QUESTION: Where could this method be useful? Check out our Heat Set Inserts and Tools at cnckitchen.store (Free shipping worldwide starting at €100).

Hello from Montreal. So what I noticed is two things. 1- you need an air hole exit to allow the hot filament to make its way instead of possibly building pressure in the mold. 2- Why not lower the gantry to push down on the mold while sitting in the hot plate? Thanks for the video.

1. Air vent hole needed at the other side of hole for injection 2. You can inject plastic not at one end of detail, but at the middle - so plastic need just half path for fill all mold. Of couse you need air vent holes at the all end of detail.

<a href="#" class="seekto" data-time="650">10:50</a> injection molding operator here. You may need to optimize the mold first, like the placement of the sprue. currently, you have it at the very top of the part meaning it is quickly cooling down causing undershoots. move it more or less in the middle of the parts volume and you will be fine. I'm gonna be doing my tests soon enough as well so I'm gonna check it

I'm an injection molding designer. Here are some tips from my point of view: 1. The air vents like everyone said. But this may not be the only cause of failure. Since you have a parting line all around the part. The air can escape through there. 2. The mold needs clamping force so it doesn't flash, BUT the injector machine (in this case the nozzle) MUST HAVE injection pressure, and it's not a small one. We are talking about more than 2 MPa. This is because you need to inject quickly (around 1 to 3 seconds) and the solidifing plastic will give resistance to the filling. 3. In this case in particular, the mold should be heated around 80°C to 100°C temperature so the plastic stays liquid until it gets till the end. If you need some help, please contact me.

injection mold design engineer here. -First, heating up the mold before injection is crucial, as you noticed yourself. It will delay solidification of the part. -A vent at the end of the part might be a good idea, depends how tight your splitting surfaces are. -You should feed the part into the thickest part, like head of the screw in your last try. You always want the melt to flow from thick to thin since it tends to solidify in the thin part first. -Pressure is not the key actually, speed is more important since melt solidifies with time. Viscosity also tends to decrease with increased speed. Normal injection of a part takes max few seconds. -350bar of pressure is not the lower end in industry. Most of the parts I made molds for were injected with lower pressures. Machines go up to 2000 bar but it's rarely necessary. If part needs that kind of pressures then it's (mostly, not always ;) ) a badly designed part.

Keep in mind actual injection and ISBM machines charge the barrel before injection. That is to say, they melt enough plastic to fill the mold in advance, then inject it all at once.

The first application to this that immediately comes to my mind is reusing filament waste (failed prints, calibration prints, purge waste/ams poop, etc). Would love to see some experiments in how to melt and inject those into a mold, considering your past videos on reusing waste for filament extrusion

I hope you make a part 2. The comments are full of great wisdom and I can't wait to see how this develops

I really hope you revisit this with an air escape hole, the hotend assembly down and pressing onto the mold sitting on the printer bed (potentially heating it) so you don't need to hold it and also a bigger mold. I wonder how far this technique could be pushed!

You need a hole on the top on the mold to get trapped air out

Suggestions: slather some two part silicone or 500C gasket sealant around the edge of the entrance hole to create a squishy seal around it. That should increase your pressure. Additionally, the exit hole others have mentioned will help a lot. And yeah, pushing the preheat to the limit will help. Home injection molding machines typically require that you bake the mold in the oven before injection to keep the plastic from cooling part way through.

You can easily print shapes like that directly on the plate of the resin printer. No need for putting it at an angle with supports, it just slows the print down, uses more material, requires cleanup, etc. Even dimensional stability is better when printing functional parts directly on the plate, you just have to dial in the elephant foot compensation and properly set the exposure for the first couple of layers.... It was a revelation for me when I started doing it that way...

Always gate into the thickest part of the geometry, especially when as pressure and flow limited as you are with your setup. The flow channel that develops inside the part will rapidly narrow as the material cools, so you’re fighting an uphill battle.

