Temcor domes, revisited
320 views
Skip to first unread message
Gerry Toomey
Jan 4, 2012, 7:10:38 PM1/4/12
to geodes...@googlegroups.com
Background....
Many moons ago Taff made a SketchUp model of the Temcor dome that was formerly part of the US science station at the South Pole. (The dome was torn down a few years ago.) He later modeled another Temcor dome, the Dietrich Activity Center at Walla Walla Community College in Washington State, USA. The South Pole dome had 12 rows of triangles, with five triangles converging at the building's apex, while the Walla Walla dome has 10 rows of triangles, with 6 triangles at the apex. Adrian Rossiter also created a Windows binary for the Temcor subdivision method that can be used with his Antiview/Antiprism software. I haven't tried it yet as I'm a total neophyte on the programming side (LOL).
More Temcor layouts...
I've continued to play around with the Temcor subdivision method in Excel and have put together four simple layouts that further illustrate its geometric flexibility. In the attached jpgs, the number of chord factors equals n + 1 where n is the number of rows of triangles. The number of triangle types equals n. The ratio of dome height to footprint diameter, expressed as the theta angle in each caption, is arbitrary; that is, it's a design variable.
These are "low-frequency" domes -- just 3 to 6 rows of triangles. But I'm currently doing calcs for domes up to 16 rows, and (probably) a choice of 4 to 8 triangles converging at the dome apex.
Excel was used for the spherical trig calcuations and to compile the OFF files (thanks for the earlier help/feedback, Dondalah). I then imported the OFF files into RoffView (www.holmes3d.net/graphics/roffview) to generate the wire-frame images. For anyone not familar with OFF files, they are numerical descriptions of two- or three-dimensional shapes such as a triangle (a polygon) or a geodesic dome (a polyhedron). Basically, an OFF file lets you define a set of points in virtual space, using their cartesian coordinates, and then "connect the dots" to create edges and faces.
Anyway, the Temcor geodesic subdivision method seems to be quite versatile for the design of large domes ... I'm surprised not to have heard much about it other than on this discussion group via Taff.
- Gerry T. in Quebec
TaffGoch
Jan 4, 2012, 7:50:10 PM1/4/12
to geodes...@googlegroups.com
On Wed, Jan 4, 2012 at 6:10 PM, Gerry Toomey <toomey...@gmail.com> wrote:
> ... I'm surprised not to have heard much about it other than on this discussion group via Taff.
> ... I'm surprised not to have heard much about it other than on this discussion group via Taff.
I suspect that Temcor/Richter wanted the subdivision method kept a
proprietary "secret" -- thus no public documentation.
That is surprising, since Clinton listed/described, at least, eight
different subdivision methods. In his descriptions, he progresses
right up to describing the Richter method, but stops short. Perhaps
there was a gentleman's agreement, among the
geodesic-wizards-of-the-day, not to "reveal" the Kaiser/Temcor/Richter
subdivision method.
_____________________
Regarding the Antarctic dome, I was initially impressed at how smooth
were the arc transitions, when crossing icosahedral boundaries. That
sparked my further explorations. (If I had utterly failed to
reverse-engineer the subdivision method, you wouldn't have heard about
it!)
_____________________
Gerry, the arcs in some of your images appear to "wander" a bit, away
from smooth "great circle" arcs. In previous discussions, we never did
get to the point of describing the technique mathematically. (I
composed all of my constructions by "virtually" rotating &
intersecting planar, circular disks.)
Have you concluded your spreadsheet mathematical-derivation method, or
are you still experimenting?
If you're interested, I can compare outcomes in SketchUp models, if
you provide the cartesian coordinates.
-Taff
Gerry in Quebec
Jan 5, 2012, 7:29:05 AM1/5/12
to Geodesic Help Group
Hi Taff,
Two possible reasons why the images wander a bit from the smooth lines
of the South Pole dome.... First, when there's a small number of rows
of triangles spread out over a large theta angle, the variation in
triangle size is large. Second, there seems to be some kind of
perspective-related distortion in the 3D images generated by RoffView.
I noticed this when looking at dome images in plan view. Some of the
triangles near ground level that should have been visible were not.
It's as if the perspective has somehow been exaggerated (opposite what
you get with an isometric view).
I'll post Excel csv data files (cartesian coordinates) of the four
domes in a separate message. I'll also include a fifth csv file to
illustrate what happens when a small number of rows of triangles (six)
is spread out over a large theta angle (144 degrees, between the dome
apex and the dome footprint). This contrasts with the South Pole dome
where 12 rows of triangles were distributed over only 65.45 degrees.
I'm still working on the trig (in Excel) for a Temcor-style dome
calculator. I'm doing a separate worksheet for each odd/even pair of
domes according to the number of rows of triangles. For example, one
sheet for 3 or 4 rows of triangles, one sheet for 5 or 6 rows, etc.
While the pattern of spherical trig calculations is essentially the
same for each sheet, I haven't been able to figure out how to
generalize the equations into a single algorithm in which the number
of rows is a design variable.
What I expect to end up with (time permitting) is a set of Excel
worksheets that can generate vertex coordinates, chord factors and OFF
files for Temcor-style domes, with the following design limitations:
- Number of rows of triangles: 3 to 16.
