I never got it to work reliably. My aim with the current machine is to go back to basics and build something more rugged and reliable, not necessarily as sophisticated.
@@AndysMachines Great machine though, I can't imagine a world without youtube to help. There's only so much a 3d printer can do, transparent lenses aren't one of them. Moulding indicator lenses is my goal so I might be trying to make my own at home as you've done. Good work.
Can we order this machine if you already perfected it,nice one man,..what you did is that you make it simplier machine compared to the existing ones that way more complicated yet same principle.
great hard work
Excelente inyectora, si lo vende cual es el precio saludos desde Perú
fantastic your machine! congratulations.
does the control act repeatedly too, without having to press the start button every cycle?
do you provide the source code you used on the Arduino for study?
In this video the control system is not finished yet. I am operating the machine manually. When it's finished it will run continuously by itself, in order to do this I need to add several sensors to detect eg. when the part falls from the mould, how much plastic is in the barrel etc. I have broken the control system across several microcontrollers that perform specific functions. I'm not using arduinos but the code should run on most arduinos (I'm using bare ATMega 328's programmed in assembly). For instance one chip controls the heating and cooling running 3 PID loops, another controls the clamp, another controls the user interface. It wouldn't be possible to run it all on one arduino due to the amount of I/O needed.
Wow an expensive and labor intensive button maker
....and yes I know that other mold designs can be used
Coll machine and nice job though
I'm beginning to realize the machine I want to build is going to be mammoth.. if it takes this much just to make a button.. What would it take to make a 6 inch mold with about 4 oz of plastic per shot x.x
This machine can make bigger things too, the absolute maximum shot size is probably around 2oz. The main thing that determines the size of the machine is the cross sectional area of the parts you want to make. A 6x6 inch part would require a clamp force of around 75 tons.
@@AndysMachines I went down this rabbit hole massively lastnight. I thought I was losing my mind until I realized I had accidentally doubled the weight of the part I want to make. The largest part is 1.8oz, but it has quite a bit of surface area. It is about 3x5 inch but thin walled.
Long story short, My family used to make an item, but the dies were lost in a warehouse fire decades ago. After 2 years of pulling said item out of storage and selling online I have gained some momentum and am now finally running out of what was once a few dozen cases of said item, with people constantly mailing me asking for more.
So I would need to get dies made up, and depending on the costs involved I may end up acquiring a bridgeport as well.
Also I have a bunch of other large heavy dies for the same business that would make it more then beneficial (pay out in the long run) for me to acquire my own (presumably) 50 ton machine.
I have a fairly decent understanding of fusion 360, and can always learn more. But my knowledge base mostly revolves around 3d printing, with a background in welding/fabrication/mechanical/hydraulics for 20 years and about 6 months working in a machine shop. Plus I have a smithy mill/lathe (not accurate enough for mold making)
Anyway.. long story long this is fascinating =) and the only thing holding me back now is funding. (isn't that always the case)
nice work
Can you explain a bit about how final Torpedo and valve work?
The torpedo helps mix the plastic by narrowing the gap it has to flow through. The valve just stops the plastic 'drooling' when not pressed against the mould, it keeps the mess down but is not essential.
Amazing videos....a lot to learn from it..
Well done! What happened to the flakes? I got the same challenge. I work with recycled hdpe and its flakes and at the moment I am still manually feeding them into the funnel, where I then have to stuff them into the cylinder. I nightmare but it works. Next week my setup is also completely automated apart from the flakes problem. I will propably use the screw from an old meat mincer to transport and push the flakes to the compression cylinder, but still have to do a proof of concept. How did you solve this problem?
I'm working on this at the moment and will cover it in the next video. I'm currently using a 25mm wood auger bit as a screw to feed the flakes, but it seems the flakes need to be within a certain size range. Too large and they clog in the hopper and don't get fed by the screw. Too many fine particles and they eventually jam in the screw and just rotate with it instead of being fed along. I think a bigger screw would work better, so your meat mincer screw might do better. Producing flakes of the correct size range is a pain, needing sifting and re-shredding. I'm almost thinking of building an extruder and pelletiser to first turn the flakes into pellets. The screw does work really well and consistently with pellets, which would allow me to precisely monitor how much plastic is added.