Stefan, I believe you forgot a small vent hole towards the end of the mold (to let air escape) that and achieving higher temp in the mold

Couple of ideas, which come from my experience working for a molding company: Keep the mold hot at around 10 to 20 degrees (celsius) below the melting point of the material. This will give you plenty of time to inject. Make tiny channels at the corners and dead ends of the form to let air escape. This will reduce the needed pressure and it can serve as an indication when the material has filled the mold. You can make the mold outline smaller, reducing the needed resin. Inject from center outwards, not from one end. This will spread the material more uniformly. Best is to make the input port on the larger side of the mold - that way it will be as short and as close as possible, thus reducing the heat loss before filling the form. We are using thin motor oil as releasing agent. Sometimes petroleum jelly. Depends. These are cheaper.

You should consider venting your mold so you aren't building up pressure by compressing air. p.s. You don't need a draft for very short walls, as there is sufficient shrinkage to allow release.

SUPER!! It honestly surprises me that no one has done a 3D Printer mod kit for something like this yet. It seems like one could rather easily develop fixturing and use bed heater to preheat molds and then custom g-code to move the nozzle to each sprue hole and start shooting molds. Just a matter of time I suppose before we see it. Well done as always Stefan!

A local company has done simular as you. They take all the waste plastic from milk jugs, packaging etc, and then injection mold it into 3d PRINTED(FDM) molds of combs, hair clips, cloths line clips, etc and sell them at local farmers markets. Its really cool

I used to teach injection moulding theory at university and we would print a few moulds and run them with a small injection moulder, like this, as a demonstration. As most people have noted, position of the gate and airs are key. The only thing I would add is that with an injection moulding tool, you actually leave a tiny margin where between the two halves of the mould so that air can escape but plastic cannot. That's why you almost always see a tiny bit of a step along the seam where flashing eventually occurs, if the pressure becomes too great or the mould deteriorates. Hope this helps.

I feel like every expansive manufacturing techniques has started coming home, 4 axis CNC with the carvers, EDM with the power code and now injection molding

the fact that you can resin print a mold now is enough for me to get a small injection molding extruder tbh!

Wow! Your creative idea on 3D printer is amazing and endless. What a smart move to leave some room for everyone here giving their input and gain more engagement, which you tried not to put an air-release-hole to the mold as well as lower the print head to fix the mold in position with down pressure instead of hold by hands. 👍🏻

Wow, what a ton of responses and suggestions; I believe everyone would appreciate a follow-up video on the same subject. Thanks Stefan.

put the clamp ont he print bed, then lower the extruder into the mold and use the Z axis to hold it in place, and the nozzle and print bed can both pre-heat the mold from both sides :D

How about a nozzle that mechanically keys to the mould? And moulds that is designed to be clamped by screw hardware... That way you could put everything in a heated chamber and press until the hotend skips steps.

I love this. We need to continue the desktop injection mold technology. Stephon is doing it 💪🏾💪🏾💪🏾

I think you need a clamp with heating elements inside to keep the mold warm. That, and you really need airducts. It will flow way better that way.

Okay so, from what I’m seeing in the comments, he MIGHT have forgotten a vent hole. However, only a few people have said it so I’m not sure…

<a href="#" class="seekto" data-time="445">7:25</a> where is a channel to relase the air?

What an awesome idea! I have a suggestion: Lay the mold on the print bed, and lower the extruder down to the hole, clamping down the two halves with the extruder. Then crank the heat up on the bed, and since it will be touching the bed throughout the entire injection, that will keep the mold hot, and will probably allow the plastic to flow more easily, and longer before the plastic inside cools. It might allow you to do it with less nozzle pressure too.

This is creative engineering spirit is exactly why I love the 3d printing community, I love this! Please keep sharing these awesome projects with us :)

Why no air escape channels? The trapped air pushes plastic out of the mold.

Venting isn't required. You could add an extra chamber to be trimmed off later (often called a material saver, for some odd reason) that the gas could vent into. Remember, any gas inside the mold is being compressed. The seal between the halves will not be perfect enough to prevent gas escape at these low pressures. The suggestions to use the gantry to hold the hot end against the mold and to use the hot bed to maintain mold heat are very good ideas. Now I am wondering about standard 3DP resins......I don't think a glass filled 10k resin would be required for TPU.I've used resin printed molds for epoxy and composites, but hadn't really considered filling them with hot plastic without turning to specific tool resins. Good stuff!