- Number of triangles converging at the dome apex (n-fold symmetry): 4
to 8.
- Any ratio of dome height to footprint radius, with the understanding
that extreme values (very short or very tall) may result in weird
looking or impractical dome designs -- e.g., the near-sphere with a
theta angle of 144 degrees.... which might make a nice lampshade ;-).
- Gerry
On Jan 4, 7:50 pm, TaffGoch <taffg...@gmail.com> wrote:
....>
Two possible reasons why the images wander a bit from the smooth lines
of the South Pole dome.... First, when there's a small number of rows
of triangles spread out over a large theta angle, the variation in
triangle size is large. Second, there seems to be some kind of
perspective-related distortion in the 3D images generated by RoffView.
I noticed this when looking at dome images in plan view. Some of the
triangles near ground level that should have been visible were not.
It's as if the perspective has somehow been exaggerated (opposite what
you get with an isometric view).
I'll post Excel csv data files (cartesian coordinates) of the four
domes in a separate message. I'll also include a fifth csv file to
illustrate what happens when a small number of rows of triangles (six)
is spread out over a large theta angle (144 degrees, between the dome
apex and the dome footprint). This contrasts with the South Pole dome
where 12 rows of triangles were distributed over only 65.45 degrees.
I'm still working on the trig (in Excel) for a Temcor-style dome
calculator. I'm doing a separate worksheet for each odd/even pair of
domes according to the number of rows of triangles. For example, one
sheet for 3 or 4 rows of triangles, one sheet for 5 or 6 rows, etc.
While the pattern of spherical trig calculations is essentially the
same for each sheet, I haven't been able to figure out how to
generalize the equations into a single algorithm in which the number
of rows is a design variable.
What I expect to end up with (time permitting) is a set of Excel
worksheets that can generate vertex coordinates, chord factors and OFF
files for Temcor-style domes, with the following design limitations:
- Number of rows of triangles: 3 to 16.
- Number of triangles converging at the dome apex (n-fold symmetry): 4
to 8.
- Any ratio of dome height to footprint radius, with the understanding
that extreme values (very short or very tall) may result in weird
looking or impractical dome designs -- e.g., the near-sphere with a
theta angle of 144 degrees.... which might make a nice lampshade ;-).
- Gerry
On Jan 4, 7:50 pm, TaffGoch <taffg...@gmail.com> wrote:
....>
Gerry Toomey
Jan 5, 2012, 8:58:57 AM1/5/12
to geodes...@googlegroups.com
Taff,
Here are the Excel cvs files for the four dome layouts I posted -- plus one extra, an extreme example, where theta = 144 degrees.
In each case I've given the cartesian coordinates for one symmetry cluster of triangles. So if the dome has five or six triangles converging at the apex you have to intersect five or six clusters respectively to model the full dome.
- Gerry
Gerry in Quebec
Jan 5, 2012, 4:26:40 PM1/5/12
to Geodesic Help Group
Taff,
Removing one or two rows of triangles from a Temcor layout (e.g.,
using only 5 rows of an otherwise 6-row layout, or only 12 in a 14-row
layout) may result in a smoother cross-sectoral arc, at least at the
bottom of the dome. It's also possible that when the end points of the
reference geodesic (chords of which are the white struts in the
original South Pole dome diagrams) have a theta angle of 37.3774
degrees, and when the number of triangles converging at the apex is 5,
you will have smoother cross-sectoral arcs. Ditto when the reference
geodesic is at 63.4349 degrees. Those angles correspond to the great
circle interesection points for class II and class I icosa domes
respectively. Similar patterns may apply to octa domes.
Just a guess.
- Gerry
On Jan 5, 7:29 am, Gerry in Quebec <toomey.ge...@gmail.com> wrote:
> Hi Taff,
> Two possible reasons why the images wander a bit from the smooth lines
> of the South Pole dome.... First, when there's a small number of rows
> of triangles spread out over a large theta angle, the variation in
> triangle size is large. Second, there seems to be some kind of
> perspective-related distortion in the 3D images generated by RoffView.
....
> > -Taff-
Removing one or two rows of triangles from a Temcor layout (e.g.,
using only 5 rows of an otherwise 6-row layout, or only 12 in a 14-row
layout) may result in a smoother cross-sectoral arc, at least at the
bottom of the dome. It's also possible that when the end points of the
reference geodesic (chords of which are the white struts in the
original South Pole dome diagrams) have a theta angle of 37.3774
degrees, and when the number of triangles converging at the apex is 5,
you will have smoother cross-sectoral arcs. Ditto when the reference
geodesic is at 63.4349 degrees. Those angles correspond to the great
circle interesection points for class II and class I icosa domes
respectively. Similar patterns may apply to octa domes.
Just a guess.
- Gerry
On Jan 5, 7:29 am, Gerry in Quebec <toomey.ge...@gmail.com> wrote:
> Hi Taff,
> Two possible reasons why the images wander a bit from the smooth lines
> of the South Pole dome.... First, when there's a small number of rows
> of triangles spread out over a large theta angle, the variation in
> triangle size is large. Second, there seems to be some kind of
> perspective-related distortion in the 3D images generated by RoffView.
