Groups keyboard shortcuts have been updated
Dismiss
See shortcuts

Universal connected

89 views
Skip to first unread message

Ashok Mathur

unread,
May 10, 2025, 11:20:27 AMMay 10
to geodes...@googlegroups.com

Dear DX
The photo  at the link shows a type of a connection in a hub that I have not seen before.
It can easily modified for a hexagonal hub.
Dies it qualify as a universal hub?

Charles Lasater

unread,
May 10, 2025, 12:58:58 PMMay 10
to geodes...@googlegroups.com
I am searching for affordable hinges. Thank you.

--
--
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

---
You received this message because you are subscribed to the Google Groups "Geodesic Help Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to geodesichelp...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/geodesichelp/CAD-a3Og7NHj%2BEJXHexbCFLgghjd%2BpuzDBhsXz5hBmHCWcCUHHw%40mail.gmail.com.

Dick Fischbeck

unread,
May 10, 2025, 12:59:51 PMMay 10
to geodes...@googlegroups.com
Nice. I like it.

Levente Likhanecz

unread,
May 10, 2025, 1:03:16 PMMay 10
to geodes...@googlegroups.com
maybe look for asymmetrical hinges, or manufacture them from flatbars DIY. 
(asymmetrical as one half longer than the other. and the longer to fix on the wood) 
image.png

Dx G

unread,
May 10, 2025, 1:12:27 PMMay 10
to Geodesic Help Group
Ashok,
  Good photo. Quite a good step forward in many ways.  I looked at that one last year and have had discussions with several hinge manufactures.  One issue that needs review is the stability of the axial angle(s) since hinges are designed for rotation in that very direction.  If used as shown, there is no bracing there to prevent axial movement, although it would be limited in an assembled dome.  A certain amount of wobble might be tolerated by the structure as long as that freedom of movement did not contribute to failure/inversion of one or more hubs and/or a related form of collapse.  However, there are multiple, reasonable ways to lock a given axial angle in such an assembly, even perhaps make those lock positions adjustable.   
  Since this approach nicely checks quite a few of the boxes, it is certainly deserving of further consideration, and continued development.  It also presents some interesting possibilities for a "folding" structure which benefits shipping and portability.
  Keep up the good work  :-)
Dx G

Aakash Shah

unread,
May 11, 2025, 1:57:27 PMMay 11
to geodes...@googlegroups.com
Hinges are bad idea. You can try, and learn.

Thanks and Regards.
Aakash Shah
Celebrate lifethat is the purpose of life.


--
--
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

---
You received this message because you are subscribed to the Google Groups "Geodesic Help Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to geodesichelp...@googlegroups.com.

Dx G

unread,
May 11, 2025, 2:48:10 PMMay 11
to Geodesic Help Group
Aakosh,
  Would you care to say why hinges are a bad idea?  It sounds like we would see something if we assembled a model.  What would we see ?

Just as importantly, are there ways to prevent that failure mode(s) and make it work?

thx Dx G

Ashok Mathur

unread,
May 11, 2025, 8:49:37 PMMay 11
to geodes...@googlegroups.com

Dear Dx G
A geodesic structure is both a machine and s structure at the sane time. It is always jitterbugging around an equilibrium. That equilibrium is disturbed by the passing wind also.

The universal hub should allow the struts to re-algain themselves in any direction as the equilibrium vector requires.

The hinge in the photo allows the strut to move freely in one plane .

But it fully restricts movement in other planes.
So that hinge is only partially good.

We need a ball and socket type joint to allow maximum flexi it of movement.
Sorry for not having figured out this earlier.
Regards

Ashok




Eric Marceau

unread,
May 11, 2025, 8:53:25 PMMay 11
to geodes...@googlegroups.com

As Aakash stated, hinges are a bad idea!