Very cool idea. I to was 3D printing but wanted a way to make parts faster and stronger which was why I started developing DIY injection molding machines. I think one of issues that you will run into with using a 3D printer is how fast the plastic can be injected before cooling as you have seen. The other issue is that even if you are able to fully fill the part you would want to keep pressure on the part so the plastic has less shrinking and pitting as it cools. That also might be harder to do with an extruder over a plunger or screw type injection molding machine. Preheating the mold is important even on a regular injection molding machine but typically the heat from multiple injections is enough to keep it hot. I also can't remember if I saw vent holes for the air to escape but that is also very important. Last thing you could possible try is to switch the location of where the sprue is injecting the plastic in to the mold. If you did it from a more central location the plastic would have less distance to travel and thus potentially give you better results. Still super cool and can't wait to see what you are able to come up with!

Suggestion: Why not use a threaded, hollow bolt as your nozzle and then design complementary threads into one side of your mold. Screw one side of the screw-like nozzle into your hotend and screw the other end into the threads on your mold. This should create a tight seal between the hotend and the mold, preventing leakage. Then clamp your mold together and proceed. In addition to preventing leakage, by using a hollow, threaded bolt, you will also have more diameter to allow for maximum flow into the mol.

I'm being gaslit! I thought the topic of vents would eventually come up when it was the right time in the script, so I held back the urge to scream "VENTS!" at the screen for 18 minutes, but then the video just ended 😭

Tested this myself as well. SLA printed mold for a gear and that did work pretty well. I had a hole for the air to get out (and when plastic came out of it I know the mold was full), was using some grease for easier release and I had the mold on the heat bed and I just used the Z-axis to lower nozzle tightly to the mold. Ended up breaking the mold by moving the Z-axis the wrong way...

Wohooo, nice idea. Classic CNC kitchen stuff. Loved it

A always great Stephan! Thank you for sharing your experiments. You've got very nice tips from the comments, I hope seeing more work about this subject, it's very interesting.

you could try heating the mold in a toaster oven and too 150C, you can also change the place where you fill it and widen the sprue to allow more flow. Lastly adding air vents will help prevent air getting trapped. oh and you could make bolt holes to clamp it down they will keep a constant pressure.

I love this proof of concept; great work. I do think that the best part is the use of a resin printer to make molds. I don't know if you are aware but long before 3d printers were being made by hobbyists, some were making tabletop injection molding machines. Check out the Gingery book on the topic, its a much easier build than a 3D printer in every way except making the molds. So using a 3d printer to make the molds is the real game changer here.

I've had this idea for so long I'm so glad somebody did it 1st so I can see how it works

I had this idea where you use few generic hotends and extruders to melt filament and inject it into larger heated cylinder, and when it's ready, a piston pushes the plastic to inject it. Because even if you get hold of a hobby grade desktop moulder, you will struggle to buy the pellet material, and if you find it, it will probably be more expensive than filament or you will be required to buy entire euro palet at once.

i think the pressure is not the problem - your problem is speed and coolingrate in the mold. you have to melt all the needed volume and then shoot it in. a slow printer extruder might be ok for very small parts. maybe some air vents would help to reduce the airpressure in the mold, the plastic has to "fight" against. some small 0,5mm holes might help a lot.

Maybe change the inlet position so plastic doesn't have to travel too far? For example try printing the hook from the middle? Also adding a very small channel to release air pressure could help. This could be interesting although it also can release the pressure too well which may result in under-fills. I'm amazed how well this worked.

Worked a lot better than i expected, this gives hope that for smaller parts a much smaller and cheaper machine than the huge usual DIY ones could work

Great Video. Even in a professional injection molding shop if your machine is on the edge of filling a part, what we did is to place air vents in the areas filled last. Add a sacrificial hole on the end of the hook and you will be amazed how easily you will fill the part

Very interesting concept, did you try putting the fill hole on the long side so the filament has less distance to travel? It would also allow active heating on the build plate

In industrial molding pressure is everything. If the mold temp is below optimum and/or the material is not up to the desired temp for optimal flow rates, putting 75 to 100 tons of pressure behind the plastic WILL get the job done. Of course there is no way an extruder hot end can produce anywhere near that sort of pressure... BUT if you can get temperatures of the melt and the mold right it should work for small parts. Industrial machines have a screw that is moved back while the melt chamber/reservoir is filling and then gets forced forward quickly when injecting so maybe one of the hot ends that use raw/regrind material could be modified to work in a similar way. But I really like the idea of 3D printed molds - if only you (or maybe someone else) could perfect the process it would be a massive leap forward for makers.