> - Gerry
>
> On Jan 4, 7:50 pm, TaffGoch <taffg...@gmail.com> wrote:
> ....>
> > Gerry, the arcs in some of your images appear to "wander" a bit, away
> > from smooth "great circle" arcs.
....
>
> On Jan 4, 7:50 pm, TaffGoch <taffg...@gmail.com> wrote:
> ....>
> > Gerry, the arcs in some of your images appear to "wander" a bit, away
> > from smooth "great circle" arcs.
> > -Taff-
Dave Kruschke
Oct 30, 2012, 1:27:24 PM10/30/12
to geodes...@googlegroups.com
Well, I'm not surprised that Temcor's versitile designs for large domes haven't attracted much interest. When it comes down to building one of these domes, one needs very deep pockets and a lot of confidence in their geometry. Moreover, I worked at Temcor around 1969 on derivations and never heard of any rule where the number of chord factors equals n+1 where n equals the number of triangles in a row - or whatever...
Dave Kruschke
Oct 30, 2012, 1:48:34 PM10/30/12
to geodes...@googlegroups.com
Having worked at Temcor, on the South Pole Dome, no less (I was the very most junior member of this group), I worked on derivations of chord factors and can say that the Richter method almost always involved working in the Sphere (radius = 1) rather than using some kind of x/y/z approach that needed a lot of computer power that Temcor didn't have. My strong opinion is that Clinton didn't list the Richter method because he didn't know it. Yes, this subdivision was certainly kept a proprietary secret. In fact, I wasn't allowed to even look at a "P Drawing" at Temcor until I had worked there for a while. And I doubt if there was any kind of "gentlemen's agreement" regarding Richter and Clinton. Richter probably never heard of Clinton. Finally, I strongly believe that there was only one "geodesic-wizard-of-the-day," and that was Don Richter - not Kaiser, not Temcor, not Clinton and certainly not me.
Dave Kruschke
PS; Sometimes just a little x/y/z work was done so one could draw the Plan and Elevation view of a given dome. And I think some x/y/z work was done (not by me) on the South Pole Dome because the government wanted some special stress analysis done on this dome (hey, heavy snow loads) by outside consultants whose software needed x/y/z data to operate...
Ken G. Brown
Oct 30, 2012, 4:07:15 PM10/30/12
to geodes...@googlegroups.com
Can anyone point me to an explanation of how the Temcor subdivision is done?
Thx,
Ken
--
You received this message because you are subscribed to the "Geodesic Help" Google Group
--
To unsubscribe from this group, send email to GeodesicHelp...@googlegroups.com
--
To post to this group, send email to geodes...@googlegroups.com
--
For more options, visit http://groups.google.com/group/geodesichelp?hl=en
TaffGoch
Oct 30, 2012, 4:35:28 PM10/30/12
to Geodesic Help Group
Dave,
I used to employ cartesian-coordinate calculations (exclusively,)
years ago. Once I mastered the rotational & intersection tools offered
by SketchUp, I haven't done any "calculations" at all. (No need.)
SketchUp has 6-decimal precision, when typing in values, but it is
apparent that 9-to-12 decimal accuracy is inherent to plane
intersections, but is not displayed.
I, especially, marveled at how well the Temcor-method "manipulations"
turned out.
-Taff
I used to employ cartesian-coordinate calculations (exclusively,)
years ago. Once I mastered the rotational & intersection tools offered
by SketchUp, I haven't done any "calculations" at all. (No need.)
SketchUp has 6-decimal precision, when typing in values, but it is
apparent that 9-to-12 decimal accuracy is inherent to plane
intersections, but is not displayed.
I, especially, marveled at how well the Temcor-method "manipulations"
turned out.
-Taff
TaffGoch
Oct 30, 2012, 4:44:46 PM10/30/12
to Geodesic Help Group
Ken,
Here's the full description:
Be sure to read each post, all the way through the discussion. It was a trial-and-error process, and you should note that my early impressions of the Antarctic dome subdivision were erroneous. I corrected my mistaken assumption, later in the discussion, which lead, finally, to a correct assessment of the subdivision method used by Temcor, as devised by Don Richter.
Note that there are 3 pages of discussion, and you have to use the "newer" link at the bottom of pages 1 and 2, to fully access all the posts.
-Taff
TaffGoch
Oct 30, 2012, 4:46:45 PM10/30/12
to Geodesic Help Group
Ken,
If you use SketchUp, you can study the final-result 3D model that I posted at Google's 3D Warehouse:
-Taff
Ken G. Brown
Oct 30, 2012, 5:56:13 PM10/30/12
to geodes...@googlegroups.com
I have Sketchup but haven't yet gotten very familiar.
Ken
Gerry in Quebec
Nov 1, 2012, 11:51:55 AM11/1/12
to Geodesic Help Group
Hi Dave & others,
The n+1 "rule" doesn't refer to the "number of triangles in a row" but
to the number of rows of triangles. This excludes the chord factors of
the struts forming the add-on skirt of triangles at the bottom of the
dome to make the dome sit flat. The rule is based on Taff's multiple-
plane description of the Temcor method, which in turn reflects the
strut information originally supplied by by Don Richter for the
Antarctic dome you worked on.