You need to have joint connectors that are "rigid".  The reason being, you need to "clamp" the geometric shape into a single configuration for stability.  Some connector configurations might allow minimal linear movement along the struts, but any rotary "freedom of movement" can lead to instability of the structure, which is why those are not used for larger structures where the overall structural weight will cause movements to an extent that the "rigid" connectors are intended to constrain or prevent altogether.

The idea of ball-joint connectors, as we often see on smaller domes using plastic materials, is not really one which is scalable for that very reason, hence the avoidance of those in the larger structures designed to operate in a gravity well.

Hope that clarifies the non-scalability of the hinge-type joint.


Eric

Paul Kranz

unread,
May 12, 2025, 7:40:45 AMMay 12
to geodes...@googlegroups.com
Eric: A triangle is a locked hinge. It doesn't matter if the connectors are rigid or not. The hinge is locked. The same applies to the 3rd dimension. I have built a few locked hinge geodesics and they appear to be just as rigid as the domes with rigid connectors.

Paul sends...

Aakash Shah

unread,
May 22, 2025, 2:17:54 AMMay 22
to geodes...@googlegroups.com
Hinges do not hold struts at certain angle, as hinge allows movement. So at the end you have to weld it. Which is not result into the perfect or even near to perfect angles.


Thanks and Regards.
Aakash Shah
Celebrate lifethat is the purpose of life.

Levente Likhanecz

unread,
May 22, 2025, 4:39:45 AMMay 22
to geodes...@googlegroups.com
hinges connecting different planes suppose to lock each others.
i can not describe it (like strut triangles are self locking). 
for e.g sarrus linkages:
sarrus2.gif
sarrus3.gif

Aakash Shah

unread,
May 22, 2025, 5:01:13 AMMay 22
to geodes...@googlegroups.com
No it does not. Weight of the struts and their assembly sequence will not support it. That's why I said that you try once and you will realise.
Thanks and Regards.
Aakash Shah
Celebrate lifethat is the purpose of life.

Paul Kranz

unread,
May 22, 2025, 7:04:21 AMMay 22
to geodes...@googlegroups.com
Aakash: Do you have to weld the loose-pin hinges of a triangle to make it rigid?

Paul sends...

Dx G

unread,
May 22, 2025, 11:27:07 AMMay 22
to Geodesic Help Group
I posted this under the hinge discussion on May 15 -
     consider, if you will, that there are already lots of examples of "hinges" allowing rotation, but then are locked in place.  Paul pointed this out earlier.
  So consider the example of almost any folding table.  The legs are on a hinge or pivot of some sort.  After opening, the "third" leg of the triangle locks the other two sides in place, preventing any further rotation between the table and leg.  If its something like a flexible cable, this third leg would prevent opening any further, if solid, could be used for both compression and tension.
  As I said in my comments earlier, there are likely over a dozen ways to "lock" a hinge, or even fit it with springs of sorts (like a torsion spring) to resist a given limit of force.  It does not look like the panels in the photo utilize a brace or lock for their hinges, but it would not take much to add such a feature.  Some of those folding locks they use on folding tables are typical examples of what you can buy off the shelf for lighter applications, but there are many such options.  The hinge concept offers many opportunities. 

-Dx G

Eric Marceau

unread,
May 22, 2025, 8:51:02 PMMay 22
to geodes...@googlegroups.com

As I mentioned earlier, on a "small" scale, where failure does not cause catastrophic results, hinges can be a solution although not, in my mind an ideal one, but workable on that small scale because the "pinned triangle" configuration has more force "leverage" because the "cone" of the joint is more "pointed".

As you go large scale, that joint "cone" is flatter and the hinge configuration becomes a liability, one that grows in terms of risk of negative consequences as the design configuration grows larger.

To support my stance, you need only look at all large-scale "space frame" designs.  These all make use of tetrahedral constructions (the "pinned" triangles which keep being mentioned) ... and yet ... you never see hinged joints.  Knowing that hinges on joints, at small scale, give advantages for assembly, surely, if there were no risks associated with such, we would see wide-scale application of those for the myriad space frame structures that have been erected.  But we don't!  