I'm glad other people are mentioning air. I was sitting through every failure thinking about where the air was suppose to go.

For solutions to your issue: I believe you were on the right track by heating the mold prior to injection. As for why there are voids in the cavity leading to incomplete parts, I think what is happening is you are compressing the air left in the mold while simultaneously heating the air causing greater pressure than the rapidly cooling plastic can push. The reason why I think this comes from making soft plastic fishing lures. The molds used in the injection method as well as the open pour method have small grooves cut from the end of thin fine detail parts to the mold edge. These fine grooves allow the air out so the plastic can flow into the finer points of the mold. There are plenty KZfaq videos on home made soft plastics to reference. Some lead sinker molds also have thin grooves. I dare to make the comparison as soft plastic lures when at liquid temperature have a similar viscosity to maple syrup. I believe you could easily score your test hook mold with a dremmel cutting wheel from the up turned part to the edge. Less than a mm deep should be good. When or if a hair of the injection plastic comes out of the vent then you know the cavity has been full to the bottom. As for the sucking in of the top after cooling that is from the shrinking of the cooling plastic. It happens in soft plastic injection molding as well. Often a bait maker will top off the sprew so there will be less chance of cavities in the top. As for making molds, I have personally made a 5 cavity minnow open pour mold from elegoo water washable resin to see if it would take the heat of soft plastic and lead. The mold has been used several times for plastic and once for lead so far. It has held up well to the temperature I had plans to make a 2 piece mold like yours this past winter but found myself without the time or energy to 3d model the mold after a long days work. Hopefully I will get the chance later this summer.

You should have homed the printer then use the clamping force between the nozzle and the bed to get a better seal on the nozzle.

Can you add a port at the other end of the part to let the melted filament all the way through then finish with a cut in post like casting?

Your silicone sock and this episode are most outstanding!!! You did and share a lot of great experimentations! Vielen Dank!

Molds usually have breather holes to let the air escape. Also, you should inject from the middle, so the plastic doesn't have to travel as far to reach all the corners

I have a question, why not shape the hole after the nozzle and use the z-screw to hold it in place?

Hi Stefan, es ist super wichtig das die Luft aus der Mold entweichen kann, oder du eine kleine Kammer einbaust wo das Material überlaufen kann. Ich habe viele solcher Molds im 3D Druck für meine Arbeit gebaut. Falls du interresse hast können wir uns hier gerne austauchen.

i have a background in injection molding some tips i have found in industril cases 1 you need a holle for air excape this allows for better filling 2 the problem you have for the undershot is from the masive gate this like the support on resin is easier to remove the sprew 3 a mounting plate on the back side of the hex for the hot end will allow for heating the mold and betting injecor presures 4 when injecting you want the material almost to the point for burning this allows for faster injection and better fill

Really cool. Two things of note. (1) You need a second hole to allow air/pressure to escape from. When doing stuff like metal casting using a mold without that second hole you get only partially filled molds. All the 1 hole system is doing is causing air pressure to build up toward the bottom of the mold and that prevents it from filling. Without the hole all their air either gets trapped and won't allow the plastic to fully fill it or it tries to escape, either through the top hole where the plastic enters or by forcing the mold apart. (2) You need to find some way to continuously apply heat to the mold. No matter how hot you have the plastic it means nothing if the mold rapidly cools it and blocks the passage.

Wow, I found this on my recommended 8 seconds after it was posted.

My brother in christ you need a hole for the air to exit the mold LMAO

I have had this idea before I’m actually excited to see your findings

Really love these videos, high quality shots, good testing, with theory behind it and not overly serious. Keep it up man!