- Gerry
On Oct 30, 1:27 pm, Dave Kruschke <theabunda...@yahoo.com> wrote:
> Well, I'm not surprised that Temcor's versitile designs for large domes
> haven't attracted much interest. When it comes down to building one of
> these domes, one needs very deep pockets and a lot of confidence in their
> geometry. Moreover, I worked at Temcor around 1969 on derivations and never
> heard of any rule where the number of chord factors equals n+1 where n
> equals the number of triangles in a row - or whatever...
>
>
>
> On Wednesday, January 4, 2012 6:10:38 PM UTC-6, Gerry in Quebec wrote:
>
> > *Background....*
The n+1 "rule" doesn't refer to the "number of triangles in a row" but
to the number of rows of triangles. This excludes the chord factors of
the struts forming the add-on skirt of triangles at the bottom of the
dome to make the dome sit flat. The rule is based on Taff's multiple-
plane description of the Temcor method, which in turn reflects the
strut information originally supplied by by Don Richter for the
Antarctic dome you worked on.
- Gerry
On Oct 30, 1:27 pm, Dave Kruschke <theabunda...@yahoo.com> wrote:
> Well, I'm not surprised that Temcor's versitile designs for large domes
> haven't attracted much interest. When it comes down to building one of
> these domes, one needs very deep pockets and a lot of confidence in their
> geometry. Moreover, I worked at Temcor around 1969 on derivations and never
> heard of any rule where the number of chord factors equals n+1 where n
> equals the number of triangles in a row - or whatever...
>
>
>
> On Wednesday, January 4, 2012 6:10:38 PM UTC-6, Gerry in Quebec wrote:
>
>
> > Many moons ago Taff made a SketchUp model of the Temcor dome that was
> > formerly part of the US science station at the South Pole. (The dome was
> > torn down a few years ago.) He later modeled another Temcor dome, the
> > Dietrich Activity Center at Walla Walla Community College in Washington
> > State, USA. The South Pole dome had 12 rows of triangles, with five
> > triangles converging at the building's apex, while the Walla Walla dome has
> > 10 rows of triangles, with 6 triangles at the apex. Adrian Rossiter also
> > created a Windows binary for the Temcor subdivision method that can be used
> > with his Antiview/Antiprism software. I haven't tried it yet as I'm a total
> > neophyte on the programming side (LOL).
>
> > *More Temcor layouts...*
> > Many moons ago Taff made a SketchUp model of the Temcor dome that was
> > formerly part of the US science station at the South Pole. (The dome was
> > torn down a few years ago.) He later modeled another Temcor dome, the
> > Dietrich Activity Center at Walla Walla Community College in Washington
> > State, USA. The South Pole dome had 12 rows of triangles, with five
> > triangles converging at the building's apex, while the Walla Walla dome has
> > 10 rows of triangles, with 6 triangles at the apex. Adrian Rossiter also
> > created a Windows binary for the Temcor subdivision method that can be used
> > with his Antiview/Antiprism software. I haven't tried it yet as I'm a total
> > neophyte on the programming side (LOL).
>
Dave Kruschke
Nov 1, 2012, 12:27:30 PM11/1/12
to geodes...@googlegroups.com
Well, I remember some "rules" for creating Temcor Dome triangles and there was no mention of a n, n+1 "rule." This "rule," interesting if valid, is not necessary to derive the chord factors of the various Temcor Domes that I worked on. For example, some of the Temcor Domes don't have a five sided "pent" at the top but instead have a six sided "pent" and aren't connected with the geometry of the icosahedron. Do you think that this n, n+1 "rule" applies regardless of the initial spherical triangles as viewed from the top of the dome?
Dave Kruschke
From: Gerry in Quebec <toomey...@gmail.com>
To: Geodesic Help Group <geodes...@googlegroups.com>
Sent: Thursday, November 1, 2012 10:51 AM
Subject: Re: Temcor domes, revisited
--
You received this message because you are subscribed to the "Geodesic Help" Google Group
--
You received this message because you are subscribed to the "Geodesic Help" Google Group
--
To unsubscribe from this group, send email to GeodesicHelp+unsub...@googlegroups.com
Gerry in Quebec
Nov 1, 2012, 1:49:30 PM11/1/12
to Geodesic Help Group
Hi Dave,
Yes, the same rule applies whatever the number of triangles converging
at the apex. I will post a diagram to show what I mean because it's
hard to express this clearly in words. Gotta run -- the brakes of my
old beat-up GMC pickup have finally been repaired and I have to go get
it now.... It will be nice to have wheels again. I will post that jpg
a little later.
- Gerry
On Nov 1, 12:27 pm, Dave Kruschke <theabunda...@yahoo.com> wrote:
> Well, I remember some "rules" for creating Temcor Dome triangles and there was no mention of a n, n+1 "rule." This "rule," interesting if valid, is not necessary to derive the chord factors of the various Temcor Domes that I worked on. For example, some of the Temcor Domes don't have a five sided "pent" at the top but instead have a six sided "pent" and aren't connected with the geometry of the icosahedron. Do you think that this n, n+1 "rule" applies regardless of the initial spherical triangles as viewed from the top of the dome?
