51 points by wjb3 13 days ago | 9 comments
gravescale 13 days ago
That stress simulation is very interesting. I wonder would happen if they just filleted all the corners, inside and out.

They've added a lot of direct compression "paths" through the vertical web so it's not surprising it's unloaded the corners where the stress used to "go around" the void.

r14c 13 days ago
Integza posted an interesting video about novel applications of porous metal printed parts. Really excited to see how this develops.
aeonik 13 days ago
What temperature does it melt at? I'm curious to know more failure more.

I'm also curious how the human immune system reacts to it, though I'm sure it's not known yet.

rsfern 13 days ago
It’s a titanium alloy (Ti-6-4) that’s commonly used in medical implants, and it probably melts at pretty high temp - the wrought form is used in turbine engine compression disks

This is like a 3D printed metal scaffold, in the paper they’re doing compression tests. Probably more likely structural applications than medical I’m guessing

https://en.m.wikipedia.org/wiki/Ti-6Al-4V

capitainenemo 13 days ago
https://pubmed.ncbi.nlm.nih.gov/7634595/ ← from a search for qualities searched for in implants. "The primary advantage of TiAlV over CoCrMb is a lower modulus of elasticity. This results in decreased stress shielding and subsequent favorable femoral remodeling." This new material might do well there.. "TiAlV has been shown to achieve excellent bone ingrowth into porous surfaces." Well, given they printed this structure almost as a sponge it probably would do well on that front too.
rsfern 13 days ago
To be honest I don’t know too much about the biomedical side, I’m probably jumping to conclusions on what the materials research directions are there - originally I was just going to joke about how you might hope it doesn’t hit its melting temperature if it’s used in a medical implant as it starts to melt around 1600 C
0cf8612b2e1e 13 days ago
What kinds of parts are currently made with these of printers? I assume they are slow that it takes really interesting shapes or tolerances to warrant not using traditional manufacturing.
gravescale 13 days ago
GE's LEAP engine, for example, uses 3D printed parts, I think for the fuel injector nozzles.

This is a laser method, but binder jet printing is getting cheaper and easier by the year but it is still fairly slow.

j_m_b 13 days ago
Strong enough for a space elevator?
beambot 13 days ago
The lattice structures shown in the article are likely optimized for compression strength. Tensile strength is the limiting factor for space elevators. Very different design constraints.
DigitalHackOp 13 days ago
Not even close. They optimized for strength-to-weight ratios, which is great for a ton of things, but a space elevator needs insane levels of brute strength. Were not sure the material can exist in our current understanding of physics
Qwertious 13 days ago
Don't forget that space elevators need more than brute strength: they need to be immune to all sorts of cosmic radiation and space debris, they presumably need to be electrically conductive since the elevator carriage will need a lot of power to travel upwards a few thousand KMs before their passengers starve to death, they need to be robust against the elevator carriage's friction, and probably a few other things.
andreasmetsala 12 days ago
> they presumably need to be electrically conductive since the elevator carriage will need a lot of power to travel upwards a few thousand KMs before their passengers starve to death

Even a slow space elevator that can only be used for transporting cargo would already be extremely useful. You could even load it with a container filled with life support if you really need to ship living cargo.

Turing_Machine 13 days ago
An orbit to ground space elevator would indeed be challenging, but the cable doesn't have to go all the way to the ground to be useful. Shorter tethers have many applications, and those can be made out of non-exotic materials (e.g. Kevlar).

https://en.wikipedia.org/wiki/Space_tether

There are still challenges, of course.

readthenotes1 13 days ago
"Great for a ton of things" heh
avmich 10 days ago
Not strong enough to survive collisions of satellites with the space elevator.
wpollock 13 days ago
I thought limpet teeth were the strongest material known?
e44858 13 days ago
Has anyone tried printing this with an FDM printer?
rsfern 13 days ago
It’s not that kind of 3D printing, it’s laser powder bed fusion [0]. I don’t know a lot about FDM but I’d be kind of surprised if you could do Ti-64 and get the same mechanical strength (but I’d be interested to know if I’m off base) because I think it’s like metal embedded in polymer that you have to sinter post-build? Compared to basically doing a continuous laser weld with the powder bed method

However, here [1] is a cool paper looking at optimizing similar meta material structures using polymer filament printing - if you could do something similar in metal FDM that might be interesting?

0: https://en.m.wikipedia.org/wiki/Selective_laser_melting

1: https://doi.org/10.1126/sciadv.aaz1708

Turing_Machine 13 days ago
I'm not sure what difference that would make?

I mean, obviously metal is going to be stronger than fused plastic, but I gather here that what makes it stronger than normal is the structure.

Wouldn't FDM plastic objects made with this structure be stronger than FDM objects made with a more conventional structure? That would be very useful even if it's not as strong as metal.

I think I'm going to print up some test pieces and do destructive testing.

rsfern 12 days ago
Sorry - I didn’t mean to come off as discouraging, that’s part of why I linked the BEAR paper!

I agree in relative terms, I’d think this cellular structure should have better compressive strength than a different structure made of the same material

I guess where I’m coming from is that FDM titanium (if you can realistically print that, I don’t know) isn’t going to have the same properties as SLM titanium, so the headline strength (in absolute terms) is probably not achievable in FDM Ti

KevinMS 13 days ago
It already exists, its called infill
pbronez 12 days ago
Are you aware of any slicers that implement this tube-grid + plane-grid pattern?

Seems doable, but I haven’t seen it yet.

bborud 12 days ago
Observation: it took longer to respond to the GDPR popup than read the article. And to be frank, I just skimmed/quick read the descriptions of each choice.