I suggest that having such a majority of designers/architects leaning away from hinges, for joints, is suggestive of important insights which, while not shared explicitly, beg to be learned upon further reflection and study.


Eric

Dx G

unread,
May 22, 2025, 9:46:44 PMMay 22
to geodes...@googlegroups.com
Well Eric, suspension bridges are pretty big, and their failure would be pretty catastrophic. Just like trusses, some of which are extremely large, all use interconnected structural elements to handle compression and tension, many of which would rotate if freed from the assembly.  So the real key is to design the assembly to handle the load, and as pointed out earlier, to accommodated earthquake, wind, temperature contraction/expansion and other forces that are better managed by planned movement rather than trying to make things lock in place so they "can't" move.   Actually, if one studies epic collapses in ancient and more recent history, it is clear there are consequences for a belief that the forces of nature can be defeated by us puny humans.  Hinges are not the only way to go, and might not even be the best, but they offer some interesting options

Dx G

You received this message because you are subscribed to a topic in the Google Groups "Geodesic Help Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/geodesichelp/_OP_4mtROqk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to geodesichelp...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/geodesichelp/872b60d4-de28-4821-b926-99b1e013b2d3%40gmail.com.

Aakash Shah

unread,
May 23, 2025, 4:53:16 AMMay 23
to geodes...@googlegroups.com
@Paul Kranz
When the load is applied on the opposite side of the dome's junction, the deflection will take place at this junction. So, yes, one has to make the hinge rigid to withstand the impact.
Have I clarified, or am I mistaken?

Thanks and Regards.
Aakash Shah

Robert Clark

unread,
May 23, 2025, 2:33:35 PMMay 23
to Geodesic Help Group
Hinged connectors would work jus fine in a geodesic dome.
The connectors I used on a 23 foot diameter dome are laser-cut from thin 2mm aluminum.
The arm joint can be bent by hand (10 degrees).  They are not much different from hinges. 
In fact, during assembly, the weight of a strut pulled down by gravity was enough to bend the angle of the "hinged" arm excessively.
Once fully assembled, the structure was absolutely rock-solid without any give at all.  I was surprised.
Compression and tension loads run in the direction along the length of the strut, converging at the central part of each connector.
The material of the connector just has to be able to resist crumpling. Assembly automatically bends the arm angles to the proper amount.
The screws are taking a lot of sheer force. That is often an overlooked part of the connector hub design.
This 23 foot diameter, 12 foot tall dome covered in the top half with plastic, easily weathered a winter of heavy snow and gale force winds.
My chickens will vouch for the safety and structural integrity of these near hinge-like connectors. These are not welded heavy gage steel.
This is why these connectors only cost $5.00 each instead of $20.00.

Screenshot 2025-05-23 140645.jpgScreenshot 2025-05-23 140856.jpgScreenshot 2025-05-23 140747.jpg

Dx G

unread,
May 23, 2025, 11:25:21 PMMay 23
to Geodesic Help Group
Robert,
 I've actually seen those hubs for sale, but only outside the USA.  Although they don't check all the boxes for the Universal Strut Connector, they certainly have some great assets for simplicity, ease of use, lightweight, and several more.
 So here is a question for you.  Suppose instead of cutting out the starfish pattern, you just put that pattern inside a circle, cut out the circle, cut a radial slit in the circle, and pulled it into a cone.  There we have the equivalent of one of Dick's random plates, but only the vertex tip. That approach does offer a few advantages. For one, the same cone can accept different numbers of struts, 4, 5, 6, even others. So you are cutting just one shape for use at a number of different hubs. It also has more strength for some settings than a connector with protrusions.  Maybe you could try a few, and install them in the dome to see how that works.  We may discover things we didn't know.  I expect to try it myself. 

