Cheap, 3d-printed inner grids

[kvedera]

I know I've been overly active over the last few days, and I'm sorry for hogging the airwaves (this is my last post) (probably).
I just wanted to inform you friendly fellas that I put up a design on Shapeways.com for a inner grid, so if you want a perfectly symmetrical and circular inner grid just follow this link:
http://www.shapeways.com/model/907221/i ... =my-models
(The picture's a little perspectively misleading, look at the 3d animation for full view)
So yeah, just get a stainless steel one of those printed, and bob's your uncle.
Best of luck in all your endeavors,
Anthony Kveder

[Chris Bradley]

The surface roughness from a 3d print would probably mean that it will be sparkling and 'electro-cleaning' itself for ever.

[kvedera]

Yeah, maybe. It was just an idea.

[kvedera]

Anyway, is that really that bad?

[Chris Bradley]

Try it! There seems to be little to be gained by having a perfect grid, however, but that is not well researched - a few anecdotes either way. A printed grid would likely provide some much needed consistency between different grids, so as to see what forms compare with what.

Spark-erosion is one means to perform very accurate engineering, so maybe it would all come good once this printed grid 'settles in' and all the asperities are burnt off.

Some fusing of the pieces will also, likely, happen making it more solid, but possibly distorting your perfectly shaped grid as the micro-voids drop out of it. By the time it has burned in, internal distortions and external erosions might mean that the geometry is just as uncontrolled as a wire grid!

... but, yeah, do try it if you are motivated and can do it! I'm sure you will get due kudos here in the attempt, whatever the outcome.

[Rich Feldman]

Yes, Anthony, start getting your hands dirty!

I'm sure many readers would love to see pictures & cost data of a 3d-printed grid. If you get one or more made, maybe someone else could try it in a working fusor for you.

Another option for a symmetric, seamless, reproducible, spherical grid
would be to start with a thin hollow sphere, such as a stainless steel float.
Then machine or etch away unwanted material.

You are free from the constraint of overlapping great circles.
Why not make a grid not easily done with wire?
How about a soccer ball (buckminsterfullerene), or the pattern from some other familiar sports ball?

[kvedera]

Ah, good point! Thanks guys, I'll try that out. Unfortunately I won't be able to give results, because I haven't actually BUILT the damn thing yet, but I ordered the grid, and modeled a new buckyball-shaped one to try out. Thanks for the encouragement.

[Richard Hull]

All grids made of metal suffer extreme hydrogen imbrittlement over time. I have mentioned this before. Thus, springy grid wire forced into loops and circles, as in the normal construction, gets to such a degree of embrittlement that it winds up like a piece of glass, hard, brittle and forever formed into the geodesic.

I silver soldered my first grids together back in 97-98. I worried that the heat would pop the circles loose so I babied them along, never allowing them to get red hot. Over time, I noted that the silver solder was gone or seemed to have disappeared, yet the grid stayed intact. Finally, I did over temp a grid to a degree that a section of one of the geodesic wires melted and fused to two beads on each end of the wire arc.

I removed the grid and studied it. It was frozen and so brittle that a pair of pliers allowed me to snap off the melted arc sections without harming the rest of the structure or poping the loop open.

We observe that a process that would normally be considered a route to structural failure, turns out to be beneficial in this one instance.

Richard Hull

[kvedera]

So, do you think this would impair the efficiency or structural soundness of a 3d-printed one? It's made from powdered stainless steel, and has about 30% bronze mixed in with it. If you look at the one I made, it's pretty thin (I'm beginning to regret making it so thin, I may make another one). Might hydrogen imbrittlement damage it to the point at which it is almost useless? Be aware that it is not soldered together, but it simply one singular mesh of wires.

[kvedera]

Also, as I mentioned, the perspective in this picture makes it look like it has a rod running from one pole to the other -- this is not the case. The thing that looks like a pole is simply another ring, viewed sideways.

[Carl Willis]

I just don't see how 3D printing would result in a solid, durable piece of metal. Wire is solid metal drawn through a die. 3D printing is usually an accretion of particles in some kind of binder. With a site like that one, you can upload any design you want and it will figure out a price automatically, but it can't tell you if the piece is actually viable mechanically or has materials in it that are suitable for being used in a fusor. You should probably get someone familiar with the 3D printing process at that website to advise you on whether the part can actually be made as drawn.