Wow, I'm looing for a home injection molding solution and this idea is blow my mind

Excited to see a part 2 with all the suggestions from the comments but it seems like a great start!

To achieve the necessary pressure, you could shorten an overly long nozzle, create a thin section immediately after the hotend, and then screw on the mold. Incorporating air holes and other suggestions from the comments, this approach could work very very well.

Once upon a time before 3D printing machines were cheap I tried injection molding with hot glue and I actually got it to work after a fashion using milled aluminum molds. The thing that really helped was preheating the aluminum mold. In the end I got a small Creality machine when they became available and now I make all my parts that way.

If you get around to medal molds then you could probably also make a custom nozzle that can link to the mold so you can latch it on to help with that issue of leaking, and this would allow you to have it heat up with the nozzle so you could make bigger molds without it hardening too quickly to fill it. And don't forget vent holes because if it is made airtight, then you are going to need them if you want to fill the molds up all the way.

Agreed, I work with molds at my job and all of them need venting some times in multiple places inorder for the mold to fill completely.

This is an amazing way to use a 3D printer. I would love to try this, if I had the correct setup. My 3D printer is way too old. I can't wait for what you and the community come up with to advance this.

I wrote prusa about this exactly about six months ago. Never heard back. Glad to see you tried it out!

I did this months ago, but I used an aluminum mold and secondary heater blocks to maintain temperature to keep plastic molten. It worked okay, but ultimately the limitation was maintaining heat. It's just a continuation of the hot-end, but without the high powered heaters and pressure used in injection molding, you're limited on part size.

It's already been said but injection molds have micro air vents. It would allow the plastic to flow better

A way that may help alleviate the pressure issues is adding little branches coming off the part, see how the PLA wasn't getting to the end of the mould, extend where the end is, so at the end of the hook have a small 0.3mm channel that will help the end fill up

Love how much feedback this is getting, if this improves so much in a short time the possibilities are endless

I'd be interested to see more!

I love how at the very end you get actual nice models injected

I have watched a LOT of DIY molding videos with metal and they always have Vent holes to let the air escape so it can fill easier and faster before the plastic hardens

Wow! Creativity never ends, i interesting on test all filaments in molding version it realy shows some interesting results

Really surprised as an engineer you didnt consider having a vent. Would thought that would be common sense. Without a vent you're just compressing the air inside.

I love this idea!!! Please do more!!!

For the M4 screw mold where the head never completely fills, try a couple of well-placed weep holes through which air can escape as the head portion gets filled. The weep holes should be tight, perhaps 1/3 mm, because you want the air to escape the head allowing the head to fill with plastic, and you do not want a lot of plastic escaping through the weep holes. The goal is to fill the head portion and quickly plug the weep holes so that you can build pressure as quickly as possible. You may even experiment with weep holes that don't fully reach the edge and allow air to escape. Maybe the weep holes should have just enough empty volume to contain the air that isn't escaping your current M4 screw mold.

I’ve been thinking about this for over a year after a horrible failed print where plastic was injected everywhere. I think I suggested this to you a while back

Bravo Stephan!! Your channel is pushing the boundaries of cretivity!! Please don't ever turn into a 3D Printing reseller like the rest.. You inspire us!!

Good to see you have a Q1. I hope a review will come.

I have tried a similar method not to inject mold, but to strengthen vertical on my prints. If you leave a channel in your print where you can extrude molten material from top to bottom, it does have much more strength than the regular inter-layer adhesion. I raise the temperature when doing that by about 20 degrees above the regular printing temperature, so the plastic could flow a little more and connect to the existing structure. Also, for your method, try heating your mold above the melting temperature of the plastic.

2 things. At the end of your hook, create a tiny air channel so the air can escape as you push plastics in. Otherwise I think the fluid plastic needs to fight the air pressure if the mold is to airtight. Secondly, put the mold against the build plate and turn its temp up real high, this will ensure longer fluent filament. Then lower the nozzle till it locks the nozzle against the molds entry point. This way you don't need to rely on your muscle strength for the pressure. It also allows for the pressure to go straight down and not diagonally. As added bonus, you don't risk hurting or burning yourself. As a bonus suggestion, calculate the volume of the object to mold and just push that amount of filament in.