I've been wondering about this, I'm messing with the concept on paper with a view to building but can't afford a custom compression screw and was looking at masonry bits, specifically SDS bits. Have you tried this yet? Because if it worked with flakes then it should be possible to go flakes -> filament -> pellets all in one operation and use that for injecting.
Awsome. Shall we also expect the Terminator series to resume soon as well :)
Yes, I'm still working hard on that. It's just that everything is taking much longer than normal at the moment, for obvious reasons.
@@AndysMachines Completely understandable. Stay safe
Now to start making penny skateboard decks!
Too big for this machine unfortunately😖
nice machine
This is awesome. I am actually trying to do something similar. One thing I am unsure of is how to make sure I am getting the air out of the melted plastic. Can you explain in a little bit more detail how your nozzle is designed?
I haven't really had problems with air in the plastic. At the pressure it is injected all the air seems to be forced out, even with a simple nozzle with just a hole (no torpedo or valve). Some plastics absorb moisture and need to be dried before use. Moisture can turn to steam when heated and this could be confused with air being trapped.
I experienced this. He explained well the moisture issue. But, also, for his design, it has a side feeder tube, the plunger won't fully comes out of the heater tube hole, so the hot air and melted material won't burst out there, instead it goes to feed tube, but there is a long length to travel, it won't cause problem.
Great Build would love to see more!
Its funny that the protolabs ad showing an automated injection mold machine turning out buttons brackets this video
Coincidence?
@@AndysMachines Spies are everywhere! LOL.
Would it help to heat the mold ahead of time?
Yes, depending on the complexity of the mould sometimes heating is required (I did include mould heating in this machine).
Very nice vedio thanks bro
good job!
Great job.
Excellent Work.
Well done what did the hole setup cost
damn bro I really appreciate your content!
Thanks, I really appreciate your appreciation!
Wouldn't you get WAY better thermal conduction if you'd have removed the oxide layer off the pipe and applied some kind of thermal transfer compound?
Something like the thermal paste used between heatsinks and CPUs?
I'm not sure how much it would help in this situation. The parts are clamped tightly together and there's a relatively large surface area in contact. I think thermal grease is for when you need to push as many watts as possible through a small contact area or where the parts have gaps in between. I could try it I guess.
@@AndysMachines Per these two pages it looks like
Iron Oxide -- 0.58 W/(m*K) -- Per: www.engineeringtoolbox.com/thermal-conductivity-d_429.html
Iron (Cast) -- 52 W/(m*K) -- Per: www.engineeringtoolbox.com/thermal-conductivity-metals-d_858.html
So the iron itself is roughly a factor of 100x more thermal conductive than the oxide surrounding the pipe.
Even grinding it down (until you see raw iron), cleaning it with something like isopropyl alcohol and then smearing the thermal compound would get you much better heat transfer.
If you really wanna get nuts you could machine them to a super close tolerance and with enough clamping force they might cold weld. Not sure though.
My gut feeling is that the oxide layer is so thin compared to the surface area that it wouldn't make much difference, maybe not even measurable, but I don't know. I might experiment with this next time I strip the machine down.
I've read through the installation instructions from several manufacturers of band heaters and none of them mention using thermal compound, just that the surfaces should be clean and free from contaminants.
The tube I'm using for the barrel has a 'bright' finish, not a thick black 'mill scale' oxide layer. I believe this type of tube sometimes has an electroplated corrosion resistant finish, though when heated it produces oxide colours on the surface that look just like steel.
@@AndysMachines Yeah the numbers above were for iron oxide, as I just guessed on the material. But any oxide is chemically different from the metal itself, and so I'd guess (huuuge uneducated guess) regardless of the metal, it would be worse at heat-transfer.
We are talking about the pipe @ 3:51 which fit inside the custom machined sleeve you made right?
