We've just launched our map. Add yourself by clicking here!


Laser and cnc parts for moulds

  • This topic has 14 replies, 9 voices, and was last updated 2 years ago by Anonymous.
Rūdolfs rudolfsskuja

Laser and cnc parts for moulds

09/11/2018 at 10:56

Dear Machine builders, engineers and product designers,
I’m proposing to make an archive of the experimental and working mould parts as files (that can be used to make the parts)
This problem has bothered our team, making good quality moulds.
I would like YOU to share your knowladge and experience making moulds.
Share your ideas and suggestions!
(I share my experience making compression machine moulds -coasters. If the moulds need to be good quality, then the materials sould be good quality too?)

@davehakkens @jegor-m

14 replies
2 subscribers
0 saved
sort on most likes
23/11/2018 at 13:48

Hi, actually we are looking for Mould-CNC Files!
Would be great if there would be a site (something like thingiverse, github, …).
Where are you in the developing of this archive? What would you need for developing this?

23/01/2019 at 20:17

Moulds are hard to come across by, indeed. Most moulds end up a huge investment, and you’ll only recuperate the investment if there is a market for whatever the mould makes.

23/01/2019 at 22:06

There are few other recent topics about alternative mold materials around (epoxy, wood, ceramic), please search the forum too.

On the other hand I wouldn’t be too much scared to get a CNC, you can build a quit decent one for 2-3K (another 1K over time for tooling). I’d say, if you made your way through plastic you will make your way through aluminium also 🙂

The limits of a low-budget CNC are mainly deep cuts ( > 10 cm) and surface finish which is yet not that a big deal because often you can’t even fill plastic that size with the PP machines and you can polish molds up your self and even if, there are workarounds like stacking slices. I can compile a playlist of decent DIY builds if you like… possibly there are also ready to use machines on the second market.

Looking at the math it’s also pretty simple! A relative simple mold is around 800 Euro, more likely not even tested against your machines nor do they have all the mechanics needed for a production (release plate, etc..), ending up sometimes in ordering another mold. Then you’ve just one project done. To make things viable, you will gonna repeat again. The sunglass maker (well, more workshop stuff than production) in Barcelona paid 2-3 K for their mold.

It’s not just about molds btw. Having a CNC at hand (tool wise) is an absolute must when things get serious. Or how else you wanna build quickly some semi-automatic injection system or whatever is needed (ie: https://www.youtube.com/watch?v=zTjbs3Hg2PA) ? I am using mine daily for all sort of things, marking hole patterns, finishing badly grinded edges, some new bearing mounts, heck, it’s became my right hand … mostly controlled by a stupid xbox-controller and some manual GCODE only.

as ignorant it sounds, please just get over it, do your self a favor and just get one 😉

most of the cases this will be enough : https://www.youtube.com/watch?v=K9UA9ZRFwWU


02/02/2019 at 05:27

I am curious if anyone tried to make experimental (not production) molds using soft materials and one of the cheap 3018 CNC engravers.

02/02/2019 at 14:54

not molds but I’ve cutted Delrin to make lead-screw nuts on a SBR20 (supported) railed mini 3d-printer conversion (3cm plywood) and I won’t do it again. possibly i used the wrong speeds and feeds but yet, it didn’t look good  or promising.

anyways, we’ve dedicated some budget to develop small [email protected] machine sets (500e max). if you think the router on amazon is a good choice, we could buy it just and make some tests for you 🙂

02/02/2019 at 19:04

I think I agree with you. That unit is probably too weak and too slow to be useful for machining molds. I was just curious if anyone had one and tried it.

02/02/2019 at 20:03

let me try the little SBR-20 router again (self made, 250 E or so) I’ve mentioned.

too bad I did run out of plastic and will have to order some pellets first. any specific material you have on mind to make molds? I always wanted to give epoxy a try but possibly overkill, no idea yet.

A SBR-15/20 rail system on something better than plywood could do. those standard printer leadscrews (larger ones) need just better nuts I think, ‘Delrin’ (used also for larger mills) is a great material for that.

02/02/2019 at 21:31

One easy to machine option may be to make the final part with machinable wax and then cast a plaster mold in a container with a rigid frame. For the type of home projects that may work well.
One advantage may be that you don’t have to generate a CAD file for the mold. This could make a lot of the thingiverse designs available though I have not worked with those yet.

A long time ago I made up some machinable wax and was surprised how good a surface finish you could achieve. It is also reusable.


05/02/2019 at 15:41

I’m currently working with these guys to see if they can scan a piece and design and 3d print a mold to be casted in aluminum. I’ll know ifits possible soon I hope.


01/03/2019 at 13:49


We are also developing laserfiles for moulds and would like to share our designs. Is there an open source community somewhere? it seems to be hard to find.


01/03/2019 at 14:02

@sheeponwheels not really. also in v4 there is no real versioned directory planned for this or machines. so for now I’d suggest to store it on github and keep updating here in the forum. github however allows others to update your files.

21/03/2019 at 13:07

Thanks for bringing this up. I’ve been looking for designs for molds and was having a hard time. Ideally, if we put them in other places like Thingiverse, GitHub, or other places, we should find a way to make them easy to find for everyone else. Most places offer some kind of tagging system. How about tagging them all with #preciousplastic to make sure we can find them easier later?

Here are some possible locations:
Thingiverse: https://www.thingiverse.com/tag:preciousplastic
YouMagine: https://www.youmagine.com/designs/tags/preciousplastic
GitHub (GH doesn’t call these identifiers tags, but topics): https://github.com/topics/preciousplastic

What do you think?

21/03/2019 at 14:01

Sounds good, i was thinking recently about making a bot for this. The google search API is comprehensive enough despite that human searches still give you better results. You could create custom searches on certain sites, score them and if there is enough material or backlinks you could add it to a machine generated catalog, takes a few days for a senior developer.

The major problem is that there are just a few here in the community who share enough details to reproduce molds (and thus success to some degree). Even when asking some people for more details like the sunglass guys in BCN, you’re ending more likely nowhere but being forced to start from scratch, which is a real pity.  In 50% of the cases I’ve encountered you can get the details, just takes a few days to that details.

The other big problem is that there are quite a lot of molds which need special machines and that’s very often not really visible. Folks on Instagram tend to show only the output but rarely any details about the process or machines involved. For experienced people it’s not a real problem.

We can just hope that the v4 team in Eindhoven rolls out enough infrastructure to streamline this very community efforts so that success can be duplicated in other regions of the world  🙂


21/03/2019 at 14:58

for anyone who want’s to get hands dirty on this 🙂

Using a Github (or just git) based approach seems the best fit:

– offline-availability
– decentralized
– no proprietary data, services dependencies
– future prove since it’s file based
– user versions

the only problems in the way is a pretty interface for non-technical users to add/modify data but this can be solved by many of those markdown/git/file based CMS systems which read the repository directly. You can also just make file – manager (i have done this) which uses git as backend (virtual file system), it auto-increases the version with each change… That way you only need to give users a link to upload their zip instead of an inflexible interface coming with lots of new problems. There are also thousands of already existing tools out there, localized 🙂


Viewing 14 replies - 1 through 14 (of 14 total)

You must be logged in to reply to this topic.

Support our projects on Patreon so we can keep developing 💪