3D Printed Connector Kit

43 views
Skip to first unread message

Robert Clark

unread,
Jul 12, 2026, 7:08:41 PM (5 days ago) Jul 12
to Geodesic Help Group

This box of connectors is just sitting here waiting for my lazy ass to go out and purchase some 3/4" EMT conduit to build a 16' test dome.

During this past winter, I printed these out of TPU 64D filament. It took 10 days and almost three 1kg rolls of filament on my one Bambu Lab A1 printer. I used tear away brim ears to prevent curling and lifting up.

The TPU 64D is incredible tough, like a hard rubber. Almost indestructible and will return to its shape after being bent. Weather and UV resistant. You will not break one of these.

This is my attempt at a semi-universal connector. There are 3 shapes here. 5-star, 6-star, and 4-star base. They have a designed in 10 degree axial angle. These connectors can be used on a wide range of geodesic dome types and frequencies.

I'm posting all this because it might force me to get out and actually build the dome.

-Robert


Screenshot 2026-07-12 190045.pngScreenshot 2026-07-12 190257.png

Robert Clark

unread,
Jul 12, 2026, 7:11:21 PM (5 days ago) Jul 12
to Geodesic Help Group
I wanted to include an image showing the iteration of prototypes. I think it's interesting to see the process and not just the final result.

-Robert

Screenshot 2026-07-12 190500.pngScreenshot 2026-07-12 190701.png

Levente Likhanecz

unread,
Jul 13, 2026, 4:00:17 AM (4 days ago) Jul 13
to geodes...@googlegroups.com
dear Clark,
if your connector is solid mass of material, not some lattice with skin, then the salt method remelt can increase the strength by making layer adhesion like traditional casting.
or "lost wax" casting with very lite infill. you compress the sand on it, apply heat and drain the melted plastic, then cast it metal.
https://youtu.be/nRLJ4ylGTFc

--
--
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/8eaf874d-4d17-4f89-8abe-659dc0d5d629n%40googlegroups.com.

Robert Clark

unread,
Jul 13, 2026, 9:59:07 AM (4 days ago) Jul 13
to Geodesic Help Group
Levente,

Thank you. I watched the video. The PETG part seemed to come out best. It is a very labor intensive method, good for doing a couple of parts at a time. A kit of 60 or more dome connectors would not be practical. I don't know if TPU would work as well with this method. Also, I don't think there would be much benefit because the layer lines of TPU are already completely fused together and cannot be separated like PLA parts. Very fascinating reheating method though! I would considerate for a specialty PETG part that I want to have a frosted glass look. Thanks again.

It takes on average about 4 hours of 3D printing per connector. It costs about $1.50 or less in filament per connector.

-Robert

Screenshot 2026-07-13 093645.png

Dx G

unread,
Jul 13, 2026, 12:38:05 PM (4 days ago) Jul 13
to Geodesic Help Group
Robert,
  Interesting choice of parts.  I get that you referred to these as "semi" universal, so a few comments regarding "universal".

1) Assume the "arms" are flexible. If so, that would address variable axial angles to some extent.

2) Perhaps you can consider a tapered-flats end connector, rather like a step bit.  This would allow use of different inside diameters of tubing or pipe for a given arm, rather than accepting only one ID.

3) Also consider making these as separate arms that could be stacked at the hub, perhaps using a bolt. 
a) That would prevent getting handcuffed to different parts for 4, 5, 6 or more struts.   Unlike what you show here, this allows one part to work for different strut counts.
b) Likewise stackable arms works for different face angles, rather than being fixed like the parts shown.  In particular, useful when a given hub has to accommodate several different face angles at a given hub connection point.

I've got a few more ideas on these, but that's a start.  The original list I posted for the USconn is a pretty good yardstick, pending any downstream additions, corrections, deletions.

I always see the Einstein tee shirt that says, E=MC^2, not just a good idea.  Likewise, I'd say, Universal, not just a good idea.  :-)

Dx G

Robert Clark

unread,
Jul 13, 2026, 4:13:01 PM (4 days ago) Jul 13
to Geodesic Help Group
Dx G,

Thank you for your comment and insight. I'll try to address a couple of your bullet points.