An easy test would be to simply test the temperature of the pipe containing the pellets.
If its not as hot as your thermocouple says (bc the thermocouple is mounted to the machined sleeve, not the pipe) then that's conclusive evidence you're not transferring heat and some kind of boundary-layer is stopping the transfer.
great stuff! lovin' the build!
Steel has a bad thermal conductivity. Can't you use an aluminium tube instead ?
You're right, and that's actually a good thing as it limits how much heat travels up the barrel above the heating zone. I don't want that area to get too hot or plastic might start to melt in the feed tube from the hopper. The part of the barrel that's heated is encased in the aluminium sleeves (much better heat conductor) which are much thicker than the walls of the tube and have more thermal mass. The lower part of the barrel tube itself could be made from aluminium, which would give even better heat conduction, but aluminium is a relatively soft metal and the inside of the barrel would wear much faster. Using a PTFE piston instead of a bronze one might reduce wear, but if the gap between barrel and piston gets too big then the plastic starts to bypass the piston (molten plastic flows around the piston and ends up above it).
would adding a 1 way valve to the buffer tank help with the pressure drop returning to normal? so you didnt have to fill it back up again
I don't think a valve would help unless the buffer tank were backfeeding the airline, which it can't in this case as there is nothing upstream taking air. A one way valve between the buffer and what it's supplying would only restrict the flow. The buffer tank needs to refill after each shot and has plenty of time to do this. The tank is really only a quick fix anyway and I aim to improve things so I can do without it.
Just wondering what are you going to make next after the terminator project is done
I usually have several projects on the go at once and I don't think I'll ever run out of things to do! The terminator is going to take a while and will probably never be 'finished' I'll just keep developing it. I'm already thinking of better ways I could have done it and maybe even starting a 2nd one!
@@AndysMachines well then at least I got my answer. By all means keep up the good work and if you do end up working on that T-800 more all I can say is have fun and I mean that 👍
Sorry, just seen your reply and not sure if my answer above was very clear. When I said the terminator will never be 'finished', I meant because I'll still be working on it and yes there will be more videos, I'm working on the next one now.
@@AndysMachines oh no don't worry about it I know what you mean you don't have apologize
@@djbandicoot4798 hell yeah the terminator is how i found this channel and its awesome!
Is there anything that closes the path for the plastic once you stop injecting? If so can you please explain how it works?
There is a spring loaded valve at the end of the nozzle which lets the plastic flow when it is pressed against the mould and stops it when the piston and barrel are raised. This valve still leaks a little as it doesn't close immediately due to the viscosity of the molten plastic. I've also found that just making the hole in the nozzle smaller and not having a valve at all is fairly effective.
very nice video
Simply genious
dumb question but would it be possible to close the mould with an air piston and do the "final" pressure with a shorter screw, making the cycle time alot fast and the ejection of the part a bit more snappy. :)
Close (piston) -> apply pressure (halfturn/fullturn of screw) -> open piston and retract screw in one go.
Yes, you could do something like that. Whatever you use to apply the higher clamping pressure would have to 'move out of the way' of the air piston, or perhaps be on the other side of the mould with some sort of ratchet or locking pawl on the piston side. However reducing the cycle time is not always needed, the plastic needs time to melt in the barrel after each shot, for larger parts I had to introduce an extra delay for this to happen.
Actually the motor driven screw I'm using could go a lot faster too, I'm driving it with something like a 25% duty PWM, but I don't actually need it to go faster.
@@AndysMachines Ah ok, that explains alot then.. Yeah i was looking at commercial offerings and their cycle time is "ka-thunk", but i guess they have some active cooling in their dies as well then.
Andy, awesome video. I’ve been wanting to do this myself. Did you create any plans based off your design or follow any plans? I’ve been looking around and have not come across anything as good as your setup.
Thanks! Yes, I did draw some plans, mainly for the clamping section, it was my own design. I have the drawings on Patreon.
AndysMachines awesome! Thanks for the reply. Ill be heading there now.
Have you ever had any previous experiences with injection moulding? As working with it before or such?
