Hex infill? (finding a infill good for final parts)

[joergen 2]

Hi Cura team,

 

Feature request: please implement a proper hex comb infill into cura

(if it is already implemented, please let me know how and where to find it)

 

Reason: hex comb infill is an old and very reliable infill for practical 3D prints that get used directly. within the hex comb parameters (in other slicers) one can create very strong parts with only 2 shells and a 20% infill.

none of the current CURA infill options come close to the strength and reliability, as well as ease of printing, compared to the hex comb infill

 

the lack of hex comb infill in CURA has been the main reason why I am still slicing with other slicers.

 

I hope that hex comb is not plagued by some nasty intellectual property issues from i.e. the netfabb days, or because hex comb wasn't part of the original skeinforge code...

 

the part i am printing today had a volume of 60x7mm and about 40mm tall, and all of cura infills (even when set to 65% infill) fail to produce a structurally valid infill. "cubic subdivision" came closest to create something meaningful, but the infill prints very bumpy, causing tons of vibrations.

Edited September 12, 2021 by joergen

[GregValiant 1,330]

For strength I use the Grid with the line directions at [0,90].  This is with 15% infill, "connect infill lines" enabled, and "Infill Line Multiplier" at 2.  I can park my car on prints with this infill (I know that because of a blind test my wife conducted).  It also doesn't matter how the block is oriented.  Unlike Hex, the load capability of any face is the same.  I don't use it often because of the hit on print time.  At 2 walls I would get marks at the infill lines.  With a car parked on top they are hard to see.

 

 

 

 

 

[Smithy 1,145]

14 minutes ago, GregValiant said:

With a car parked on top they are hard to see.

🤣😂

[nallath 1,124]

We just never bothered to implement hex infill because we didn't really see the point. It might make things a tiny bit stronger, but you would also need to stop extruding or do other tricks to get it to work.

So as far as we've been able to tell, it's better in theory, but in practice it doesn't really matter. The cross infill gets you the same bang for your buck, but is a lot easier to implement.

But if someone shows some tests that prove that hex infill is actually better (strength tests, tensile tests, etc), we can be convinced.

[joergen 2]

Quote

doesn't really matter. The cross infill gets you the same bang for your buck, but is a lot easier to implement.

 

never mind the 10 years of commercial 3D printing experience I have ever since I got my UM in 2011.
As mentioned, other slicers (from the good old/defunct netfabb, to simplify3D (also abandoned) and others), hexcomb works great. I personally like a low infill for larger parts (20%) with a 1mm line width, and 35% with 0.6mm lines for smaller parts...

as I said, it was a feature request, if there is no need for it in your opinion, you don't have to.

[joergen 2]

Greg, I did try the grid. I had gradual infill steps accidentally turned on, which screwed up the infill completely. without it, grid works nicely, thx for the hint

Edited September 13, 2021 by joergen

[GregValiant 1,330]

With many models the line direction matters.  When I have to spin a model at 45° to get it to fit the build plate then I rotate the infill and skin to match.  Keeping the Infill and Skin at 45° to each other works for me.  For parts that don't really need the inner strength then using an Infill Layer Height of 2X Layer Height makes for quicker prints.  The same can be done with support infill layer thickness.

[joergen 2]

Hi Greg,
yes, the last 10 years have taught me that the direction matters a great deal.
one of the advantages of the hex comb infill is the independence of the infill from the orientation

[nallath 1,124]

17 hours ago, joergen said:

 

never mind the 10 years of commercial 3D printing experience I have ever since I got my UM in 2011.
As mentioned, other slicers (from the good old/defunct netfabb, to simplify3D (also abandoned) and others), hexcomb works great. I personally like a low infill for larger parts (20%) with a 1mm line width, and 35% with 0.6mm lines for smaller parts...

as I said, it was a feature request, if there is no need for it in your opinion, you don't have to.

Never mind the 9 years of commercial experience that I have by working for Ultimaker then 😉?

I'm not arguing that it has no value (or doesn't work at all), but rather that it doesn't provide any meaningful benefit to what is already out there. This especially when considering the amount of work that needs to be done to implement it. If the infill is significantly better, we could (and should) reconsider this decision, but as I said, I've not seen any evidence that that is the case.

But for every thing we build, there are a number of other things that we don't build. So as much as that sucks, we have to always weigh the value of building something to the other things we could be spending our time on.