Last time we looked in on Injekto — a homemade plastic injection machine — it was at version 1.0. A recent video from the team that you can see below shows version 2 which is much improved and can work with 3D printed molds. Injection molding takes a lot of pressure and the machine certainly looks stout with lots of machined aluminum.

If you want to skip the build process, you can skip up to around the 9-minute mark. That’s where they show a machined mold and a 3D printed mold being used with the machine.

The first attempt overfilled the mold, but with a little clean-up, it looked pretty good, and, of course, there’s a way to adjust the fill amount. The 3D plastic molds cost about $100, cheap for a mold, but expensive for a 3D print. They also tested some cheaper printers and resin to create molds at a lower cost which also worked well.

The second version of the machine is a step up from the old version, as you’d expect. There are other ways to build an injector, of course, but this does look like a well-polished setup. We liked the dual injection buttons to keep your hands clear of the pinch points, for example.

Plastic injection molding is nice but aluminum injection molding is much better. Unlike plastic it doesn’t become trash after being used.

You are correct. not many people realise that unlike plastic 75% of the product ever mined is still in use today. So that beer can you are drinking out of would have been the part of the fuselage of an aircraft at one stage.

The correct term for aluminium injection molding is Die Casting.

The initial smelting from ore, melting for moulding and shredding for recycling of aluminium require orders of magnitude more energy than plastics creation, moulding or reuse.

In addition, die casting is a far more labour and tooling intensive (read:expensive) process that usually does not create ready to use parts like injection moulding does. Cast parts often require significant post processing.

Businesses exist to make money though efficiency. Injection moulding is king because it is efficient. No body has been sleeping on the benefits of aluminium, it’s just not as efficient.

Now that’s not to equate efficient with “better”, however you define that. If you want more use of the “better” process, you have to change the playing field through regulation to force out the incumbent, accenting that it’s not just an easy hand wave type process.

There’s a very specific reason plastic recycling is still a thing: it’s cheaper to both create and recycle plastic than it is to even recycle metal. It requires higher temperatures, more labor, etc.

Watch Climate Town’s video called “Plastic recycling is an actual scam”. It’s a great watch.

https://youtu.be/PJnJ8mK3Q3g

Not really Will, Plastic recycling is very much a thing, and is and can be done very effectively and very very cheaply in some circumstances.

What isn’t happening very effectively is the recycling of the vast volume of plastic created, which has often been ‘dumped’ in poorer nations under recycling contracts, but where nothing is actually done to recycle it. Not ideal, certainly some elements of scammers making money out of ‘recycling’, but then you get the same thing in any industry that isn’t sufficiently well regulated and has serious money to be made by not actually doing what you were paid for…

Maybe that can change if they could use FDM printers with recyclable materials, which doesn’t seem possible without polishing considering the surface quality.

This is about plastic injection molding for the hobbyist. If you want to drag this out about what is best, let’s just skip molding altogether, and go straight into forging or CNC matching, or a combination of both.

I didn’t understand why the Formlabs prints were much more expensive and what advantages did it bring ?

Moreover, resin printers don’t scale print duration with volume, so considering the max volume it is useful only for production runs above 10-15 (or if you don’t want to make resin parts), but we don’t know if the molds will resist…

Is that true, I can see they wouldn’t scale with layer area, but they would with height I thought? Print something 1cm high it takes time x, no matter area, up to machine limit, but print something 10cm high it would take 10x I would think.

On most UV cure resin machines that is exactly how it would happen – though there have been and I assume still are galvo driven laser spot cure machines where the time per layer is entirely about how long the squiggly line the laser is traced over needs to get.

Al Williams said: “If you want to skip the build process, you can skip up to around the 9-minute mark. That’s where they show a machined mold and a 3D printed mold being used with the machine.”

Thanks for adding that comment, it’s a time saver. This URL automatically cuts to the 09:00 time (9 minutes times 60 seconds per minute = 540 seconds):

https://youtu.be/M28Pbrm7wzE?t=540

https://youtu.be/M28Pbrm7wzE?t=540

Holly f*** this ground wire is sketchy! A loop end wire crimp, to anything in mains range. If that simple piece was omitted, I would operate this machine with 11 foot pole.

That’s why in developed country you can have RCD protecting the whole electrical system of your house.

It seems like the software to design the mold (channels, sprues, etc) is hard to get or find. And that’s for a 2 part mold, im not even considering a 4 part one. Possibly even harder than the machine.

Even if I built such a machine, using it effectively seems hard. As I understand it, without proper sprues and channels, you shatter your mold after a use or two (and have worse injections until then with air bubbles) Which is too bad, because I do have some fun ideas for injection molding or for casting.

This machine lacks ejectors and pull pins. Aluminum structure looks to make this a 0.5 ton max mold clamping force machine (how most IM are rated, e.g. a 150 is 150 metric tons of clamping force). No chance any of these machines ever get enough shots to make fatigue an issue, so he’s got that going for him.

Most importantly the injector lacks a screw feed and a slipring. There is no way they can ever get bubble free shots. The most common drill press based hacks have these, so I don’t know what he’s doing except just setting out without a survey of what’s already done. Does he even understand that you pack the injector with the tip blocked by the previous shot as it’s cooling?

Also _pneumatic_ cylinders? Really? Melted plastic is a _compressible_ liquid! Missing that fact produces much misunderstanding. Real injection molding machines approach 10kpsi injection pressures.

Also he is apparently unaware of some of the ways injection molding can go wrong. He is unprepared for a clogged nozzle and plastic decomposing to gas in the injection cylinder. That can release in a dangerous way out the nozzle. In the case of this machine it would likely back/shoot the injector piston out the top, fountaining phosgene, char and melt right into the operators face.

Mold temperature control? Never heard of it…Good luck with your thinner walls.

Am I the only one who looked at the first picture and wondered just what the builder knew about his machine that encouraged him to wear a full-face respirator and welding gloves ?

Concerns about hot plastic giving off fumes suddenly coming out of a mold at high speed?

Indeed, even if the fumes are too small in volume and so of no concern you want eye protection, and hot plastic is hot so gloves makes sense – its nearly always better to have overkill PPE than inadequate (you could even argue always if you exclude from inadequate PPE actively incorrect PPE – the stuff that is very much not for this task and for the job at hand actually increases the dangers).

“We also always use proper hand and face safety….”

Eh, really? I saw at least 4 instances in the video where the machine was being operated without gloves. :)

In a way this is really cool, but to me the cool party is that some people are doing some grassroots type engineering and being successful with it. I send my project to the hackaday tip line and they apparently rejected it.

When I consider how far fetched it is that anyone buy my kit to make my project, it’s a little comforting to know that these guys are apparently absolute to do stuff with a far more expensive esoteric and less likely to work unit…

Although I applaud what they have accomplished, I think guys that we all have to have self discipline in making sure we are working on stuff that’s really important. Let’s face it, the use case for such a machine is a tad limited. Anything you can make with this you can make through other means that are probably more practical unless you are producing quite a large number of units. Units of plastic stuff, not very accurate or whatever else. Of moldable geometry. I don’t see who could be making such a product. It seems like a market niche that doesn’t exist much, but if it works for you that’s great.

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.

By using our website and services, you expressly agree to the placement of our performance, functionality and advertising cookies. Learn more