>
> Dave Kruschke
>
> ________________________________
> From: Gerry in Quebec <toomey.ge...@gmail.com>
> To unsubscribe from this group, send email to GeodesicHelp...@googlegroups.com
Yes, the same rule applies whatever the number of triangles converging
at the apex. I will post a diagram to show what I mean because it's
hard to express this clearly in words. Gotta run -- the brakes of my
old beat-up GMC pickup have finally been repaired and I have to go get
it now.... It will be nice to have wheels again. I will post that jpg
a little later.
- Gerry
On Nov 1, 12:27 pm, Dave Kruschke <theabunda...@yahoo.com> wrote:
> Well, I remember some "rules" for creating Temcor Dome triangles and there was no mention of a n, n+1 "rule." This "rule," interesting if valid, is not necessary to derive the chord factors of the various Temcor Domes that I worked on. For example, some of the Temcor Domes don't have a five sided "pent" at the top but instead have a six sided "pent" and aren't connected with the geometry of the icosahedron. Do you think that this n, n+1 "rule" applies regardless of the initial spherical triangles as viewed from the top of the dome?
>
> Dave Kruschke
>
> ________________________________
TaffGoch
Nov 1, 2012, 2:16:47 PM11/1/12
to geodes...@googlegroups.com
Example of a 6-sided Kaiser/Temcor dome -- Chinese pavilion at the 1974 Spokane world's fair. (Now serves as roof of Dietrich Activity Center at Walla Walla Community College.)
Subdivision of the hexagonal face of a cuboctahedron, using Richter's method.
-Taff
Subdivision of the hexagonal face of a cuboctahedron, using Richter's method.
-Taff
Dave Kruschke
Nov 1, 2012, 4:14:11 PM11/1/12
to geodes...@googlegroups.com
Thanks...
From: Gerry Toomey <toomey...@gmail.com>
To: geodesichelp <geodes...@googlegroups.com>
Sent: Thursday, November 1, 2012 2:06 PM
Subject: Temcor domes, revisited
Excluding skirt triangles, the number of different strut lengths equals the number of rows of triangles plus 1. This is independent of the number of triangles converging at the apex.
- Gerry
Dave Kruschke
Nov 1, 2012, 4:16:16 PM11/1/12
to geodes...@googlegroups.com
Hmmm...
From: TaffGoch <taff...@gmail.com>
To: geodes...@googlegroups.com
Sent: Thursday, November 1, 2012 1:16 PM
Subject: Re: Temcor domes, revisited
Example of a 6-sided Kaiser/Temcor dome -- Chinese pavilion at the 1974 Spokane world's fair. (Now serves as roof of Dietrich Activity Center at Walla Walla Community College.)
Subdivision of the hexagonal face of a cuboctahedron, using Richter's method.
Subdivision of the hexagonal face of a cuboctahedron, using Richter's method.
-Taff
--
You received this message because you are subscribed to the "Geodesic Help" Google Group
--
--
To unsubscribe from this group, send email to GeodesicHelp...@googlegroups.com
Blair Wolfram
Nov 1, 2012, 4:16:56 PM11/1/12
to geodesichelp
Dave;
It's great to have you participating on this newsgroup!
Blair
It's great to have you participating on this newsgroup!
Blair
Adrian Rossiter
Nov 1, 2012, 4:35:46 PM11/1/12
to geodes...@googlegroups.com
Hi Dave
On Thu, 1 Nov 2012, Dave Kruschke wrote:
> Well, I remember some "rules" for creating Temcor Dome triangles and
> there was no mention of a n, n+1 "rule." This "rule," interesting if
> valid, is not necessary to derive the chord factors of the various
> Temcor Domes that I worked on. For example, some of the Temcor Domes
> don't have a five sided "pent" at the top but instead have a six sided
> "pent" and aren't connected with the geometry of the icosahedron. Do you
> think that this n, n+1 "rule" applies regardless of the initial
> spherical triangles as viewed from the top of the dome?
I have written a program which I believe produces all the isoceles
triangle models of this kind. I have checked a few to confirm that
they have the n+1 chord lengths as described by Gerry.
The frequency and number of sides around the axis are free variables,
and there are also another couple of continuous variables that
determine the final model
The program link and discusion are here
https://groups.google.com/forum/#!topic/geodesichelp/xpzbp5qMLXA
Here are a couple of example heptagonal models I posted
http://www.antiprism.com/misc/geo_temcor_ex01.png
http://www.antiprism.com/misc/geo_temcor_ex02.png
Adrian.
--
Adrian Rossiter
adr...@antiprism.com
http://antiprism.com/adrian
On Thu, 1 Nov 2012, Dave Kruschke wrote:
> Well, I remember some "rules" for creating Temcor Dome triangles and
> there was no mention of a n, n+1 "rule." This "rule," interesting if
> valid, is not necessary to derive the chord factors of the various
> Temcor Domes that I worked on. For example, some of the Temcor Domes
> don't have a five sided "pent" at the top but instead have a six sided
> "pent" and aren't connected with the geometry of the icosahedron. Do you
> think that this n, n+1 "rule" applies regardless of the initial
> spherical triangles as viewed from the top of the dome?
triangle models of this kind. I have checked a few to confirm that
they have the n+1 chord lengths as described by Gerry.
