Spline Loft Sculpture

ZooHead

Well-known member
I got all the ribs for C1d made and I decided not to do separate PDFs.
I thought scale matching each one would be a lot of extra work.

I'm still thinking it's best to make one PDF with everything on it when it's all complete.

C1d rib and feet.jpg
 

ZooHead

Well-known member
Placing the splines used in the Sweep Creator with Component Snapping.
The Sweep Creator is used with the Boolean operation to cut the holes for the carbon rods.

Tip for Component Snapping: In a very cluttered scene you will snap to anything in the
scene under the cursor, even if it's not visible. That includes stuff behind the cursor.
To avoid this, work on objects in isolation. Move only what you need to a separate file.
I think you can also do the same thing with a separate Take.

C1e sweepspline.jpg
 

ZooHead

Well-known member
I use the Split Command to make a separate object of the first rib/end piece.

In Edge Mode, select edges and use the Cover Command,
Transform and Weld Tool to build the foot.

C1e rib1 feet.jpg
 

ZooHead

Well-known member
I've started working on the first intersecting element C2a.
Using the Slicer Script I was able to create offset ribs like interlocking fingers.

C1d C2a intersect.jpg
 

ZooHead

Well-known member
It looks like I may need to edit the foot of some of the
ribs in C1d, and maybe move the carbon rod a bit.

C1d C2a intersect02.jpg
 

Helmut

Well-known member
* @GS089 / Gerald (?) and @ZooHead
* My honest (and brutal) evaluation of the current solution to the curved elements (C1d, C2a, ...) is:
* Discard and start from square one.
* You (plural, as opposed to the Shakesperean thou / singular) seem to have ignored aspects of assembly, of interpenetrating sails and of finetuning (filling / planing / sanding in post assembly) by concentrating on isolated details and ignoring all else.
* At best and IMHO, constructing the tricky parts of the sculpture on the present concepts would be extremely time consuming and quite expensive.
* The longitudinal carbon rods (again, IMHO) provide no stability as the ensuing stack can still be twisted until foam gaps have been inserted (how ?) and glued into place. Fixing the front and end ribs suffices, but then the legs for any intermediate ribs are superfluous as the rotation / position is determined. Removing these legs after assembly will also be great fun.
* Diagram #266 implies that curved channels need to be provided in the foam gaps (how ?) of interpenetrating sails. More great fun. How to add the foam gaps of C1d and C2a is not addressed.

:sick: @ZooHead: I apologise for the critique on your concept. I respect you highly, but I think that you have started on the C3D modelling of the curved elements lacking a complete and concise plan.
 

ZooHead

Well-known member
* @GS089 / Gerald (?) and @ZooHead
* My honest (and brutal) evaluation of the current solution to the curved elements (C1d, C2a, ...) is:
* Discard and start from square one.
* You (plural, as opposed to the Shakesperean thou / singular) seem to have ignored aspects of assembly, of interpenetrating sails and of finetuning (filling / planing / sanding in post assembly) by concentrating on isolated details and ignoring all else.
* At best and IMHO, constructing the tricky parts of the sculpture on the present concepts would be extremely time consuming and quite expensive.
* The longitudinal carbon rods (again, IMHO) provide no stability as the ensuing stack can still be twisted until foam gaps have been inserted (how ?) and glued into place. Fixing the front and end ribs suffices, but then the legs for any intermediate ribs are superfluous as the rotation / position is determined. Removing these legs after assembly will also be great fun.
* Diagram #266 implies that curved channels need to be provided in the foam gaps (how ?) of interpenetrating sails. More great fun. How to add the foam gaps of C1d and C2a is not addressed.

:sick: @ZooHead: I apologise for the critique on your concept. I respect you highly, but I think that you have started on the C3D modelling of the curved elements lacking a complete and concise plan.
Very interesting since I got the Idea from you.

But not to worry if you can't visualize the construction, just stick around and see how it goes.

With the "feet" all clamped or screwed together with wood blocks, the ribs from C1d will stand up on their own.
Foam can then be cut and glued between ribs leaving a channel for the ribs of C2a, which I would screw into C1d's ribs.

I am inclined to agree about the carbon rods but I'll leave that to Gerald to decide.
 

ZooHead

Well-known member
I've made some modifications to the feet of C1d to keep them away from the intersecting ribs.
Also I've made room for a pipe clamp to fit through.

Here's a shot in the working position.
If the intersecting ribs are going to be bolted in,
the bolt holes could be laser cut as well.

C1d clamp clearance.jpg
 

ZooHead

Well-known member
Since the ribs of C1d are quite wide, there's ample room for bolt holes.

And I don't have to worry about blocking the holes for the
carbon rods because I can just alter the thin ribs from C2a.

C1d C2a bolt holes.jpg
 

ZooHead

Well-known member
Here's a rendering of the parts.

Looking at this rendering alone, I can imagine a step by step process for construction.
That's something that's always in the back of my head when
building something that will be constructed in the real world.

But it's more important what Gerald thinks, and since he's the artist I must take my direction from him.

Since his idea to make "feet", essentially work stands
built into each rib, I can really imagine it coming together.

C1d C2a render.jpg
 

GS089

Member
Eric, you are so productive and fast I nearly cant cope with you. ;)

Since having the legs/stands, I think the rods are not so important any more, but I will use rods to give it more stability. It´s nice you already include them, but having the stands it would be possible to drill them after having cut the ribs, all at once.

So, when combining the two elements you fix the ribs for the second rib to the “main“ rib? I think that should work, because cutting the two elements as one element will waste a lot of hdf which would be very expensive…

The main strategy to build that thing will be one rib after another, step by step, and with every step I will already cut the foam to the shape, maybe with a hot wire or just with a knife… After the whole element is build I can rotate it and cut all the legs..