Dx G

Robert Clark

unread,
May 24, 2025, 4:53:28 PMMay 24
to Geodesic Help Group
1780376_orig.jpgDx G,
Dx G,
What you are describing is a shallow cone connector.  The dome company "Growing Spaces" already does this.  Their connectors are basically hexagonal (and pentagonal) pieces of sheet metal dimpled by an arbor press into a shallow cone shape. The holes are drilled (or laser-cut) either before or after forming. There is only one hole per strut end, so there is a possibility for pinwheeling.  But, that also allows for just one type of hexagonal connector for an entire dome. 

There is a trade off between the star-shape connector and the cone-shape connector.  The star uses less material, but the laser-cut path is longer.  The cone shape is a simpler part to cut out with a shorter laser-cut path thereby saving laser cut time costs.  However, there is increased material and added shipping weight which both adds cost.  Also, the secondary process of pressing the metal into a cone adds cost. One other thing, the geometry of the axial angles at vertices will vary slightly depending where on the dome they are. However, the angle of the cone connectors are all the same.  Tightening down the bolts will warp/bend the sides of the cone to meet the strut, so this probably won't even be noticeable.

-Rob

Dx G

unread,
May 24, 2025, 6:28:10 PMMay 24
to Geodesic Help Group

Robert,
 Very good example for the discussion. Thanks for posting it. I agree with many of your points, but want to elaborate on a few of the issues.

  The single point strut connection does seem like a design weakness, but likely the company would claim they have been making these for years and not a problem. This may rely on their glazing, as they tend to use triangular multiwall polycarbonate panels, which confer considerable stiffness to the shell.  I'm not sure that single point connection would do so well if the frame was only covered with poly film, especially if it was subjected to a big wind driven snow load.
  I did actually visit someone I know who has one of these domes.  Somehow I don't recall seeing these hubs inside. Apart from that, I don't know if the company stopped doing this some time ago, but their dome had aluminum strips along each strut, bent to the dihedral angle for each pair of mating triangular panels.  These were attached to the outside on top of the polycarbonate panels.  Likely this provides added strength and a good weather tight seal, so maybe they decided it was overkill and added unnecessary expense, or perhaps leaked due to differential expansion/contraction.
  As to the hub, cutting hexes is pretty efficient with near zero waste, although not as good for the pents.  Seems to me once the cone is formed, the perimeter would be irrelevant, since only the axial angle and holes determine the strut connection. I agree, the issue of variable axial angles around any given hub would need review.  Sure, there is a benefit to using less material for the finished product to reduce weight, but if doing so just generates waste, my preference would be to let the material go to the product rather than to waste as long as it adds strength or provides some other advantage without downsides, like unacceptably higher shipping cost or weight.
  What I like about an actual cone with a slit, like Dick's Randome, is that all you need is one "blank" and then the cone can be adjusted to various axial angles by adjusting the overlap at the radial slit. Within given face angle limits, holes can be punched so the "blank" can be used for either a hex or pent. The punch operation they use to form the cone is probably pretty fast, but once its punched, not likely it can be used for anything else.
  So I won't make any claims that what I'm proposing is actually better than their arrangement, but in terms of the feature list for the Universal Strut Connector (USConn), a cone with a slit (or fold, or what works) provides many more options for various dome hubs, rather than specific manufacture like they use. It comes closer to universal, since once made, a given hub can be used in a 4 frequency octahedral just as it can for a 2 frequency Class 2 icosahedral...and almost anything, even down on the ground line where it needs a flat side. Likely it is also easier for people to make the conical hub with common equipment, rather than requiring machines one would only find in a machine shop or factory.

Dx G

Ashok Mathur

unread,
May 24, 2025, 8:53:53 PMMay 24
to geodes...@googlegroups.com
For a 3 v dome a dimple of 9 degrees works very well. About a thousand domes gave been built using such cones.
But as you go to 4v and higher, a single angle distorts the ideal very much such that they are not safe any more.

