3
$\begingroup$

Assuming you have a suitable oven to maintain temperature at the filament's melting point and a suitable mold that can handle the temperature, is a commodity 3d printer hotend and extruder, with large nozzle, suitable for injecting material into the mold? I'm thinking of a setup like having the hotend mounted through a wall of the oven, braced against a hole in the mold inside the oven, and feeding filament via motor or manual cranking outside. Or is much higher pressure needed to make something like this work?

Certainly there are better setups to do this for manufacturing at scale, but the point of this question is whether you can do it with minimal setup effort and cost using commodity parts and filaments rather than needing expensive or custom-built equipment and material sourcing.

For relevance to the site in case it's questionable: certainly if this technique is possible, it could be used along with initial 3d printing of a design and using that to produce a (e.g. high-temperature epoxy) mold.

$\endgroup$
  • $\begingroup$ This is interesting, considering the recent spate of 3-in-1 devices that can do FDM printing, CNC milling, and laser etching in the same device, depending on the attached head. With the right set-up you could add injection to the list and create a 4-in-1 maker powerhouse. $\endgroup$ – Joel Coehoorn May 31 '19 at 16:02
  • $\begingroup$ @JoelCoehoorn: Sounds nice, but I'm not sure how you'd make that practical - it needs an insulated oven. $\endgroup$ – R.. GitHub STOP HELPING ICE May 31 '19 at 16:07
  • $\begingroup$ @R.. you can use a filaextruder as molding injection, just need the mold to be cooled according your volume rate. you don't need as much pressure, you won't make diamonds; you just need more pipes for injection to distribute the material. $\endgroup$ – Fernando Baltazar May 31 '19 at 17:25
5
$\begingroup$

Injection molding requires two major components: pressure and heat. So your question can be broken down into those two halves: can your average extruder handle injection molding temperatures, and can it handle injection molding pressures?

Let's start with pressure. Per this page on the University of Minnesota's site, plastic injection molding tends to require pressures of around 2 to 8 tons per square inch. Assuming you're using a 0.4mm nozzle, which has a cross-section of 0.126 mm^2, that works out to be 0.000195 (1.95E-4) square inches, which translates to about 3lb of pressure total at the nozzle assuming you're going for the high end of 8 tons (16,000lb). However because of the way that you're treating the molten filament in the extruder as a hydraulic fluid, you've got to deal with the fact that the "piston" on one end is actually quite a lot larger area, which means you have to multiply the force by that difference in size. The cross-section of 1.75mm filament is approx. 9.62 mm^2, or 0.149 in^2. That's 76.4 times larger, which means you need to be pushing on the end of that filament with roundabout 230 pounds, or 105kg, of force.

For reference, the Nema 17 that's on my extruder is spec'd at 76 oz-in of torque, geared down 4:1 through a Wade's extruder, and then acting on a hobbed gear with a 6mm effective diameter (3mm radius). Much to my own surprise, as I write this, that means that my little plastic extruder is actually capable of just north of 160lb of force! All these numbers would need to be recalculated for 3mm filament, and I have no experience with 3mm, so we're going to skip that one for now.

Now, that being said, my extruder is also capable of shredding filament if conditions aren't just right. The main two problems you'll have to overcome is 1) gripping the filament hard enough without destroying it, and 2) keeping the filament from buckling. I think if you got clever with some gears keeping multiple hobbed gears synced up, and a polished aluminum or steel feed tube, you could absolutely make your own extruder that's capable of consistently putting 300+ pounds of force on your plastic filament without it buckling or stripping. The downside is that your feed rates are going to be fairly slow, so each injection molding is likely going to take you quite a bit of time. A larger motor such as a beefy NEMA23 might help offset that by giving you much higher torque at higher speeds, so long as you can melt the filament fast enough. However we'll need to revisit these pressure numbers in a few moments, after I explain a few things about temperature.

Next, let's look at temperatures. Obviously we know that we can melt the filament itself as it's moving through the extruder. Using a Volcano nozzle or something, you can even guarantee molten filament at a fairly high extrusion rate. However most printers are designed such that the filament cools to solid (60-80 C normally) almost immediately. Injection molding designs require that the entire mass of plastic be kept molten. Fortunately, ABS and PLA melting temps are easily reached by literally any toaster oven, so stick your setup in there and you're golden, right?