2) Stepped end to hub arms. Yes, that might allow different size tubes. Right now, it is an optimized fit for 3/4" EMT, the most common type of EMT conduit. I could create a step down for 1/2" tubing. It would mean extending the ends and using more material and adding time to print. That extension may never be used. I think it is better to have two separate designs, one for 3/4" and one for 1/2". These are digital files. Just choose the one to match with the material size you will be using. Sometimes the problem with trying to be completely universal is you compromise design, cost, and usefulness by being one thing for everything. I compromised on "universal" by settling on one strut material, but allowing for many possible dome configurations.

3) Separating into multiple parts and adding additional fasteners. This would go against what I was trying to achieve by simplifying the hub connector to just one piece that could be used on many different angled vertices. Again, I'm accepting compromise on "universal" by having 4-arm, 5-arm, and 6-arm connectors. Perhaps a design that satisfy what you are suggesting would be a hub that is a circular metal disk with slots radiating out from its center to create 4, 5, and 6 star patterns. One single TPU arm could be designed that would snap-fit into the slots to create the various configurations of 4, 5, and 6 star hubs. The slightly flexible hard rubber arms could adjust for varying axial and radial angles. Your unique part count would now be down to two - the central metal disk, and the TPU 64D arms.

These designs are intended for modest size domes, not house structures.
The best multi-tools can do a few things very well, but not everything.

-Robert

Dx G

unread,
Jul 13, 2026, 5:19:39 PM (4 days ago) Jul 13
to Geodesic Help Group
Robert,
 Thanks for comments.  Certainly you have some valid positions on this, and are, of course, entitled to operate accordingly.  However, let me follow up with an alternative view.  I'm trying to look at this from both the viewpoint of the producer/manufacturer as well as the user/customer/market.

  For one thing, having all these different digital files and parts does create a level of complexity, albeit not a lot, that you don't have when the entire system is based on just one.  That inventory of one does present a significant reduction in complexity and offers greater economy of scale in part production, order fulfillment, etc. down the road, even if not initially.

  So picture "the part" as just the arms, which can be assembled into "hubs".  

Sure, if the arm had the stepped connector (or otherwise accomodate different size strut profiles) the same part could be used for different struts and domes.  However, it could even accomodate domes made from several strut sizes and types, where the user may have a mixture of materials on hand (or otherwise available) but not enough of any one size to make a complete dome.  This lends itself to making good use of suplus or "waste" material, which at times may come in small lots.

In addition, all the parts could be reused later for different hub or dome configurations if needed.

Then, of course the one part can be grouped to accomodate hubs for any number (within limits) of strut and face angle configurations.

So I see some value in basically making a strut "end" that will enable connecting struts of various kinds into a hub configuration, bringing it closer to a universal connector. That is why I called the thing a USconn (Universal Strut Connector) rather than a hub. IMHO, your expertise, effort time/effort would serve you and the dome community better putting out something like that.  Consider the option, and possible value, in your deliberations.

In any event, I applaud your efforts, either way, and hope you have some success.

Dx G

Robert Clark

unread,
Jul 13, 2026, 8:28:26 PM (4 days ago) Jul 13
to Geodesic Help Group
Dx G,

Here is another concept very close to what you are talking about, I think. I created this about 3 years ago. I might have even posted it to the group at that time. Here is a link to a video that shows how it would work. The arms were intended to flex slightly. Again, it was just a rough concept. The round hub is made of two of the same identical molded halves that screw and lock together like an Oreo cookie. I did have one prototype 3D printed for me out of MJF material. It was a little expensive. That's as far as I've taken this design.


I would be interested in seeing sketches or models of your USconn to spark some ideas or discussions.

-Rob

Screenshot 2026-07-13 202725.png

Dx G

unread,
Jul 15, 2026, 12:56:33 PM (2 days ago) Jul 15
to Geodesic Help Group
Robert,
 Thanks for putting that up.  Its an interesting concept and offers some much needed flexibility for different domes.  Thing is, I'm a simplicity snob.   I see a lot of perfectly good hardware like this in the patent literature.  They would all work, and all have various assets, just like the hub you illustrated.  The issue one runs into is cost (as you pointed out), ease of implementation and thus demand from  users, and likelihood of volume sales.

  It would be nice to live in a world where money didn't drive things. That's not to say that money has driven all things that are good, but its an element one can't totally ignore.  The problem with all these hubs and dome hardware I see is that they would all require custom tooling, molds, machining and dedicated production...just for the dome application. Certainly 3-D printing has helped enormously on one-off (or low volume) modeling to at least get a concept off the screen and into the physical world where it can be structurally, or at least mechanically evaluated.  In the past, one could not easily get even that far.

