Welcome to the SeeMeCNC Community

Hey everyone,

I've been working on nozzle designs for a while now. Nozzles are a critical part of filament type 3d printers. I think we should brainstorm a bunch of ideas for nozzles so we can improve them open source style. I attached some renders of nozzles below. The geometry of the nozzles can probably be optimized for future 3d printers.

This is a basic nozzle. The external taper angle is 50 degrees with a small flat on the bottom for the orifice to be drilled. This one uses a 118 drill bit for the main bore hole of 2mm. There are a limited number of drill bit angles we can choose from for CNC turning. The one below has a steeper angle between the 2mm bore and the orifice. Check out the tapered bore on this one. I think this could help improve the retraction characteristics with plastics, and help align fibers once we start using fiber reinforced plastics. Copper and aluminum have better thermal conductivity for higher melt rates. This will be necessary for larger orifice nozzles with faster speeds.
I'm probably going to get a few small CNC machines this summer, so I'll be manufacturing these for people who want to experiment. I'm limited to 5 pictures per post but I have a bunch of other nozzle designs ready to prototype. Some are integrated with a heater block, some experiment with exotic materials, some push the boundaries of thermodynamics in engineering, ect.

Post your ideas below!

Nophead did some tests with a tapered nozzle when he first tested stainless nozzles 2 or 3 years ago.
Short version I think he tapered the heated break, so the opposite taper to your diagram with good results.
The taper you show might make PLA clogging worse, the issue isn't on the retract, but during the resume when the rubbery plastic expands and grabs the walls.

I checked the inside of my nozzle about a week ago by retracting the filament out when the plastic was warm but solid. The plug of plastic in the nozzle had a smooth surface with a 45 degree tip.

Do you think that PLA is actually binding on smooth cylindrical walls, or is it a problem with surface finish/chamfers being caught? Any chamfer between the threaded parts (nozzle and heat break) could prevent retracts. A rough surface caused by bad drilling could increase the friction, but I don't know how much that would affect it.

I think if we make sure the parts that come in contact with high pressure hot plastic are well machined, parallel, and smooth, the jamming problems with go away. I wish I could start experimenting sooner. 6 weeks to go!


Also, does anyone have nozzles with orifices that have been enlarged by abrasion or wear? I think nozzle wear might be a problem once fiber filled plastics get on the market.

The theory goes that unlike most other printable plastics PLA has basically 3 states, liquid solid and a point where it's in a rubbery malleable state. This much is trivially demonstrable, the next bit is conjecture because it's virtually impossible to observe what actually happens. Though I think based on fairly solid thinking.

The thought is that as the filament is pushed down the barrel in it's rubbery state it expands, grabbing the surrounding material and jams. This seems supported by things like hotends jam even without retraction even in hotends with PTFE liners, as heat creeps up the Hotend and the size of the rubbery section expands. The PEEK fan reduces this soak, minimizing the length of rubbery filament.

It's also supported by what you will see if you strip down a really badly blocked Hotend.

There has been a lot of work on this over the years, all the original hotends were all metal, PTFE was added when people started printing PLA.

Also on the nozzle design, one of the key measurements is the length of the orifice, if it's too long it can be almost impossible to print with it, if it's too short nozzles tend to ooze, even if the melt zone is short. It also impacts die swell which can impact accuracy.