I just wanna say I love the community, the comments are just full of various experts offering really neat advice and ideas and I just find the level of cooperation and helpfulness really cool. Humans are pretty alright sometimes :)

looking forward to the next stage

by modifying the mold to change the entry point position and possibly making a y to push plastic at several entry points we could facilitate the penetration of the plastic into it, we can also create exhaust route to release the air present and make place for the plastic

You need to add vent grooves in the mold. This drastically helps fill the cavity when doing low pressure desktop injection molding. If done right (size and location of the vent as well as injection pressure) the vents won’t yield much flashing if any at all. I’ve successfully printed SLA molds with .001” deep x 6mm wide vent slots that fixed the short shot issues I was having.

I always love seeing when you drop a new video because I know it will change my perspective on 3D printing. No, I had NEVER asked myself if I could injection mold with a 3D printer but now I am thinking about it, thanks to you! Some thoughts for your mold that may help. Because of the relatively low pressures that the injector can reach it should drive slightly different mold designs than would be expected for "normal" injection molds. 1) Air vents. The very bottom of your hook was the hardest to fill, perhaps putting a decent sized air vent at the tip of that hook would reduce the pneumatic pressure acting against your injected plastic. Also, it could show you that your part was fully filled if material squeezed out of it. Adding multiple additional air vents at key point on your part could reduce the needed injection pressure 2) Multiple injection sites. If done quickly enough you may be able to inject into more than one point on a given mold with the same tool head, or even multiple tool heads. Especially with multiple tool heads I think an air vent would be important. Multiple tool heads at different angles would also address your volumetric flow rate limit. 3) Customized Injection configuration. This could be multiple tool heads, a heated cradle and all sorts of other crazy ideas. I originally though you could press the mold against the printer bed but after reflection I realized it really isn't designed to apply that kind of force... but you really don't need to put the tool head on a 3D printer. The simplest "custom injector" that I would try is actually injecting from below with open venting at the top. This could significantly reduce needed pressures while still using gravity to help fill the whole mold. My hypothesis is that if you tried this technique and injected from the top you'd be likely to not get the plastic filling the mold as well, some "backpressure" or force is needed to help fill the entire mold. I look forward to seeing what you do next!

before even watching the video; I have been thinking about making a "micro" injection molding machine from my old 3d-printer for a while now, so looking forward to see how well it works out :D

preheating was definitely needed so I am glad you did that you may also want an air path. it means a burr to deal with but it should lower the pressure needed. assuming not to small of details for the desired mold. the air is part of the reason why injection molding uses a lot of pressure. as well as to make sure that every nook and cranny is filled with the material. so I think if you have a small air path it would help lower the pressure needed and still be easy to clean. and if you need less pressure it should help a lot when holding it by hand

Love the concept, i appreciate you for shearing your ideas in a way that others can understand and improve on!! Even when you don’t have it perfect your self it still helps the community!

I'm an injection moulding machine operator and for a while now have been trying to emulate the quality of injection with 3d printing, but I remember seeing this topic before and have often thought about it. It's easier to take an injection moulding machine and turn it into a 3d printer than to make a 3d printer perform injection moulding. I do have some suggestions, however. You may want to design your mould with small vents for the compressed air to escape, also preheating the mould is key - you're on the right track there. Allowing the molecules time to relax will help reduce internal stress and warping (i.e. slow, adequate cooling time). I think that mould release helps. However, I think your main bottleneck is injection time. Normal injection moulding occurs in a fraction of a second (i.e. 0.10 sec on the faster end of the spectrum), filling the mould in 5 to 10 seconds allows the plastic to cool and freeze before it has time to completely fill the mould. The leading edge of the injected plastic freezes first. Keep in mind that temperature increases with flow rate and pressure. Maybe add a gate to maintain a high flow rate. I hope some of this insight helps, I would love to start injection molding with my FDM printers. 😄

You know it's gonna be a good day when there's a new CNC Kitchen video \0/ I know this is random but I'd love a video on resin printer's usability for mechanical and accurate part. the tolerances, shrink, predictability etc.

Hello from Canada, amazing video as always. Thanks for the content.

This is one of the coolest things you've done.