Regards

Ashok




Dx G

unread,
May 24, 2025, 9:38:16 PMMay 24
to geodes...@googlegroups.com
A thousand domes Ashok?  That's good news. Perhaps you can point us to a few or do so next time you come across one or more.  There are different ways to implement a cone hub, and I'd be interested in seeing that...like the one Robert showed us.

Dx G

You received this message because you are subscribed to a topic in the Google Groups "Geodesic Help Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/geodesichelp/_OP_4mtROqk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to geodesichelp...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/geodesichelp/CAD-a3Oikb8hoq_Uks9kYYy78vsS0_yJXygh0uWhQeR9toREQ0A%40mail.gmail.com.

Ashok Mathur

unread,
May 24, 2025, 9:45:36 PMMay 24
to geodes...@googlegroups.com
Dear Dx G 
It sounds incredible, but it is true.
About 20 years ago there was a major earthquake in an area called Latur in Maharashtra and most houses were flattened.
These 3v dies were built on a mass scale to rehabilate the victims.
The domes are still being used as homes with almost no repair needed so far.
I will post some pictures of a family living in one such dome.
I will post some pictures of such dimple hubs also.
There is a simple archimedes screw down press that houses dies and a hole punching system.

Regards

Ashok




Ashok Mathur

unread,
May 25, 2025, 1:57:06 AMMay 25
to geodes...@googlegroups.com
Dear Dx G
Here are two photos of a simple 3 V hub with a 9 degree deficit. Hubs of this sort were used to build the over 1,000 domes at Latur.
In the process of making a headlight for a scooter or a motorbike, circular holes are made in ms sheets. These hubs are recycled headlight holes.
The ridges in them are formed by the  archimedes screw set of dies.
There is a third photo showing me teaching (I am not visible in the photograph)  architecture students near Baroda how to build 3v domes using such hubs.

Regards

Ashok



unnamed-2.jpg
unnamed-3.jpg
Hex Hub.jpg

Ashok Mathur

unread,
May 25, 2025, 2:00:02 AMMay 25
to geodes...@googlegroups.com
Dear Dx G
Here are some photos and a short video of a dome built in Latur.
These photos are a year old and the gentleman in front of the dome is one of the builders of the domes.
I think I have shared this in past also

Regards

Ashok



PHOTO-2022-06-08-17-40-59.jpg
PHOTO-2022-06-08-17-41-00_1.jpg
PHOTO-2022-06-08-17-40-59_2.jpg
PHOTO-2022-06-08-17-40-58.jpg
PHOTO-2022-06-08-17-40-59_1.jpg
PHOTO-2022-06-08-17-41-00.jpg
VIDEO-2022-06-08-17-41-00.mp4

Levente Likhanecz

unread,
May 25, 2025, 4:58:08 AMMay 25
to geodes...@googlegroups.com
on your "dish" hubs the double bolt setup help to avoid pin wheeling.
(i suppose, pin wheeling is something like on charles lasater's wood dome, the twisted hub:)
pin wheeling.png

Robert Clark

unread,
May 25, 2025, 9:44:05 AMMay 25
to Geodesic Help Group
🌀 The Universal Connector Manifesto: Toward Modularity in Dome Design

This began as a design manifesto for myself — a way to clarify and document the principles I believe can guide the development of a near-universal dome connector system. I’m sharing it here in the hope that others in the group may find resonance, challenge it, build on it, or even help refine it further. If it sparks useful debate or inspires better design thinking, it’s done its job.


🧠 Reframing the Problem: It’s a System, Not a Vertex

Too often, design discussions focus on a single hub in isolation — stress tests, failure modes, bending loads on one cantilevered arm. But geodesic domes don’t behave that way. They are holistic systems, and their strength emerges from the synergy of triangular geometry and distributed force.

