Simplify3D User Forum

i guess i would be envious if all i printed were half completed parts and not actually prototyping designs

BaudR8 wrote:i guess i would be envious if all i printed were half completed parts

Infill can show through transparent filament. Ugly infill can negatively impact people when you show them a prototype that's transparent.

BaudR8 wrote:and not actually prototyping designs

I do both prototypes for structural parts and objects designed just for art/beauty. Sometimes with prototypes, beauty is more important than structural strength. It all depends on what kind of prototype it is. If you're just making prototypes of things that need strength, you can use 100% infill. But when opting for something that's both beautiful and strong with 50% infill, Simplify3D has no real options available.

When it comes to strength, 100% infill is the best option. But if you also need to save on filament costs, then you need other choices, and you should know that one type of infill pattern will NOT be optimal for all prints. You might think honeycomb is optimal for strength while also saving on filament, but it depends on the shape of your print. There are cases where it will not work well, and you need a different pattern. With custom infill I can design infill that is optimal for every single print I make. The strongest infill, other than 100% infill, is infill designed specifically for the structure you are printing. There is no one size fits all for structural strength. It varies a lot. If honeycomb was always the answer, every bridge would use it. Every building would use it. But it's not optimal for everything. That's where custom infill comes in.

BaudR8 wrote:i guess i would be envious if all i printed were half completed parts and not actually prototyping designs

If you don't need infill, or you don't care what it looks like, that's great.

But if you want maximum strength at minimum plastic consumption, then specific infill patterns matter. Generically, hex infill is strongest.

Though what I'd really like to see would be intelligently generated infill, perhaps similar to how support is generated.

Infill has two functional goals (IMO, of course):

1) Hold up the tops of the object when it's too wide for bridging. For this, I think that infill should be like the new support, which lets you configure variable density so that you can put less material in most of the interior, and make it more dense right below the top surface, which saves material/speeds printing. For example, imagine using 10% infill until a few layers from the top(s), then 25% infill to build support, then the outer shell is 100% solid. Much more efficient than using 25% infill everywhere. As with external support "top" is really anywhere that there's too steep an overhang for the material to be self-supporting. So under areas that don't need internal support, there could be 0% infill, which is as efficient as you can get!

2) Make the part stronger. For this, regular grids are pretty good, hex is stronger, but stronger still is to do a "soap bubble" optimization, and print walls that are aligned in three dimensions to maximize support with minimal material. The placement would be based on supporting the exterior shell. There's a lot of interesting research published, and it makes a lot of sense, but no slicers do this that I know of.

You'd need a combination of the two. Ideally a "soap bubble" structure, with greater density as it approaches the top surfaces that need support.

Both of these would put S3D well ahead of any other slicer. I'd love to see it!

BaronWilliams wrote:I do both prototypes for structural parts and objects designed just for art/beauty. Sometimes with prototypes, beauty is more important than structural strength.

Actually, with prototypes, I have banned anything but "perfect" looking models from being viewed by outsiders. It is amazing how many people lack the imagination to overlook an ugly print, even when you show them a computer model.

tenaja wrote:

BaronWilliams wrote:I do both prototypes for structural parts and objects designed just for art/beauty. Sometimes with prototypes, beauty is more important than structural strength.

Actually, with prototypes, I have banned anything but "perfect" looking models from being viewed by outsiders. It is amazing how many people lack the imagination to overlook an ugly print, even when you show them a computer model.

Unfortunately, when showing a print, an ugly 3D print can sometimes cast doubts in peoples' eyes about your abilities despite the model actually being very well designed. If they understand 3D printing, and it's limitations, this is not likely, but for those who don't understand, an ugly print can poorly reflect your abilities. If you are having to explain why the print is ugly, things are not going so well. That's the last thing you want to talk about. That's why I try to give myself some buffer time before a model needs to been shown. I want to at least print a few test runs before showing a model. If the print is not good despite several attempts, then it's time to do some post processing (sanding, polishing, painting, etc.).

Back on topic, the structural properties of the soap bubble and the honeycomb structure don't completely apply to 3D prints. For example, a real soap bubble has it's molecules all equally adhering to each other. If we printed a bubble on a FFF 3D printer, each horizontal slice is a ring made of 1 line of plastic, with a start and end point. Between the start and end point, the line is 1 solid well fused line, but at the end points, it is a weak bond. So it's not a complete ring. It's a ring with a weak point at one end. It will also be impossible to print all the rings from bottom to top using a single wall for each layer without adding support material.

Likewise, the typical "honeycomb" infill most slicers offer doesn't give the strength advantage of a real honeycomb. It's a simulated honeycomb, not a real honeycomb. A true honeycomb structure is not really possible on an FFF 3D printer, because you can't even make a hexagon without having a weak bond somewhere in it, and you certainly can't link these hexagons together without several weak bonds between them.

The "honeycomb" infill that I have seen in many slicers isn't even a bunch of hexagons, but rather a bunch of zigzag lines that sort of look like a honeycomb when combined together. This simply does NOT give the same structural advantages a real honeycomb offers. You could get something similar by drawing hexagons between these zigzag lines, which will give you strength close to a real honeycomb, but this also requires more filament, so it's a trade off. But either way, the infill is pretty strong.

For FFF 3D printers, I think we're still in the early stages of development when it comes to optimal infill. More experimentation needs to be done.

Having infill that becomes less dense horizontally and vertically as it gets closer to the center would help speed things up. The ability to go from 70% infill at the outer regions to 10% at the center would be nice.

+1 Million for honeycomb infill. It provides superior strength to rectilinear infill and also looks good when using transparent filaments. More infill pattern options please.

Yes, please focus on better Infill options!

Craftware in my opinion has the best solution for generating infills that exists. It is highly customizable and extremely strong even with low percentages

If there was something like this in S3D would be perfect!

Extra options such as: Different grids in the same layer (instead of overlapping layers as it currently is) and Thickness of lines!

after a couple months of trying to get S3d into my workflow I come away with one area I'd like to see addressed.
And, if you might guess, it is the infill. Its time to add the options that are available in just about every slicer thats worth its salt.
MORE infill options. Until then I can not use S3d for my production parts.

Honestly, a nice API to allow 3rd party plugins would solve this. Allow us to design our own infills (or anything else with plugin allows access to), and they can solve many many problems.