  So when a new part, system, etc. develops, there is a circular problem. If you don't have the money to gamble (and yes, risk total loss), one cannot ramp up volume to get the economy of scale to get the price down and generate the revenue. Likewise, if you don't get the product out there, there may be no way to get the money, unless of course the pitch is good and investors get interested (and hopefully they don't lose *their* money). 
    Not to say things like you illustrate here are worthless. They may find a market, or, if nothing else, contribute to a later version or adaptation that does make it.  Imagination and creativity often demand a workspace that is not limited by popularity and profit potential. If it did, there are a lot of things we use every day that would never have seen the light of day.

  That's why one of the fundamental precepts of the USconn is to make use of elements that are already made and sold in large volume, typically for some existing established purpose or industry (item 7 in the USconn list). We do this when we use PVC pipe or EMT tubing for struts.  So, the idea of the USconn is to apply that same approach to *connecting* those struts.  I've been chipping away at this for some time, and getting closer, but holding off putting anything out there at this point.  It's not so much analysis paralysis as it is shooting for a certain level of functionality.  A bit like a little kid on a long trip asking if we are there yet  :-).

  In any case, I hope your work will continue.  As they say in the R&D and invention world, the reach should exceed the grasp.

Dx G

Robert Clark

unread,
Jul 15, 2026, 8:06:56 PM (2 days ago) Jul 15
to Geodesic Help Group
Dx G,

What is your target price per hub connector? I mean, how much before you decide the cost of materials or time for labor is too high? I have given myself an arbitrary limit of $5. Ideally, I want a complete hub connector that is under $3.

-Robert

Ashok Mathur

unread,
Jul 16, 2026, 10:32:42 AM (yesterday) Jul 16
to geodes...@googlegroups.com
Sometime back I had explored making multiple parts from 3d printed molds using two part resin and hardner.,it seems that process would lower the cost without compromising quality .


Regards

Ashok



--
--
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.

Robert Clark

unread,
Jul 16, 2026, 11:00:59 AM (yesterday) Jul 16
to Geodesic Help Group
Ashok,

I had thought about doing exactly the same thing! I went as far as purchasing molding supplies last February. Unfortunately, they are still sitting in a box. I got caught up with the 3D printed Synergy Sphere and other design ideas have taken a backseat. I'm easily sidetracked and jump from idea to idea. I have a hard time to focus! 🤣

-Robert

IMG_2540.jpeg

Ashok Mathur

unread,
Jul 16, 2026, 11:03:30 AM (yesterday) Jul 16
to geodes...@googlegroups.com
As a first guess about 20 copies can be printed  by this method. 


Regards

Ashok



Robert Clark

unread,
Jul 16, 2026, 11:37:42 AM (yesterday) Jul 16
to Geodesic Help Group
Dx G,

You said: "I've been chipping away at this for some time, and getting closer, but holding off putting anything out there at this point.  It's not so much analysis paralysis as it is shooting for a certain level of functionality."

I encourage you to just show what ideas and progress you have made so far, even if not quite there yet. Group discussion can open your eyes and spark new ideas that you might not have considered. Sometimes our minds get locked onto just one path. As an example, your comments have made me reconsider the two part design I had developed a few years ago, but hat put aside.

-Robert

Charles Lasater

unread,
Jul 16, 2026, 4:33:07 PM (23 hours ago) Jul 16
to geodes...@googlegroups.com
Robert and DGx,
Your practical arguments are exactly what drove me to single-strut rotegrity. Capture the form as simply as possible and reinforce it as needed. Just two wye connectors. Not universal but minimal. 


--
--
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,
Jul 16, 2026, 5:54:16 PM (22 hours ago) Jul 16
to geodes...@googlegroups.com
Robert,
With respect to the connector pricing, I'd say that one is down the
road for now. A lot goes into setting pricing, since its not just
your material cost, and one has to consider all the other inputs -
production, advertising/marketing, order fulfilment, packaging,
shipping, insurance and legal (hmmm, tariffs), etc. long list.
Somewhere there needs to be some profit to make the whole thing viable
financially unless its some kind of a charitable or nonprofit
approach. Part of my thinking is that item nr. 7, as if one can use
materials that are already in circulation, at high volume, often at
least you are starting with attractively low prices there. Clearly,
the connector price would need to be competitive with alternative
choices, although there might be room to charge more if it provides
value that competitors don't offer...like having that
one-connector-does-it-all capability.