Till now I am moste afraid of getting a nice smooth round shape and surface when there is already the second element mounted to the first… that could be interesting to handle. I can imagine to coat everything with a layer of glassfibre or some thin wood like balsa or veneer..

Or wouldnt it be better to just build one curved element and the intersecting elements as small elements of its own, and after every element is finished I can mount them to the first element?
 

ZooHead

Well-known member
Maybe I can try to illustrate, step by step, a construction process.
Then you can evaluate and we can alter course as needed.

The problem with building the thin ribs as add on pieces is
how to attach them after the main ribs are filled in with foam.

If you bolt them on as you go rib by rib, it will be easier to attach. And if the bolt
pilot holes are cut with the laser, the thin ribs will be held in proper position.

Whichever way you think is better, I'll do it that way.
If you need a to take a break to catch up, just let me know.
I know you don't have a lot of time per week to work on this.

Have you worked with automotive body filler? It's called Bondo in the
U.S. and It will make a thin hard durable shell that can be painted.
 

ZooHead

Well-known member
Here's a quick preview of the construction setup as I see it.
A variety of bar clamps will be needed.

construct layout.jpg
 

Helmut

Well-known member
Or wouldnt it be better to just build one curved element and the intersecting elements as small elements of its own, and after every element is finished I can mount them to the first element?

* I have experimented with Gerald´s proposed idea in #272:
1 There is a primary curved element, C1d which is sanded / filled / or whatever post processing is required. This will be identical to the existing C1d, a "simple" contiguous shell.
2 Interpenetrating parts (note the plural) are modelled as separate secondary parts of X2, X3, ... ie curved swings which are assembled separately and are glued to the primary C1d after fine-tuning. There will be 2 or 3 fragments which are attached to C1d on assembly.
* Some tests imply that this is possible. It may even be less complicated and laborious than the existing solution.

* Disadvantage: On the C3D side you are stuck with some Boolean mods / creators which complicate the model.
* Advantage: Gerald´s job will be much simpler (but also cleaner and more logical). There are just individual parts which can be tooled / finished as required and which fit together on assembly. The secondary parts also function as stabilising fins to the primary shell.

:unsure: A small step for man, a giant leap for mankind!
:devilish: Replace "man" with maniacs or Mainiacs, depending on sublte semantics.

* Note that with a simple switch in the basic model you can exchange the primary and the secondary elements. In my example you end up with 3 or 4 separate shells for C1d. Gerald needs to decide which of these constructs is more suitable for constructing / assembling the exhibit.

Screenshot 2022-08-11 at 08.42.04.png

Boolean creator for primary & secondary elements
Screenshot 2022-08-11 at 08.45.49.png

Subelements of the interpenetrating shell, separate
Screenshot 2022-08-11 at 08.49.02.png


Screenshot 2022-08-11 at 09.05.06.png

Switch of the primary / secondary shell if useful for experiments. In this case shell #2 (not shown in the shot) would be the primary element.
 

frank beckmann

Well-known member
From a more practically view and given the "splines" at start and end are parallel I´d merge the 2 intersecting "bows" into one. Laser cut every 10mm - glue with 10mm distant blocks spreading along the inner peripherals and check for a filler or Filling compound that can "bridge" 10mm gaps - stay "put" and is sandable. I´d built up that structure vertically - no need for rods or supporting material which has to be removed in the end. Total material battle either way.
 

Helmut

Well-known member
* I have suggested such a combination of the splines (primary + secondary,) somewhere above.
* My prime concern with this solution would be that tooling the assembled / combined shells may be tricky to impossible. Some of the internal edges would hardly be accessible and achieving a smooth surface / precise and sharp angles would be quite cumbersome.
:sick: However, I know too little about the proposed methods of tooling and assembly.
 

ZooHead

Well-known member
In any case if element C1d is made with multi rib construction, it is in my opinion a relatively easy build.

Some random info:

Materials: There are many different types of foam 7 foam types compared
Extruded polystyrene is the winner and you can order a block the size of a school bus if you need it.

Adhesives: Professionals that sculpt foam use Great Stuff minimal expanding foam to glue parts together.
It's also essential to have to fix mistakes, as it can be carved after it cures.
I also have to mention Gliddon Gripper primer, some people have tested it with excellent results and it's very cost effective.
I think it's good to have an adhesive that expands and one that doesn't.

Professional Help: There are quite a few pro studios around that sculpt foam and it wouldn't hurt to consult.
These guy's and gal's are great: Aden Hynes Sculpture Studios
 

ZooHead

Well-known member
Here's how I visualize the construction of C1d, with or without incorporating C2a.

Establishing an accurate 90 degree corner was something I had
to do many times while I worked in the movie studios in Hollywood.

When building walls we used a chalk line, a tape measure and the 6-8-10 method.

For this case a standard carpenters square should work fine.
This setup is only about 12 or 13 inches high so it could be built on a table I suppose.

The next step after this one is to glue foam blocks between the ribs, which seems easy.

But I can see the carbon rods being helpful for alignment simply because the ribs
are very thin and I don't know about how flat or warped it may be when purchased.

Anyone who's gone to a lumber yard knows about "Banana Wood",
wood so warped and twisted it's only good for firewood.

c1dbuild.gif
 

ZooHead

Well-known member
The foam can be cut and glued around the carbon rods.
You could even leave a hollow center section.

Some pros primarily carve with a knife and some like to use a hot wire.
I've recently seen a 20 foot hot wire that was just the wire attached to two small rods.

In the pink.jpg
 
Top