# Predictability of scale....

Posted by Replace
 Predictability of scale.... December 24, 2016 01:18AM Registered: 10 years ago Posts: 861
I am fond of Scale.
But there is a question on the usage:

When I want to make a hollow object, with predefined wall thickness,
I design the object and Difference it with a scaled smaller copy, that is translated in Z, to a defined altitude that reflects the wall thickness on the build plate.

However, the thickness of other walls: left, right, front, rear and top , (or with more complex objects all other outside surfaces) is depending on the scale parameters.

Can I calculate the thickness as a function of both the original size and the desired thickness, to get the scale parameter ?

Thomas

www.3daybreaker.blogspot.com

Orca V4.4 rebuild to Ramps with Mk8 and E3D, as well as a Rostock Delta Mini and an OLO in backorder :-)
 Re: Predictability of scale.... December 24, 2016 09:58AM Registered: 6 years ago Posts: 590
Your proposed approach using a difference with a scaling version of the full object to get a predefined wall thickness will not work in general. It will only work for very specific simple shapes (such as cubes and spheres), since it will result in a variable wall thickness that is essentially proportional to the distance from the origin.

What you would need to do this computing the Minkowski difference of you object and a sphere with a radius equal to the wanted wall thickness and subtract the result from the original object. But, unfortunately, OpenScad only implements the Minkowski sum, not the difference...

The only "easy solution" I could see, is to let the slicer do the work by printing the full object with a 0% filling and the number of shells and top/bottom layers that correspond to the desired wall thickness.
 Re: Predictability of scale.... December 25, 2016 07:01AM Registered: 10 years ago Posts: 861
I was not aware of a possible minkowsky difference . Enyhow it is not there

Your solution gives me a hard time programming internal ribs and supports in the hollow shape, as far as I can think-up now.

But I will keep it in mind designing. Thanks

www.3daybreaker.blogspot.com

Orca V4.4 rebuild to Ramps with Mk8 and E3D, as well as a Rostock Delta Mini and an OLO in backorder :-)
 Re: Predictability of scale.... January 26, 2017 09:54AM Registered: 6 years ago Posts: 590
Just a note regarding the Minkowski difference: In this other thread I proposed a simple implementation of the Minkowski difference using the Minkowski sum on the "negative" (i.e. a cavity) of the original object...
Sorry, only registered users may post in this forum.