prusa3d / PrusaSlicer

G-code generator for 3D printers (RepRap, Makerbot, Ultimaker etc.)
https://www.prusa3d.com/prusaslicer/
GNU Affero General Public License v3.0
7.78k stars 1.94k forks source link

multiple support material (MSM) - Feature Request #3397

Open ReactSub opened 4 years ago

ReactSub commented 4 years ago

Behavior

Soluble interface is great for saving material. But I'd like to be able to use an even less expensive material as the bulk support.

Consider a three material setup that keeps temperatures close together (~240 C):

MatterHacker PRO series nylon $121/kg Ionic high temp soluble Hybrid $264/kg 3DXMAX ASA $38/kg

With the current soluble interface option it's cheaper to use the nylon as the bulk of the support (that's the point of the feature), since it's half the cost of the Hybrid. But by allowing the print body, soluble interface, and bulk support to be three different materials, the savings are accelerated.

FOR EXAMPLE:

Let's assume a small 25g print with 25g in support. Of the support, 20g is bulk support and 5g is interface. We'll assume a wipe tower with the exact same weights--so 100g total for all filaments used in total.

The material value of the final print is $3.02 in nylon.

Total cost for just the tower and support material

That's a 1/3 reduction in cost of waste material for a 25g print. The (overly) simple math is that for a 250g print, it's a $30 material savings (this assumes that the wipe tower grows at the same rate as the print, which is reductive, of course--still even $10 or $15 saved on a $130 print is significant).

Is this a new feature request? Yes.

bubnikv commented 4 years ago

Adding another material for the supports will trigger another tool change thus another wipe into the wipe tower. I am not quite sure whether you would save any material at the end of the day.

ReactSub commented 4 years ago

That's an important point. However, I am quite sure. I have a model in mind that my method would successfully apply to.

In the same way that sometimes a tower is three times the weight of model, and sometimes it's 50%, and can be anything between, we can't conclusively blanket that this wouldn't be useful. It's a feature request, after all.

You did get me rethinking, and now I realize it could be even better. Here's how: The across the board fix to what you outline is using the support structure as the wipe tower as "wipe support to (soluble) interface" in the same way as wipe-to-infill does--effectively using the support block as a wipe tower. It doesn't matter if the support structure dissolves in odd places during post. It's already throw-away.