Not in an industrial/commercial setting. Only as a hobbyist/maker.
So you need more experience in injection molding of complicated molds parts and some low flow materials as ABS and PC.
Subscribed. Glad I found you.
Wow i hope this will be a success so that we can have cheap buttons from now on.
Great build thanks for sharing
2:06 Nice machine swarf
Sir really excellent sir super
Can u share the details on the nozzle u made
I think I covered this in one of the other videos. I actually made several different types.
This was so awesome, I genuinely felt I should comment just to say congrats! Also if you ever decide to make your videos a bit more detailed and/or narrated noone would complain I reckon. Anyways you're doing a great job. Keep it up!
machine chai machine?
Hi i'm from Colombia and i want to ask you because i'm creating a plastic proyect, can you sell your machine ? can you do a big machine ? I will wait your answer Nice creating
How much pressure do you get at the piston?
How long did you wait to plastify the material before injection?
With the air set at maximum I can get around 200bar barrel pressure, but I usually don't need that much, 100-150bar seems enough. As these parts are small relative to the volume of the plastic in the barrel I didn't have to wait at all for it to melt, the cycle time is long enough that it can keep up when running continuously, though bigger parts would probably need more melt time. (Note that the machine is not finished yet and I'm operating it manually).
Tamiya still ain't worried 😜
No, probably not!
It's awesome mate, just adding a sprinkling of humor. 👍
😁
Andy, You have done some amazing engineering work. However you need to mix the polymer to achieve uniform heat in the barrel. You need to insulate the mould against heat loss.
The injection phase needs to be speed controlled, with an additional control to manage the packing.
Contact me via linkedin and I can help your project at no cost.
Hi Geoff, Thanks for the advice, but with this machine I'm trying to keep everything as simple as possible. In my previous machine I tried to implement some of the things you mention, but I could never get it to work properly.
With a barrel this small (the smallest barrel I have for this machine is only 20mm ID) I don't really need to mix in the barrel, a torpedo at the end is enough, plus I like the marbled look from using different colours together.
Yes, there is some heat loss from the mould, but very little and insulating it would be a lot of extra work. Though I do have a heater in the mould I find I can run it at very low temperatures and with the heat from each shot the heater barely ever kicks in.
As for controlling the injection phase, I find that controlling the pressure and time is enough. I have a sensor that can detect when the injection is finished and the packing begins (added since this video).
of course the plastic isnt' that hot - look where you put your temp sensor!
On the aluminium block, a little way from the bandheater, though it is directly above it. The aluminium sleeve has quite a lot of thermal mass and is about 2x the diameter of the barrel (and of course is a very good heat conductor). Would you suggest a better location for the thermocouple? I will eventually have one on each block, and maybe another right at the nozzle. I was surprised though, I could go over 300°C without any noticeable burning of plastic (HDPE). I've had discolouration at over 260° on my old (horizontal) machine.
i dunno if it were me i might just wrap an induction heating coil around the steel pipe and temp sensor right at the nozzle - since that's the business end? seems like you're getting lots of loss between heater and plastic and also the sensor and the plastic. the plastic pellets are remote across thermal bridges from both heater and sensor
Whaou Amazing!
i think this is over enigineered for such a small mould, I have seen some machines so much smaller and faster than this one, have you ever tried using electric motors instead?
The machine can take much larger moulds, up to 10cm x 10cm (even bigger without the ejector rods). This small button mould was really just a test. I am using an electric motor for the clamp, but pneumatic for the injection. I have seen smaller machines use an electric driven screw for the injection which is a nice feature as it combines feeding the pellets and injection in one part, but for a machine this size I think pneumatic is better, giving more pressure and faster injection.
what is the piston diameter and the bareel lenght inside , i want to make an injection machine , but i am stuck to this first question .
I have 2 barrels for this machine, one is 25mm ID and one is 20mm ID for when I need higher pressure but less volume. The stroke of the pneumatic piston is 300mm but I don't use all of that, the effective working length inside the barrel is around 100mm.