The frequency and number of sides around the axis are free variables,
and there are also another couple of continuous variables that
determine the final model
The program link and discusion are here
https://groups.google.com/forum/#!topic/geodesichelp/xpzbp5qMLXA
Here are a couple of example heptagonal models I posted
http://www.antiprism.com/misc/geo_temcor_ex01.png
http://www.antiprism.com/misc/geo_temcor_ex02.png
Adrian.
--
Adrian Rossiter
adr...@antiprism.com
http://antiprism.com/adrian
Dave Kruschke
Nov 1, 2012, 7:36:21 PM11/1/12
to geodes...@googlegroups.com
Thanks. I can see that I've really fallen behind in this area over the years. I've probably fallen behind in more than a few other areas as well...
From: Adrian Rossiter <adr...@antiprism.com>
To: "geodes...@googlegroups.com" <geodes...@googlegroups.com>
Sent: Thursday, November 1, 2012 3:35 PM
Subject: Re: Temcor domes, revisited
Dave Kruschke
Nov 1, 2012, 7:42:26 PM11/1/12
to geodes...@googlegroups.com
Thanks, Blair but I'm beginning to think that I have little to contribute here...
Otherwise, I still see an importance for some kind of tension or compression measures at the ground level of a dome.
Moreover, for years I've read enthusiastic statements about how a dome becomes stronger with higher frequencies. While this may be true for distributed loads, I don't think that this is true for a concentrated or "point" load. An example of this would be the common soap bubble...
Dave Kruschke
From: Blair Wolfram <thedo...@domeincorporated.com>
To: geodesichelp <geodes...@googlegroups.com>
Sent: Thursday, November 1, 2012 3:16 PM
Subject: Re: Temcor domes, revisited
TaffGoch
Nov 1, 2012, 8:08:58 PM11/1/12
to geodes...@googlegroups.com
Also, here, modeled in 3D:
Blair Wolfram
Nov 1, 2012, 8:47:28 PM11/1/12
to geodesichelp
The Big Outdoors People kept a copy of The Dome Cookbook under lock and key. As you live east of the twin cities, it seems likely you know Dennis Johnson, Dennis Kelly and Sheldon Wolfe? Their solution to the outward thrust of the dome, they called it splaying effect, was a steel tension cable around the base ring of struts, continuing up and over any door openings. Splaying was one of the four failures of the early TBOP domes, caused by the hub connector design which lag bolted into the end grain of the strut.
The failure of most buildings is a failure in tension, when a 2"x6 gets over loaded, snaps and pulls apart from itself. 3/4" steel conduit used in most 'doityerself' dome projects has better tensional strength than a 2"x6". (conduit has other weaknesses) When a hub connector on up to a 50' wood dome has multiple bolts with adequate spacing drilled through each end of a 2"x6 strut, a graded 2"x6 has sufficient strength in tension not to fail.
You can transfer this lateral force to the earth with concrete forming tubes imbedded with a steel channel footing plate.
The Dome Cookbook by David Kruschke had a huge impact on the developing dome world! The 4 strut 3v, and 6 strut 4v domes have significant advantages, and The Dome Cookbook is the primary reference.
Blair
The failure of most buildings is a failure in tension, when a 2"x6 gets over loaded, snaps and pulls apart from itself. 3/4" steel conduit used in most 'doityerself' dome projects has better tensional strength than a 2"x6". (conduit has other weaknesses) When a hub connector on up to a 50' wood dome has multiple bolts with adequate spacing drilled through each end of a 2"x6 strut, a graded 2"x6 has sufficient strength in tension not to fail.
You can transfer this lateral force to the earth with concrete forming tubes imbedded with a steel channel footing plate.
The Dome Cookbook by David Kruschke had a huge impact on the developing dome world! The 4 strut 3v, and 6 strut 4v domes have significant advantages, and The Dome Cookbook is the primary reference.
Blair
Adrian Rossiter
Nov 2, 2012, 9:07:42 AM11/2/12
to geodes...@googlegroups.com
Hi Taff
On Thu, 1 Nov 2012, TaffGoch wrote:
> The 7-fold radial symmetry dome you posted looks comparable to the geometry
> used for "The Land" greenhouse, at Epcot (a Temcor dome):
It would be interesting know the type of parameters and values
used for the physcal dome.
I haven't added any other parameter sets to the geo_temcor program
yet, but I have had a look at the effect of the existing parameters
by making some animations
When the second continuous variable is fixed, varying the first, it
is possible to see a stage where the model is most "even", which seems
to be the form used in physical domes. It may be the model with least
variance of triangle height, or perhaps least variance of triangle
apex angle. Perhaps it can be expressed as a simple formula in terms
of the parameter values.