Regarding the prompt release of information in the early development
phase, in general, I agree with you and this is certainly a good group
to brainstorm. I've done quite a bit of this on the forum already on
selected topics, and enjoy the generous (and useful) reciprocation.
I'm holding off for now on the USconn until testing tells me whether I
am, or am not, barking up the right/wrong tree. This can help
reduce/eliminate what HL Menken alluded to back in the 1920's...for
every complicated problem there is a simple idea that won't work. So
stay tuned, its still like I said earlier, like the little kid on a
trip asking...are we there yet. However, you will notice that some
time ago I did post the spec list for the USconn (which is still
subject to additions, deletions and changes), so hopefully that does
have some of that thinking going about the how-to among the list
readers.

Dx G
> --
> --
> 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 a topic in the Google Groups "Geodesic Help Group" group.
> To unsubscribe from this topic, visit https://groups.google.com/d/topic/geodesichelp/ITxUijdRQ_w/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/5d1892f1-e5ad-4a91-ab7f-65ba7eed697fn%40googlegroups.com.

Dx G

unread,
Jul 16, 2026, 5:56:48 PM (22 hours ago) Jul 16
to geodes...@googlegroups.com
Charles,
 Thanks for bringing that up. Yes, the simplicity snob in me certainly approves of single strut.  Some time ago, I built a small model of the one Taff posted, as I wanted to see how one of these went together. 
   One tradeoff one accepts with a single strut approach is you lose some of the flexibility you have with domes which benefit from going to a higher frequency to prevent undesirably long struts.
  Thing is, that single strut still needs to be connected to two others.  This is one of the advantages of the USconn, as it will accommodate a 3 way connection. However, obviously, if all one ever makes are single strut polyhedra which all use the same hub, there is little advantage to using a connector that is also useful for 4, 5, 6, or other nodes.  Also depends if those 3 way hubs all have the same face, axial, dihedral angles, etc.  The other wild card is whether or not the USconn actually proves to be a better choice if it does use components you can buy at any store, which provides enhanced availability, and then cost advantages, if nothing special needs to be manufactured for the connector.  What's nice about some of the rotegrities is the connector for those is sometimes simply a hole that goes through the 3 struts, and that's it.

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/ITxUijdRQ_w/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/CADgTqM5db7z2x6u%2BCehTC3Pd20JPybgSJCu-TvPsp%2B%3D3PNW%2BmA%40mail.gmail.com.

Robert Clark

unread,
Jul 16, 2026, 8:30:40 PM (19 hours ago) Jul 16
to Geodesic Help Group
Dx G,
Thanks for your perspective. To clarify on the pricing, the $3 to $5 figure I mentioned is strictly my estimated material and production cost, not the intended retail price. I completely agree with you that once you factor in marketing, packaging, fulfillment, and overhead, retailing at that price point would make it impossible to turn a profit.
It also seems we might be envisioning different intended sizes for these domes. Even a universal connector has to accept certain scaling constraints. The heavy-duty metal connectors required to build Gerry's 50-foot dome house in Canada simply aren't practical for a 16-foot backyard patio arbor, or a 6-foot net-covered dome built to protect blueberry bushes. A 50-foot dome is never going to be a DIY project.
I am intentionally aiming at a completely different tier: the novice user building a modest structure, such as a 20-foot chicken enclosure. Think of the Zip-Tie Dome concept, but with a more finished look and an easier assembly process.
My target consumer thinks domes are cool but knows almost nothing about them. They only own basic household tools like a hand drill and a circular saw—no drill presses or compound miter saws required. Ultimately, my goal is mass-produced, injection-molded parts that a customer can just pull out of a bag, snap together, and secure to standard wood struts using simple screws.
-Robert

Robert Clark

unread,
Jul 16, 2026, 8:44:52 PM (19 hours ago) Jul 16
to Geodesic Help Group
Dx G,

In the meantime, for the members here, I will continue tossing my unfinished design concepts against the proverbial wall to see what sticks. I welcome feedback to improve my designs and open myself to different perspectives.

-Robert

Reply all
Reply to author
Forward
0 new messages