Felicitaciones ! muy buen material . Debería poner subtitulos en Español y Los videos con titulos en Español GANARIAS MUCHOS SUSCRIPTORES A TU CANAL , ya que no hay muchos videos con la calidad de tu enseñanza y conocimientos ! Tienes un nivel de conocimientos , metodo de trabajo , MUY AVANZADO ! FELICITACIONES MUY BUENA CALIDAD DE VIDEO ! GRACIAS
Awesome 👍
Nice!
a beautiful creative video. the solution to India's unemployment lies in creativity in manufacturing .
more and more of ideas would start pouring in . our engineering education needs more of the project work like that of Germany . most of the teaching staff in our engineering institutions have never been to a manufacturing facility . the engineering classes have become like a history class. simplified diagrams , old notes , imp topics ,....what else could they do .
education , health sector, ....have become an industry , degrees and certificates are on sale ,....and now
passed....w/o exam , assignments ,
practicals, .....in Corona ?
Just stumbled upon your vids where did you get the design of your nozzle?
The anti-drool one? I just came up with that myself, there are plenty of other designs. I think it would actually be better to put the springs on the outside though, so the plastic doesn't interfere with the action. It just looked neater hidden inside.
@@AndysMachines was my thought aswell looks great thanks for responding.
Amazing 👏
Hi can you try a glossy finish product
The finish is really more dependant on the finish of the mould itself, to get a glossy finish it needs to be like a mirror. And of course everything else needs to be dialled in, temperature, pressure etc...
@@AndysMachines I'm making one from POM (derlin) plastic. Building a mini injection molding now
@@AndysMachines thanks for the information
très interprétant merci
Cool!!
How can this inject anything when there's no exhaust?
The pressure builds up but it can't go anywhere?
This injektion is very slow, so the gap between the two molds should be enough.
Moulds of this type often don't have vents (for the air). At the pressure the plastic is injected the air has no problem escaping between the mould halves and through the ejector pin holes, you would need a really airtight seal to stop it leaking out, but the plastic is too viscous to fit through the tiny gaps. The only place you might need a vent is at the end of a blind hole that gets filled by the plastic last, otherwise the high pressure trapped air can cause the plastic to burn.
@@AndysMachines Oh! That is interesting, thank you!
Most cool
6:01 8:34 Seems like the injection nozzle isn't properly aligned with the mould
The barrel is floating when the piston is right at the top and does move slightly to the right of the picture. When the plunger descends it aligns the barrel with the piston rod and also the mould. You can also see the effect from 8:30, it's probably because I haven't got everything lined up perfectly, though you're right it does seem to jump at 6:00, probably because I cut several shots together as I only have one camera to film all of the angles.
@@AndysMachines Oh and that's why there is so big of a gap between the piston and the barrel ? Wouldn't this results in plastic squishing on the sides of the piston and so an unpredictable flow of plastic ? Maybe reallying on the pressure sensor is enough to know if the mold is "full", I know nothing about injection molding :')
Not sure what you mean, I didn't really show the clearance between the piston and barrel, but it is quite small, there needs to be some clearance to allow for expansion and prevent binding, it doesn't need to be airtight. The gap is probably less than 100 microns, though the piston rod has more clearance, maybe that's what you are referring to? I haven't had any plastic bypass the piston, even after shortening the barrel.
I plan to measure the travel of the piston to detect when the mould is full, this isn't really necessary for a successful injection, you can just apply pressure for a set time, but it will help identify any problems such as short shots or running out of plastic and the barrel becoming empty.
Really nice ! I'm currently building a pick and place machine to assemble my prototypes, and the injection molding machine will be the next one.
Thanks for posting your videos, really inspiring ! Do you plan to share the plan and part list ?
Really nice build :)
Thanks! Good luck with the pick & place. Yes, I'll probably draw up plans when I get it all working properly. So far the clamp is the only part I'm really happy with. The rest a work in progress.
Do ypu selling this machines?
So impressed !
can you make note book edge square machine model
Any updates?
Did you see part 3? Part 4 is on the way...