Skip to content

Comment on Researchers demonstrate rapid 3D printing with liquid metal

Comments

This is exciting but incredibly niche atm. The parts it produces have to be milled and machined afterwards which in some cases could increase the manufacturing time overall. I do like their approach of keeping it hot and printing inside the sand. Much like casting. The nozzle however pushes sand all over the place and is probably pushing it into the path, causing the poorer quality prints. I spent a couple years as a young adult in a metal part manufacturing plant while going to school.

It’s exciting though because I believe we’ll have a time, not too far from now, where a design can be fed to a molten printer to print a base part, transferred to cnc mill for final refining, and then polished and painted for shipping. It won’t cost $10M to create a fabrication facility. It won’t require 25 years of industrial mechanics to build your “plant”. Just get “Bubba the Boiler” metal 3D printer, “Michael the mill” CNC miller, and the cast and crew of smaller tools to buff, polish, powder coat.

I’ll stick to wood and plastic for now. More my “level”.

I love that I can always count on the internet to tell me why cool new things are dumb and not useful.

It doesn't cost $10M in equipment costs to create a basic fabrication facility for short-run metal parts today, right? You pay for land, permitting & compliance, construction, etc. Machinery to cut "consumer-sized" parts out of metal stock, or to stamp sheet metal, usually costs six figures or so.

Lost wax casting, which is done with CNC-machined molds, isn't a capital-intensive process either.

An old but servicable lathe and mill can be had cheap on craigslist anytime. $5000 with some tooling. $10000 if you want cnc and digital readouts, with plenty leftover for more tooling. new imports can be had for similar prices - there is debate about which is better (both are compromises somewhere, new of what the craigslist stuff was new will cost you $100,000 or more).

or you can send the print to any machinist to manufacture for you. There are many around, but often they won't talk to you for small runs. (but someone will)

This is reminding me of a YouTube channel I watch, Cutting Edge Engineering in Australia. It's a shop that does machining and welding in support of things like bulldozers and other construction and mining equipment.

https://www.youtube.com/@CuttingEdgeEngineering

There's something hypnotic watching someone fabricate precision objects.

Wow, that’s an amazing channel! I expected to see big parts on big turning machines from your intro, but somehow they were even bigger and better than anticipated in the best possible way.

in equipment costs you are correct.

This is exciting but incredibly niche atm.

However, the "niche" is a non-trivial one--injection molds.

Machines already exist which deposit metal and also have a 3-axis CNC head that can then polish up the surface of that metal in the same envelope. This lets you directly print 3D molds with conformal cooling.

So, it doesn't matter if your machine is "slow" as long as it's faster than your alternatives for producing an injection mold.

It's not actually sand, it's glass balls.

Fancy sand!

Same thing

Nope, just google for "sand microscope".

Dude, when you cast molten metal into sand, you create glass micro balls. They are just starting from there. Sand is glass, heated to molten temperature and fused into a crystalline structure.

What is commonly understood as sand, has not much to do with little glass balls, even though here we have glass balls. A very special type of sand you can say, but not common "sand". But I feel no need to go more into pedantery here ... if this is sand to you, then I am fine with it.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.