These are the animations (size 4-8Mb)
Second continuous variable fixed at -0.7, 0, 05
http://www.antiprism.com/misc/anim_temc_p0_0.5.gif
http://www.antiprism.com/misc/anim_temc_p0_0.gif
http://www.antiprism.com/misc/anim_temc_p0_-0.7.gif
First continuous variable fixed at 0.5
http://www.antiprism.com/misc/anim_temc_0.5_p1.gif
On Thu, 1 Nov 2012, TaffGoch wrote:
> The 7-fold radial symmetry dome you posted looks comparable to the geometry
> used for "The Land" greenhouse, at Epcot (a Temcor dome):
used for the physcal dome.
I haven't added any other parameter sets to the geo_temcor program
yet, but I have had a look at the effect of the existing parameters
by making some animations
When the second continuous variable is fixed, varying the first, it
is possible to see a stage where the model is most "even", which seems
to be the form used in physical domes. It may be the model with least
variance of triangle height, or perhaps least variance of triangle
apex angle. Perhaps it can be expressed as a simple formula in terms
of the parameter values.
These are the animations (size 4-8Mb)
Second continuous variable fixed at -0.7, 0, 05
http://www.antiprism.com/misc/anim_temc_p0_0.5.gif
http://www.antiprism.com/misc/anim_temc_p0_0.gif
http://www.antiprism.com/misc/anim_temc_p0_-0.7.gif
First continuous variable fixed at 0.5
http://www.antiprism.com/misc/anim_temc_0.5_p1.gif
Gerry in Quebec
Nov 2, 2012, 9:53:57 AM11/2/12
to Geodesic Help Group
I have a copy of Dave's Dome Cookbook of Geodesic Geometry, first
edition published in 1972. A price sticker on the front cover -- $1.50
-- masks the original price of $2.00. This booklet is the most prized
possession in my little collection of books and pamphlets about domes.
I bought that copy several years back for US$10 from an engineer in
Texas named Hank Phillips. Hank had reviewed a copy of the booklet for
Dave, checking some of the numbers. I learned a lot about spherical
trig from that booklet -- and also from another really useful source,
Hugh Kenner's Geodesic Math. Recently on the Web I saw there was a
copy of Dave's Cookbook advertised for a price of something like $250.
Yep, a collector's item.
- Gerry in Quebec
On Nov 1, 8:47 pm, Blair Wolfram <thedome...@domeincorporated.com>
wrote:
> > *From:* Blair Wolfram <thedome...@domeincorporated.com>
> > *To:* geodesichelp <geodes...@googlegroups.com>
> > *Sent:* Thursday, November 1, 2012 3:16 PM
> > *Subject:* Re: Temcor domes, revisited
>
> > Thanks...
>
> > ------------------------------
> > *From:* Gerry Toomey <toomey.ge...@gmail.com>
> > *To:* geodesichelp <geodes...@googlegroups.com>
> > *Sent:* Thursday, November 1, 2012 2:06 PM
> > *Subject:* Temcor domes, revisited
edition published in 1972. A price sticker on the front cover -- $1.50
-- masks the original price of $2.00. This booklet is the most prized
possession in my little collection of books and pamphlets about domes.
I bought that copy several years back for US$10 from an engineer in
Texas named Hank Phillips. Hank had reviewed a copy of the booklet for
Dave, checking some of the numbers. I learned a lot about spherical
trig from that booklet -- and also from another really useful source,
Hugh Kenner's Geodesic Math. Recently on the Web I saw there was a
copy of Dave's Cookbook advertised for a price of something like $250.
Yep, a collector's item.
- Gerry in Quebec
On Nov 1, 8:47 pm, Blair Wolfram <thedome...@domeincorporated.com>
wrote:
> The Big Outdoors People kept a copy of The Dome Cookbook under lock and
> key. As you live east of the twin cities, it seems likely you know Dennis
> Johnson, Dennis Kelly and Sheldon Wolfe? Their solution to the outward
> thrust of the dome, they called it splaying effect, was a steel tension
> cable around the base ring of struts, continuing up and over any door
> openings. Splaying was one of the four failures of the early TBOP domes,
> caused by the hub connector design which lag bolted into the end grain of
> the strut.
>
> The failure of most buildings is a failure in tension, when a 2"x6 gets
> over loaded, snaps and pulls apart from itself. 3/4" steel conduit used in
> most 'doityerself' dome projects has better tensional strength than a
> 2"x6". (conduit has other weaknesses) When a hub connector on up to a 50'
> wood dome has multiple bolts with adequate spacing drilled through each end
> of a 2"x6 strut, a graded 2"x6 has sufficient strength in tension not to
> fail.
>
> You can transfer this lateral force to the earth with concrete forming
> tubes imbedded with a steel channel footing plate.
>
> The Dome Cookbook by David Kruschke had a huge impact on the developing
> dome world! The 4 strut 3v, and 6 strut 4v domes have significant
> advantages, and The Dome Cookbook is the primary reference.
>
> Blair
>
> key. As you live east of the twin cities, it seems likely you know Dennis
> Johnson, Dennis Kelly and Sheldon Wolfe? Their solution to the outward
> thrust of the dome, they called it splaying effect, was a steel tension
> cable around the base ring of struts, continuing up and over any door
> openings. Splaying was one of the four failures of the early TBOP domes,
> caused by the hub connector design which lag bolted into the end grain of
> the strut.
