Re: RP in Architecture

Date: Tue Nov 18 1997 - 03:23:18 EET


I was interested in your posting as I have been lurking around this news
group for about a month. My firm Laser Graphic Manufacturing currently
uses a digitally controlled laser cutter to build models and displays for
architects and developers. Several model firms that I know of have this
capability as it is fairly straight forward to convert architectural
drawings and get them to 2D shapes that can be cut on the laser.

In a very general sense we often use our own version of LOM (layered
object model ,- I think). We construct most of our topography bases from
layers of styrene sheet cut by the laser and manually laminate them into
a stack. The layers are from .07" to .2" so they are hardly what is
usually discussed in this arena but obviously the correspond to the
contour intervals on the site topography. We have also used this method
to create very realistic coursed stone walls by running the wall outlines
through fractal algorithms and stacking up the layers.

We are currently looking seriously at the desktop printer market and 3-d
systems has tried to model a sample for us on the Actua. We are looking
forward to trying out other machines but have not approached other
dealers yet.

There seem to be a number of issues in architecture that make it
different from mechanical modeling. First, I have actually been tracking
the threads on STL files and "repairs" quite closely. While and 3-D
drafting tools are fairly mature in the mechanical eng. industry (Ideas,
Solidworks and the like) they are not the norm in the Architectural
community, at least not that I have found in our customer base. (this is
not to say that there are not a lot of architects drawing in 3-D, only
that it is a newer approach in that industry). AS a "service bureau" for
architectural models it may mean that we have to build 3D models from 2D
sketches. We currently do this on a limited basis for many of our
projects but not to the level of creating a valid STL file.

Along those lines is a more interesting issue that we are trying to
define. Essentially at what point in the architectual design process
would RP be appropriate? Certainly in manufacturing, RP has taken off
because in many areas you can actually use the RP to make working, near
working and even production parts. I do not think that anyone has a
modeling envelope bid enough to make a RP house! However this route may
be useful in some areas to make custom fittings, moldings, hardware etc.
 What we have found in our work is that the time a model is most useful
in the architectural design process, the details of the space and form
are not fully defined. For instance I think that it might be quite
difficult to use RP to look at a roof line problem because you would have
to create some sort of "closed" object that the RP device could model.
Certainly the major issue for RP that is the same in both industries
would be the ability of RP to help prevent making a costly mistake and
help one quickly look at different design iterations.

Another issue that we find interesting is that overhanging parts are very
common on arch models and, at least in early testing of the Actua, the
physical supports are not really acceptable. If one wants to model
interior spaces with a ceiling, or even a roof eve or deck, you get a lot
of supports. There are a number of ways to get around this and I look
forward to perhaps some other comments on this problem.

That being said I do know of a model shop for a big firm that has used
SLA to create specific parts for models. I think they had some very
difficult mesh parts with complex curves and had to reproduce multiple
copies of this part. You might contact Todd Briton at Fentress,
Bradburn... who had these parts made.

I look forward to hearing the results of your study and any more comments
from this group as we feel it is a very exciting possibility both for the
RP and for the architectural industry.

Charles Overy
800 448-8808

This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:40:45 EEST