Vary Infill?

[tonybuckley]

Hello,

Is there any software where I can vary the level of infill between two given layer numbers.

For if I want a few layers to be extra tough, I can specify?

I am printing something that has a weak point and I don't want to use a high infill for the whole print as it will take ages and use loads of filament.

Thanks,

Tony

[Mjolinor]

I don't know if there is a slicer that will do this but I do it by making small holes that are too small to actually show as holes but the conversion to Gcode treats them as an outside so you get "number of layers" of solid round them.

[tonybuckley]

Brilliant tip! Thank you.

[tonybuckley]

I have just read that Simpify3D can do this. That's 150 USD (100 GBP) and I have not bought any software so far as no reason to, but this is tempting.

I wish they would do a trial version, but on the plus side, they do offer a full refund if you don't like within two weeks.

Tony

[cperiod]

See http://slic3r.org/blog/modifier-meshes for discussion.

[Alibert]

Simpify3D can do it without problems. I have done something similar just yesterday. I had a part with a very difficult curve in it. I printed the first layer at 0.25mm, then 3 layers of 0.15mm, and the rest in layers of 0.2mm to solve that particular problem. Also you can load a table with several parts and have each part with its own infill, shells, supports and other settings. I frequently use that feature as well.

[tonybuckley]

See http://slic3r.org/blog/modifier-meshes for discussion.

That's very interesting! But how does one create the 'modifier mesh' in say 123D? Google was not helpful on this.

[cperiod]

That's very interesting! But how does one create the 'modifier mesh' in say 123D? Google was not helpful on this.

I made a cube in OpenSCAD the one time I actually used the capability, but if you're desperate i think you could just take an existing shape like a calibration cube and scale it to whatever size you need. The modifier doesn't need to be an exact match for what's being changed, it just has to cover the volume... An intersection operator, in other words.

[tonybuckley]

So if I understand, I have the model I wish to print. I place the new object where I want to change things.

So I end up with three STLs:
1: A file with the object AND the block where changes will be applied. This will never be sliced or printed.
2: A file with JUST the object I am printing.
3: A file with JUST the object I am using as the modifier mesh.

So I open file 2 as per normal. Then got to settings and load the modifer file 3 and make settings changes.

Then slice and print as per normal.

Have I got that right? :-)

Learn something every day.

[cperiod]

So if I understand, I have the model I wish to print. I place the new object where I want to change things.

So I end up with three STLs:
1: A file with the object AND the block where changes will be applied. This will never be sliced or printed.
2: A file with JUST the object I am printing.
3: A file with JUST the object I am using as the modifier mesh.

Just two STLs; 2 and 3. The file with just the object you're printing, and the modifier mesh describing the area where the changes are applied (which might have been derived from another STL if you, say, built it by scaling a calibration cube and saving it out as an STL, but that's another workflow). STL 1 isn't possible; you wouldn't be able to have an STL combining the object and the block since any slicer would treat it as just a single unified object.

You load the object STL into slic3r, then (in the settings for that object) you load the modifier for that object and change whatever you need, then generate your gcode.