In a complete structure, many of the local stress concentrations we worry about — particularly bending moments on star-style connectors — become nearly negligible. Once assembled, the dome self-stabilizes. We need to shift our thinking from isolated nodes to network-aware components.

🧩 The Real Goal: Modularity Across Variation

Domes are modular, but their geometry isn’t uniform. Vertex angles shift, strut lengths vary slightly, and material properties introduce real-world tolerances. The goal isn’t one magic connector — it’s a flexible connector family that reduces part variation while enabling broad compatibility.

Picture this:

  • A single hub core with interchangeable inserts for different tube diameters

  • Elastic articulation zones to absorb ±5° angle variation across different frequencies

  • Connectors that accept PVC, EMT, or wood, without redesign

  • Twist-lock or snap-fit mechanisms for rapid, tool-free assembly

  • Color-coded or indexed guides for intuitive layout — even in field conditions

We’re not chasing symmetry — we’re designing a system that embraces variation and absorbs it with elegance.

⚙ Manufacturing: The Missing Conversation

To make this real, we must embrace manufacturing as a core design axis — not an afterthought. The best connector in the world is useless if it’s impractical to produce or scale.

We need to be talking about:

  • Injection-molded plastics like glass-filled nylon, ABS, HDPE, polypropylene — and their UV resistance, impact strength, and long-term creep

  • Metals like 6061-T6 aluminum (machinable and strong), 5052-H32 (formable and weldable), or stainless steel (for corrosion resistance)

  • Hybrid fabrication strategies: 3D printing for rapid prototyping or open-source distribution; CNC and molding for high-volume production

  • Tolerances that reflect reality — backyard builders don’t have aerospace tools. Let’s optimize geometry for forgiving fits and field-adjustability.

Let’s stop chasing unattainable precision and start designing for repeatable, manufacturable success.

🧪 Structural Flexibility: Passive Tolerance by Design

Instead of enforcing rigidity, we can design in compliance. Strategic flex, slight give, or angular drift tolerance within the connector itself — these can eliminate the need for hundreds of highly specific SKUs.

This could include:

  • Catenoid necks or notched hinges that allow for micro-articulation

  • Elastomeric inserts or collars that adapt to small angle and length differences

  • Plug systems with indexed locking flanges, allowing slight rotation and lock-in

With the right form language and materials, even “slop” becomes a smart feature — not a flaw.

🛠 Assembly Experience: Usability is Structure

A dome that’s frustrating to assemble is structurally weaker — not because of materials, but because the user interface failed. Ergonomics matter. Assembly flow matters.

So we should be asking:

  • Can it be assembled without tools?

  • Does it guide correct orientation without a manual?

  • Are the parts repeatable and intuitive, even for first-time builders?

  • Can it be easily disassembled and reused, making it suitable for education, events, or temporary structures?

The user experience is part of the system integrity — not separate from it.

💡 Context-Specific Load Design

A universal system must be scalable, not overbuilt. A 12-foot greenhouse in California does not face the same demands as a snow-bearing structure in New England. The system must adapt — through part thickness, material choice, and connection retention — to contextual structural loads.

Different dome types prioritize different things:

  • Backyard DIY kits → cost and speed

  • Educational tools → clarity and reusability

  • Off-grid shelters → resilience and repairability

  • High-frequency domes → angular precision and load-bearing efficiency

The universal connector system should be context-aware, not one-size-fits-all.

🧭 The Path Forward

We may never reach a perfect, universal connector. But we can pursue the ideal of universality — a small, elegant family of parts that flexibly bridges complexity, adapts to variation, and empowers builders at every level.

Let’s align around these principles:

  • Think like a system, not a part

  • Embrace geometry and variation, don’t fight it

  • Prioritize manufacturability and cost-efficiency

  • Optimize for intuitive, ergonomic assembly

  • Design connectors that flex, shift, and adapt

  • Build modular families, not monolithic solutions

  • Consider your builder — every time, every use case

  • Share test data and failures, not just polished prototypes

This is more than engineering. It’s about making domes more accessible, scalable, and delightful for everyone who dares to build one.

