On 05/02/2022 10:32, nop head wrote:

Exactly that! In theory, magic is possible...

If you have ever gone through the very basic programming training wrt
the difficulty of giving detailed instructions for making a pot of tea,
(or whatever the current instructional methods are) then you would
realise that what seems obvious to you, needs very detailed
explanation/definitions  to others, and others include computer
programs. Assumptions should not be made, there are always exceptions to
what you have assumed. Not a problem if exceptions are well defined, but
then it becomes the snake swallowing its tail. Klein bottle? How would
you design a rack for holding those using it's negative space, or even
generate its positive space - just saying...

Best wishes,

Ray

A related thing I often see a need for is something like a concave hull
that works like shrinkwrap. For example when a battery is made from the
shrink wrapped cells one might think the outer shape is the convex hull of
the cylinders but actually it is concave at the top and bottom.

[image: image.png]

I have no idea how to model that. A similar requirement for modelling
insulated crimp terminals.

[image: image.png]

On Sat, 5 Feb 2022 at 11:47, Raymond West raywest@raywest.com wrote:

I'm not sure how this is related to the original question, but for the
battery, why can't you just start with three circles, round the shape with
two applications of offset() and then apply linear_extrude?

The crimp terminal seems more difficult to model.  It could be done by a
skin operation.

On Sat, Feb 5, 2022 at 7:18 AM nop head nop.head@gmail.com wrote:

On 05/02/2022 12:17, nop head wrote:

In general, a 'shrink wrap vacuumed hull' would be useful, but in the
terminal specific case, other than looking nice, why would you need that
detail?* (I know the terminal is not shrink wrapped, perhaps just a
difficult shape to generate in openscad). But, with a significant bit of
effort, such terminals could be modelled. But, it borders on the 'need'
to be able to model physical characteristics of materials.

Ray

• I guess maybe the length to determine where the wire could be bent,
  the area around the spade to decide on the spacing of connectors, maybe
  the waist in case you want a fixing clip or something there, but the
  smooth fillets?

It isn't a linear_extrude for the battery because in the middle of the
cells the shortest path is straight, giving a regular convex hull, but at
the top and bottom the tension is from two planes, so the minimum surface
energy is concave.

It is related to the original problem, because if you could have the
shrinkwrap hull of the object you could subtract that from the panel but I
don't know if the "shrinkwrap" hull or concave hull is possible to define
for all shapes.

Perhaps minimum surface hull is well defined and would solve both problems.

On Sat, 5 Feb 2022 at 14:50, Raymond West raywest@raywest.com wrote: