Re(Calibrator) Phase16

Re(Calibrator) Phase15

3D scanner, is a relatively recent invention; a medium size, handheld device allows the user to scan a three dimensional object, as one would  with a bar code reader. But unlike the bar code reader, this piece of equippment works with its own  software, which means the device is constantly connected  to a latop/computer and the output is  monitored  on the screen at all times.

Artec 3D scanner

The scanner essentialy takesphotos one after another, it is constantly flushing when in use, and later  this images or frames, are stitched together to represent the scanned geometry.

Scanned geometry

Above image is a snapshot  of the software that works with the 3D scanner. Here the  geometry is already stitched together and  afterwards a mesh  skin is  created  based on  this structure with the method of "fusion". It is fascinating that  this software works with 2D  eraze mode with a 3D object.  Another  interesting aspect is that any of the scanned frame can be  removed from  the stitched  final object.

On the examples below I have 3D scanned the glass and lead model and took it into Z brush to assign different materials.
Since the 3D scanner uses the frame technique to assemble the geometry, there are variety of ways to scan the same geometry, by manipulating the handheld scanner and setting it to custom directions.
For these reason I have obtained a number of gemoetric outcomes with scanning the same object over and over, when the data as similar and yet didfferent every time.

3D scanned geometry  test 1  (Spherical intensity)

3D scanned geometry test 2 (Metal dots)

3D scanned geometry test 3  (Reflected foil)

I  have discovered that the shininess of the scaned object also affects the scanning process, since the scanner directs flushes of light to the  model  and reads back the geometry, so depending on reflectivity and  trasparency of the object, the results were  different.
The above images show the lead construction, that  was later sprayed with mat white paint.

3D scanned geometry test 4 (The fragmented)

3D scanned geometry test 5 (The fragmented)

3D scanned geometry test 6 (The fragmented)

The above are the glass scans, which seems a lot more fragmented, due to the mentioned phenomena of shininess and reflectivity. When the scanner sends the light rays, some of the flushesare  lost and  the  software registers an absence of a geometry but then another set of frames create the same obsence from a different view when the direction of the scanner is manually changed,  so in essence this is the same data, but modified and distorted.

Re(Calibrator) Phase14 Part2

Below are some abstract conceptual composition illustrating "translations" of the forms achieved with glass and lead experiments from the previous chapter.

And the scenarios below  are speculations of what it would have looked like if those structural elements were implied upon buildings etc.



Re(Calibrator) phase14

The biggest concern  at this point was to source a kiln. After countless meeting, letters with suggestions and  numerous debates, I had to give up on the idea of installing a kiln in the Bartlett workshop. So what next? This unleasant limitation  pushed me towards an  improvised slution ofusing a gas torch to slamp certain kind of glass on the negative moulds i have created based on the positive CNC  milled pieces.

glass and lead model

Above is my latest test with slumped glass flakes and molten  lead.
But before I got to this piece, I went through series of experiements with different glass.
First I tried glass pebbles and the results were rather innsignificant.

Glass peblles

Firing the glass pebbles

Apparently my handheld torch was  not powerful enough to affect the strong structure of the pebbles. Next I tried frit glass.

Glass frit firing

With the glass frit I soon realised that some kind of binder is required  in order to keep the granules from being blown away by the heat.  I tested few binder liquids, including gum arabic, corn starch etc.  and they gave various results.

Glass frit  with gum arabic

Glass frit with UV glue

Afterwards I have tested glass flakes of various sizes and thicknesses.

Firing glass flake

One of  the biggest isssue with hot glass is the anealing, the gradual decreasing of the temperature, to espcape cracks and breaks, therefore after firing i was expecting this fragile piece to collaps shortly.

The Glass flakes holding on the shape

But however the  cracking and  breaking  did not happen. Hence I decided to proceed  with building up on this technique as it seemed to be the best materialfor conveying the initial design idea.

Glass flakes  build up on  the mould piece

During this  process, I have re-introduce the notion of the bodily  gesture. While working on these peices i have been controlling the hand held gas torch in certain ways, always changing my direction and shaping the piece to my desire. This gave me an  immese freedom which i would  have not had, if i was to use a kiln!

glass  flakes  fired with  a UV glue

the design intent

Glass test piece reflecting the digital intent above

There were still many issues to consider; the right binder, the weight and the  gravity, was  the piece going to  collapse under its iwn weight? What is the right mechanism to perhaps link the different pieces together? There are many question yet to be explored, however I feel that this is the right route and perhaps at some point I could take the physical model back to the digital world by  scanning it in and observing the differences.
I  decided to introduce a heavy metal to the process, to see if there was a potential to it and just to be away from glass for a moment.

Lead model  showing the convex side of the mould

Lead model showing the concave surface of the mould

Again I was  able to control the torch in such a way, that with one hand I was feeding the lead wire and with the other hand melting it into the chsen areas of the mould. It was quite incredible how I was directingthe flow of the molten metal, and yet  I did not have control  over the exact shapes and details that  it was forming due to its physical characteristics.

Lead embracing glass

The above test model is a speculation on how the glass could communicate in the organic cage of the lead. There is still more experiments to follow, with bigger scale.
There are few things I intend to do from here; scan the model into a digital software and place this structure into various suggestive environments.

Re(Calibrator) Phase 13

Now that I had some idea  of the 3 axis  cnc milling and the potential of the geometries that was possible to achieve with this method, I  looked at the geometry again and realised that I will need to  employ different glass making techniques.
I have gone  back to more complex version of the  initial  geometry and broke it down to 22  components.
First and foremost, I will most likely  use the direct  glass slumping in one piece technique for some of the parts.

Direct slumping method

I estimated that about  13 out of 22 elements could be generated with this  method.
There is also  the  glass slumping on a  wire mesh method, which some  of the forms  might benefit more.

Glass slumpig on a wire

In the proposed example, it is obvious that a 2nd material will be  needed  to keep the elements together.
Some of the forms  will have to go through  the core glass casting.

Intricate  core casting

In this  scenario, a two piece mould will be  produced with plaster and  quartz/silica mix.
For some of the forms, I am considering experimentations with rather  uncontrolled hot wire slumping.

Through  wire slumping

It is an exciting  method to test, however might  not be necessary for the proposed shapes.
Another thing  Iwas hoping to  achieve was the glass  slamping  in such a way, that would allow for a certain fusion to happen between the mould  material and  the glass.

slamping on a "permanent mould"

Although on the above example an actual fusion is not taking place, but it is nevertheless illustrating an intimate relationship and a  union betweeen the glass and  the mould.
I decided to test the range of polyurethane blocks that are effectively machinable on a 3 axis milling machine. I thought it  would allow me to fabricate negavites of the  geometry and then get the positive glass based on those moulds.

Polyurethane range 1 exposed to temperature

The experiments revealed that the all the range below the M440 density start  to burn under the temperatures as little  as 500 degrees Celsius.

Polyurethane range 2 exposed to temperature

Those samples with higher densities also showed signs of weakness around temperatures 600 and above.

Polyurethane range 3 exposed to temperature

 The most resilliant in the range gave in  at about 850 to  900 degrees.
These experiments do not necessarily mean  that I  couldnt still try the fusion of the moulds with the glass, however it will be difficult to find  the facilities for such experiments that can be harmful for the  glass kiln.

Re(Calibrator) Phase 12

Now that I have done the tests on a small piece, it is time to start fabricating the actual geometry, piece by piece. I will be using the 3 axis CNC router and this means a great limitation of the geometry of the outcome. Each piece needs to be precisely set up in such a way, that it can only on side of the surface with the Z axis moving up and down and back and forth.

cutting the mould pieces in 3Ds Max

The scale is 1.4 meters long and about 0.7 meters wide, which means that there are going to be hundreds of small pieces to be CNC cut separately. There are very many factors that make the cutting very difficult as one each stage I face the limitations of the software, the stock material thickness, the length and availability of the cutters. 

Digital mould pieces ready assembled for fabricating

This stage of the project proves to be the most time consuming and labour intense, and it requires of me coming up with new solutions when things go wrong very unexpectedly on each step of the fabrication. It is going to be a great challenge to put these pieces together for further development.

About 10% of the overall geometry cut

Finally I have approximately 10 % of the geometry cut and ready to be assembled together. This means the end model is going to have about 30-150 pieces, which I will use to create the negative cast out of plaster and silica and then will start firing glass. As I have mentioned, I am targeting on creating about one and a half meter long sculpture, which will have a good enough resolution for even the smallest details. The amount of practical knowledge I have gained after producing these pieces, is immeasurable, especially when it comes to understanding the limitations of the tools and trying to turn them into your favour. This is a lengthy process and perhaps not the most exciting part of my research. Software is very allowing, and it is easy to design the most complex and intricate shapes, but how much of that geometry is possible to produce? I will know the answer soon.