>
> The failure of most buildings is a failure in tension, when a 2"x6 gets
> over loaded, snaps and pulls apart from itself. 3/4" steel conduit used in
> most 'doityerself' dome projects has better tensional strength than a
> 2"x6". (conduit has other weaknesses) When a hub connector on up to a 50'
> wood dome has multiple bolts with adequate spacing drilled through each end
> of a 2"x6 strut, a graded 2"x6 has sufficient strength in tension not to
> fail.
>
> You can transfer this lateral force to the earth with concrete forming
> tubes imbedded with a steel channel footing plate.
>
> The Dome Cookbook by David Kruschke had a huge impact on the developing
> dome world! The 4 strut 3v, and 6 strut 4v domes have significant
> advantages, and The Dome Cookbook is the primary reference.
>
> Blair
>
> On Thu, Nov 1, 2012 at 6:42 PM, Dave Kruschke <theabunda...@yahoo.com>wrote:
>
>
>
>
>
> > Thanks, Blair but I'm beginning to think that I have little to contribute
> > here...
>
> > Otherwise, I still see an importance for some kind of tension or
> > compression measures at the ground level of a dome.
>
> > Moreover, for years I've read enthusiastic statements about how a dome
> > becomes stronger with higher frequencies. While this may be true for
> > distributed loads, I don't think that this is true for a concentrated or
> > "point" load. An example of this would be the common soap bubble...
>
> > Dave Kruschke
>
> > ------------------------------
>
>
>
>
>
> > Thanks, Blair but I'm beginning to think that I have little to contribute
> > here...
>
> > Otherwise, I still see an importance for some kind of tension or
> > compression measures at the ground level of a dome.
>
> > Moreover, for years I've read enthusiastic statements about how a dome
> > becomes stronger with higher frequencies. While this may be true for
> > distributed loads, I don't think that this is true for a concentrated or
> > "point" load. An example of this would be the common soap bubble...
>
> > Dave Kruschke
>
> > *From:* Blair Wolfram <thedome...@domeincorporated.com>
> > *To:* geodesichelp <geodes...@googlegroups.com>
> > *Sent:* Thursday, November 1, 2012 3:16 PM
> > *Subject:* Re: Temcor domes, revisited
>
> > Dave;
>
> > It's great to have you participating on this newsgroup!
>
> > Blair
>
> > On Thu, Nov 1, 2012 at 3:14 PM, Dave Kruschke <theabunda...@yahoo.com>wrote:
> > Dave;
>
> > It's great to have you participating on this newsgroup!
>
> > Blair
>
>
> > Thanks...
>
> > ------------------------------
> > *From:* Gerry Toomey <toomey.ge...@gmail.com>
> > *To:* geodesichelp <geodes...@googlegroups.com>
> > *Sent:* Thursday, November 1, 2012 2:06 PM
> > *Subject:* Temcor domes, revisited
>
> > Excluding skirt triangles, the number of different strut lengths equals
> > the number of rows of triangles plus 1. This is independent of the number
> > of triangles converging at the apex.
> > - Gerry
>
> > --
> > You received this message because you are subscribed to the "Geodesic
> > Help" Google Group
> > --
> > To unsubscribe from this group, send email to
> > GeodesicHelp...@googlegroups.com
> > --
> > To post to this group, send email to geodes...@googlegroups.com
> > --
> > For more options, visithttp://groups.google.com/group/geodesichelp?hl=en
> > Excluding skirt triangles, the number of different strut lengths equals
> > the number of rows of triangles plus 1. This is independent of the number
> > of triangles converging at the apex.
> > - Gerry
>
> > --
> > You received this message because you are subscribed to the "Geodesic
> > Help" Google Group
> > --
> > To unsubscribe from this group, send email to
> > GeodesicHelp...@googlegroups.com
> > --
> > To post to this group, send email to geodes...@googlegroups.com
> > --
>
> > --
> > You received this message because you are subscribed to the "Geodesic
> > Help" Google Group
> > --
> > To unsubscribe from this group, send email to
> > GeodesicHelp...@googlegroups.com
> > --
> > To post to this group, send email to geodes...@googlegroups.com
> > --
> > For more options, visithttp://groups.google.com/group/geodesichelp?hl=en
> > --
> > You received this message because you are subscribed to the "Geodesic
> > Help" Google Group
> > --
> > To unsubscribe from this group, send email to
> > GeodesicHelp...@googlegroups.com
> > --
> > To post to this group, send email to geodes...@googlegroups.com
> > --
>
> > --
> > You received this message because you are subscribed to the "Geodesic
> > Help" Google Group
> > --
> > To unsubscribe from this group, send email to
> > GeodesicHelp...@googlegroups.com
> > --
> > To post to this group, send email to geodes...@googlegroups.com
> > --
> > For more options, visithttp://groups.google.com/group/geodesichelp?hl=en
> > --
> > You received this message because you are subscribed to the "Geodesic
> > Help" Google Group
> > --
> > To unsubscribe from this group, send email to
> > GeodesicHelp...@googlegroups.com
> > --
> > To post to this group, send email to geodes...@googlegroups.com
> > --
Reply all
Reply to author
Forward
0 new messages