Not a perfect part —
but the perfect path.

-Rob

Dx G

unread,
May 25, 2025, 12:32:03 PMMay 25
to Geodesic Help Group
Ashok,
 Yes, I can see you indeed did share this in the past.

So I am quite curious. In some situations, there is no good means to get any traction with the idea of using domes. In the minds of too many people, the idea is dead-on-arrival, and won't even be considered.  However, what you illustrate is a clear contradiction and looks like a nice success. So I'd be interested in the back story.

-How did the idea of using domes for this originate?

-How did this idea to use the headlight plates come about? Someone, group, organization, company, government agency - had to be connecting the dots and putting this together.  It was no accident, there was thought and creativity at work here. Quite a great example of using "discard" material for a productive purpose few people could imaging.  One has to see the invisible to do the impossible. 

-Someone had to arrange, and pay for, angle iron to be obtained, cut, holes drilled/punched, color code and deliver.  Likewise for the bolts/fasteners, exterior coating materials and application tools.

-Then there had to be some technical and logistic project support to put all this together, get all the parts to the site and train people.

There are often lots of analytics after a disaster to determine what happened and why.  I have the same interest in a success.

Ashok Mathur

unread,
May 25, 2025, 8:52:08 PMMay 25
to geodes...@googlegroups.com
Dear Dx G
That person was Dr Kalbagh , s chemist who devoted all his life working for a MNC and after retirement wanted to teach science to rural India.
He choose the worst/ economically most backward area of India and set up an institution he named Vigyan Ashram at Pabst, dist Pune, Maharashtra.
The dome are his and his teams work.
The work continues with lot of success.

Regards

Ashok




Dx G

unread,
May 25, 2025, 11:43:44 PMMay 25
to geodes...@googlegroups.com
Ashok,
  That's quite interesting.  If appropriate, perhaps you could invite him and colleagues to join in any our group discussions that might be of interest.  Seems to me they sure do have a track record with domes, and likely have learned valuable things that might enlighten us all.

Dx G

Ashok Mathur

unread,
May 26, 2025, 1:27:31 AMMay 26
to geodes...@googlegroups.com
Dear Dx G
They have given up domes.
For what it's worth, I am what is available.

Regards

Ashok




Dx G

unread,
May 26, 2025, 10:22:09 AMMay 26
to geodes...@googlegroups.com
Ashok,
 When you say "given up", do you mean they have completed their work on domes and moved on to other projects, or that they ran into roadblocks or serious issues that prompted them to terminate their work with domes? 
  Additionally, what was your role in the dome project or if only an observer, in what capacity did you serve in that role?

Dx G

Ashok Mathur

unread,
May 26, 2025, 12:21:48 PMMay 26
to geodes...@googlegroups.com
Dear Dx G
The Latur earthquake took place in Sept 1993.
My first contact with Vigyan Ashram was in early 2000 as best as I can remember.
I wanted to try out paper Crete technology there.By the time I started to teach there, the people with theoretical knowledge of geodesic domes were gone.
But they had trained excellent students who have been all sorts of practical domes all over India.
The students knew only the formulas for lengths of a 3v dome with uneven truncation.The dies for the hubs were simple and intact. So they could build 3v domes.
Vigyan Ashram moved on to other rural technologies.
I taught one ex-student how to build a 4v dome which knowledge he used to build a planetarium space .



Regards

Ashok




Paul Kranz

unread,
May 26, 2025, 12:34:17 PMMay 26
to geodes...@googlegroups.com
Have any of you heard of the late Vyom Akhil?

Paul sends...



--
Very high regards,
 
Paul C. Kranz, LMFT
Kranz & Associates, LLC
Reply all
Reply to author
Forward
0 new messages