Message: Re: holes in layers Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

None Re: holes in layers 

Forum: Geometry
Re: None holes in layers (Balint Radics)
Re: None Re: holes in layers (Michael H. Kelsey)
Date: 05 Dec, 2012
From: John Apostolakis <John Apostolakis>

I would add a couple of caveats about using multiple Boolean operations to create one solids:

In the current implementation of Boolean Solids:
  -  it is inefficient to make a volume out of many boolean operations.  The tracking time for a volume composed from N Boolean operations at best will be proportional to N, and potentially it will be slower;
  - using multiple subtractions ( A - B - C - D - E - F  ) creates volumes which are not very robust.  ( We are investigating some cases, and aim to improve it.)  If you decide to do something like this, it is much better to make a union first:
   U =  B + C + D + E + F + 
and then subtract that union from the original A volume.   ( Please read + as the union operator above. )

Developments is underway to speed up multiple unions, and it is planned for inclusion in one of the releases of 2013.


On Dec 5, 2012, at 5:45 PM, Michael H. Kelsey wrote:

> *** Discussion title: Geometry
> On Wed, 05 Dec 2012 14:32:17 GMT, Balint Radics wrote:
>> I have some layers of G4Tubs tube thin objects. I would like to make small 
>> holes in it (e.g. 50 um size holes with 50 um distance from each other). Would 
>> you please give an example how to do that?
> You want to create G4SubtractionSolids. Look in Section of the
> Application Developer's Book on "Boolean solids" for some simple
> examples. Note that Boolean solids are strictly binary (i.e., they take
> two existing G4VSolids as input), and that one of the two inputs must be
> a simple CSG solid. With this method, take care that the G4Tubs you use
> to generate the holes extends beyond the surfaces of the solid in which
> you are making the holes.
> If you have a very complex geometry, which is the same for all of your
> tubes, it may be more performant (but more difficult!) for you to write
> your own G4VSolid subclass which implements the geometry directly, in a
> parameterized way. The Toolkit Developer's Book discusses this in
> Section 3.1.2.
>  -- Michael Kelsey
> -------------------------------------------------------------
> Visit this GEANT4 at message (to reply or unsubscribe) at: 

 Add Message Add Message
to: "Re: holes in layers"

 Subscribe Subscribe

This site runs SLAC HyperNews version 1.11-slac-98, derived from the original HyperNews

[ Geant 4 Home | Geant 4 HyperNews | Search | Request New Forum | Feedback ]