But wait, there's more! One of the problems you'll run into immediately is that extruders are carefully designed so that the plastic is molten for as little time as possible, because molten plastic against a metal tube introduces a bunch of friction, hence the need for super high pressures during injection molding. If the plastic melts too soon, then you'll clog up your heatsink (the "cold" side of the extruder), and won't be able to extrude at all. This is a fairly common source of jams in 3d printing, where you're extruding too slowly and there's not enough cooling on the heatsink. Fortunately, E3D sells a water-cooled Titan extruder that would keep the heatsink cool. However the rest of your gearing assembly, and the motor, will also need active cooling, as heat damages the permanent magnets in the rotors, and the printed geared assembly obviously will melt if put inside an oven. Your best bet might be a water-cooled Bowden setup, assuming you can find tube fittings that can withstand several hundred pounds of force. You might look into using solid tubes like brake line rather than your normal PTFE shenanigans.

TL;DR: Get you a water-cooled extruder, make a super-strong bowden setup, and gear down a huge motor with a bunch of synchronized hobbed gears, and you might actually pull it off! There's plenty of Thingiverse extruder files you can use as a starting point.

As far as commercially available extruders go, however, I don't think you're going to find anything that's immediately available that can handle what you need it to without some level of modification depending on your selected injection pressures.

| improve this answer | |
$\endgroup$
0
$\begingroup$

An injection molding injector melts all the plastic needed for the shot and pushes it into the mold and through the sprue very quickly. Perhaps that is why it is called a "shot".

Injection molding machines do not heat the mold to plastic-melting temperatures. This works because the plastic is injected quickly, and fills the mold before the plastic cools. Molds are designed so that this happens, and often include multiple thick sprues to direct plastic to all parts of the hold.

Injecting with a 3D printer extruder will be a slow process. If the mold is not above the melting point, the plastic will cool and likely become a tangle of thread at the entrance of the mold. To combat that, you could heat the mold. This is doable, and will suitable insulation the temperature of the cold end and the extruder should be acceptable.

With the mold heated, you would inject plastic until the mold is full. The mold heater would then be disabled and the mold would cool. This would take a long time.

Injection molding machines typically have water-cooled molds to cool the plastic more quickly. Time is money for an injection molding factory, and cooling quickly is key to productivity.

For the 3D printer injection molding machine, the time when the plastic is hot could be fairly long -- longer than would typically be found in injection molding. I am concerned that some plastics, perhaps such as PLA, would degrade or burn during the long molten time. Experience would be required.

| improve this answer | |
$\endgroup$
0
$\begingroup$

Yes, technically you can but only for small parts.

However the size of the object would be limited (about the size of a button). It has to do with the power of the heater element. Its too small to deliver enough thermal energy to heat enough plastic fast enough to fill a large cavity (i.e. anything larger than a button in my humble opinion). The previous answers give a breakdown as to why. Normally in an injection molding machine, the plastic starts cooling when it hits the walls. As the first set of plastic hits the wall of the mold, it sticks and starts cooling. You have to get the rest of the plastic in before that area cools back down and solidifies. Practical for a small part, but not for a large part.

As for your idea about keeping the mold itself hot, yes that would work, if you could keep the temp within range. Overheating the plastic destroys the bonds, weakening the part. Too cold and it will clog.

But I say try it with a Volcano hot end and an actual mold. The plastic will melt faster if you use a preheater (a second hot end, that is upstream), to print the plastic up to 80% temp before it enters the final extruder.

| improve this answer | |
$\endgroup$
  • $\begingroup$ Why would it start cooling when it hits the wall of the mold? The mold is at the temperature of the oven. $\endgroup$ – R.. GitHub STOP HELPING ICE May 31 '19 at 15:29
  • $\begingroup$ Normally in an injection mold the plastic starts cooling when it hits the walls. $\endgroup$ – user77232 May 31 '19 at 17:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.