-Carl

[Rich Feldman]

More along the lines of what Carl said.
What I would do is make a test part with a series of bar thicknesses and orientations. The bars would be long enough for destructive tests of their stiffness, bending strength, and ductility. And would go thin enough to probe the limit of fabricability. As if it were a new foundry-for-hire, offering to cast an unfamiliar material.

In either case, the first step is to ask for material specifications and dimensional design rules. For a less technical, consumer- or artist-oriented site, do what Carl said and just ask if your design is likely to come out as drawn. If the price is really under $10, the answer might be "try it".

I'd like to see if your 3D print shop compensates for shrinkage in the firing process. If the shrinkage is sensitive to bar thickness, then parts with large disparities might end up with internal stress, or even break while hot.

[Richard Hull]

Carl and Rich have valid points. We have no history related to powdered metal assembly resulting from 3-D printing thermal cycling stresses, etc. A powdered metal, even strong sintered metal would represent a new high in hydrogen absorption. Only testing in the real environment would tell the tale.

We know that drawn wire responds well even with extreme embrittlement, which for it is a plus, as all the tensions due to flexure during assembly within the structure are released over time in the fusor environment, freezing it to a specific shape.

Richard Hull

[John Futter]

Anthony
metal additive processes can make 99.98 percent dense output with around 30 micron granularity.
the laser sinterer near here makes SS 316 and Ti and Ti 46 alloy parts with the above performance and yes I have seen an impossibe to build any other way buckey ball within a buckyball test piece.
It is interesting to note that the parts are at least 99.8 as strong as extruded /forged parts.
I'm not sure of how the ebeam sintered parts compare but they should be comparable

[Chris Trent]

Here is the admittedly non-technical reference page for the stainless material at shapeways.

http://www.shapeways.com/materials/steel

It is not e-beam sintered stainless steel, more like bonded and brazed stainless powder. I have by doubts if they can actually manufacture the shape, but if they can it will be interesting to see someone try it.

[Donald McKinley]

Honest question. If it works as a grid, does it really have to be strong and durable? I don't recall any information about how strong except if its too thin, it's probably too easy to melt.

Other items work pretty well without much strength. A coleman lantern mantle for instance. And it gets much hotter.

Go for it Anthony, let us know.

D

[Richard Hull]

We have talked about a lot of ideas related to fusor structures, components and other minutia, but about 80% of the bold ideas are pretty much "wind over the decks".

As always, we might not really expect to see this idea actually implimented in a functional, fusing device here.

Much like power ready fusion, we will all just have to wait and see what really transpires down the road a bit.

Richard Hull

[Dany_B_Drolet]

First of all, sorry for necro post, but the forums here aren't that active.

Second, I work in a 3D printing and additive manufacturing shop. The owner, which I know well since we're just a handful of people there and since I'm pretty much his R&D assistant (my title is pretty ambiguous), is a proven physicist specialized in nano-scale deposition systems. He shown me some printed metal components and we have all the machinery to make the grid. I'm not sure I'll take your grid model, but I'll talk to him about making one (probably using SLS) and try it in my Fusor project once I get there.

I can't believe how much such a device gets more and more complex as I understand it more and more... But hey, gotta persevere.

-Dan

[John Futter]

Hi all
Dany is right
open your minds to impossible to machine items

I have had Ti6Al4V items made that are extremely good more later when it becomes public

[Dany_B_Drolet]

Heard me when I talked about using *additive manufacturing* ? Using SLS, I am certain to be able to print it in aluminium, then smelt it to create a perfect grid. I am not sure about other alternatives. But my boss is working on much more elaborate techniques using nanoparticles. Gotta ask him this week.

Impossible to machine yes, but you can still build it from the ground up!

EDIT: Oops, looks like I misinterpreted your message, John. Sorry for that!

[Adam Szendrey]

Has there been any progress in this matter? I have been considering printed grids as one of the experiments for my slowly re-emerging device. It's not THAT cheap tho'. I made my first grid out of welding wire. I'm not convinced of the actual benefits, but it'd be an interesting thing to throw at a fusor. Shapeways charges by volume, I think in the case of stainless tho, it's not the envelope volume but the actual volume of the geometry. At any rate I can't seem to find any followup to this on the forums, so if anybody has some experimental data, or at least an actual printed grid of any type, I'd be interested. Thanks.