Re: fiberglass plate domes

["J & D Goldman" <jmgoldma at dwx dot com>]

>
>--- J & D Goldman <jmgoldma at dwx dot com> wrote:
>> > I looked
>> >> at a fiberglass
>> >> unit a while back that used round panels, but the
>> holes
>> >> were just punched
>> >> where the struts would land, so it was pretty easy to
>> >> follow.
>> >> -Dan G.
>> >
>> >Dan -  Can you say anything more about this unit?
>> >
>> >Dick
>> >
>>
>> Several years ago I got brochures and pricing from the
>> company shown
>> below. I pulled this off the dnaco dome website, and at
>> this point
>> have no idea whether this is still a going concern or
>> not.   If you
>> try to contact them and it bombs, let me know and I'll
>> send you copies
>> of what I have on file (probably have phone numbers on
>> the
>> literature).  Hopefully they are still operating,
>> although, as I
>> recall, their domes were definitely not budget types.  I
>> suspect the
>> same technique could be used with other materials.
>
>What is the technique?
>
>
>Dick


I was just referring to this type of assembly in general.  I'm no
authority on this, and have no elegant method in mind to share, but to
help answer your question, it goes like this.  You start with a
material that can be curved in two directions.  (Even corrugated metal
can be curved in two directions, but the ease of doing this varies
with the material, thickness, etc.)  Now picture something like a
conduit dome, or any hub and strut type frame.  Take your bendable
material, and make circles, or patches of some kind, large enough to
span all three points of each triangle in the conduit dome, with
plenty of overlap.  Lay them over the framework, like large fish
scales (you can try this at home, but don't do it in the wind!
)    )  centering the patch over the center of each triangle. Bend
them so that they lay a bit more like a hemisphere than the hub n'
strut frame, or, make the patches really big, so that each pent or hex
is a single patch.   Then fasten the "scales" together using rivets,
bolts, screws, glue spots, spot welds, whatever.  You can join them
anywhere you want, but one possible pattern is to join the patches
along lines precisely where the struts would be.   If some of the
patches are left too small to completely over lap, you have the
makings of a window, skylight, or other opening.
   That said, I'm sure somebody out there (Steve Miller?) has a much
more quantitative and methodical approach to this.  I haven't check
the math or looked up Fuller's patent yet, but I hope to find some
improved approaches there.  The approach I describe here is a bit
clumbsy. In any case, I hope that makes it clearer what I meant by the
"technique".  If you see some pictures of the fiberglass domes, you
will notice they quite resemble what I have described here.  I'm sure
some work with a paper model would lead to an improved system.

-Dan G.