Custom Glass Castings from Digital Designs

This post is a follow up to a previous one about techniques that I have been exploring to transfer digital designs into cast glass objects. This began last year at a fellowship at  Wheaton Arts’ Creative Glass Center of America and was expanded upon this year while teaching at Pilchuck Glass School and in a workshop at Detroit’s College of Creative Studies. In particular, I have  focused on one particular technique; using a low cost desktop CNC machine to carve reusable graphite molds for glass casting multiples. There’s a few design limitations to this approach, but it is an amazingly cost effective approach to creating small to medium scale runs of custom designed glass tiles. The molds hold up for hundreds of castings, and possibly even more, so this is an exciting way of creating custom glass design objects and custom tiles for architectural applications. This will be a big focus for much of my work in 2017.

With a clean and simple design, these new geometric tiles channel the 1980’s era video game Qbert, Islamic mosaics, and leverage the material’s clarity and sparkle. I love the simplicity and the illusory way we see through the smooth top surface to see the relief texture on the backside. I am thinking this will become a set of tabletop design objects, with 6″,9″ and 11″ sizes that interlock. However, I am almost more excited to think of them as architectural tile. How cool would it be to have a wall made of these, or have them as accents embedded in concrete?

Below you will see some of this new carved graphite mold work, as well as some student work from teaching at Pilchuck this summer.  Design constraints of this method center around the fact that this process does not support forms with undercuts. Likewise, the machine can only mill material up to ~2.5″ thick and can only do straight plunge cuts as long as the longest router bit you can find. For most 1/8″ bits this means you cannot do any straight cuts more than 1.5″ deep. However, because graphite is a lubricant and it pretty impervious to heat, once the moisture leaves the material after the first few casts, these molds can be used over and over again, with beautiful results. For any schools or glass studios who are looking to create such a setup, I have created a bill of materials for creating such a setup, totalling under $2500. (BTW, I am available to teach workshops! )

The top of the casting is flat, perfectly magnifying the relief texture of the underside.

img_6550

CNC milling the graphite mold

Completed rough pass on CNC, ready for finishing pass

Ladling molten glass into the finished mold

img_4978

The tile on the left is upside down, showing the relief on the backside. The relief side is as nice as the front and they would make a beautiful glass brick wall.

I’ll end with some images from the TaDDDaa!!! class at The Pilchuck Glass School this summer. It was a three week deep dive into 3D modeling, scanning, printing and CNC carving. Here’s some of the class’ work with graphite for glassmaking:

img_6421

This was my first test with this technique at Wheaton Arts and the process used by students for the class at Pilchuck. Here we see rough and smooth CNC carving of the graphite mold, hot glass in the mold, and final product at room temperature.

img_0365

The TaDDDaa!!! class at Pilchuck was a three week deep dive into 3D modeling, scanning, printing and CNC carving.

img_9621

My TA Christian with Phirak and Rebecca, who will be teaching a 3D printing clay class at Pilchuck next summer

img_3624

Rebecca’s mold based on a victorian pattern

img_9784

Jameszie made a Ouija planchette

img_3645

Brent supervising his first carve

img_9746

John, Lee, Phirak and Nikki trying out their molds

img_9827-2

My student John was a master mold maker so he undertook making a two part blow mold and spent quite some time finishing the graphite to a polish

img_9829-2

Blowing glass into the mold

img_9847

Michael’s two part blow mold

img_9749-2

A pile of hot casting molds cooling down at the end of the class casting session

img_0375

A fishscale tile design I was playing with

3D Scanning at the Brooklyn Museum

I was recently at the Brooklyn Museum for the opening of the exhibit Coney Island: Visions of an American Dreamland, 1861–2008. Its a great exhibit and I encourage everyone to go check it out! While there, I took the time to take pictures of several objects to use to generate 3D models of a few Coney Island Artifacts, as well as some beautiful architectural details.

This process of photogrammetry  or “physical photography” as I have come to call it involves photographing an object many times from all angles, taking care to ensure that each image is in full focus. Once photographed, software analyzes the image to find the same point in multiple images and generates a 3D model of where in space each camera was. From there, a point cloud and 3D mesh can be generated. Its a laborious process but its a very accurate way of generating 3D models of still objects like sculpture.

Here’s the processed scans:

Screen Shot 2015-12-04 at 6.45.31 PM

Spook-A-Rama cyclops head from Deno’s Wonder Wheel Park (Courtesy of the Coney Island History Project)

Screen Shot 2015-12-04 at 6.44.25 PM

Pegasus Statues from the Coney Island pumping station. (read about these here)

Screen Shot 2015-12-04 at 6.15.49 PM

Bacchus Keystone from the Brooklyn Museum Sculpture Garden. This scan came out amazing, with incredible detail to it!

Screen Shot 2015-12-04 at 6.34.57 PM

Another Great Keystone!

Screen Shot 2015-12-02 at 7.10.08 PM

Telamon (Male Caryatid) #1

Screen Shot 2015-12-02 at 5.59.32 PM

Telamon #2

Screen Shot 2015-12-04 at 8.14.15 PM

Architectural Detail from the Brooklyn Museum Sculpture Gardens

3D Printed Luna Park’s Latest Addition is the Helter Skelter!

It’s been months in the making but I was able to update the installation at the Coney Island Museum today with the Helter Skelter Building. I plan to have a reception event at the museum on Memorial Day, so stay tuned!                               

A Process for Glass Casting from 3D Printed Positives

I wrote recently about being selected for a Creative Glass Center of America Fellowship, but thought it was about time to show off some experimental castings I made there last fall that helped get me the fellowship. Last Summer I blogged a first test of this process, but this Fall I was able to do some more substantial castings and learn some more. Here’s some process shots:

Wheaton Arts is home to the Creative Glass Center of America in Millville, NJ, where I will have a fellowship this year, based on these experiments that I did this past fall to create a workflo for casting glass from 3D printed objects.

Wheaton Arts is home to the Creative Glass Center of America in Millville, NJ. It lives in this building, a recreation of a 19th century glass factory.

Hank Adams is the Creative Director of CGCA and he does an amazing job running a world class glass shop. Boola just likes eating scraps and thinks she's a person.

Hank Adams is the Creative Director of CGCA and he does an amazing job running a world class glass shop. Boola just likes eating scraps and thinks she’s a person.

While at Wheaton, I got to stay in this lovely vintage Airstream Trailer!

While at Wheaton, I got to stay in this lovely vintage Airstream Trailer!

These were my two successful casts. This slideshow will walk you through the process of creating them and attempt to share my learnings from the experience.

These were my two successful casts. This slideshow will walk you through the process of creating them and attempt to share my learnings from the experience.

I started with 3D prints. This is Makerbot's Artist figurine Zee, which was given to me by Rob Steiner at Bold Machines. I customized him with some devil horns made out of wax.

I started with 3D prints. This is Makerbot’s artist figurine Zee, which was given to me by Rob Steiner at Bold Machines. I customized Zee with some devil horns made out of wax.

I had three 3D Printed PLA pieces to cast altogether- A 24" tall Luna Park Tower (printed in sections), Mat Fraser and Julie Atlas Muz 18" tall figure (printed on Makerbot's Z-18 printer), and the Zee figurine. I used some cardboard and hot glue to piece together a reservoir for each print before going to make a mold around each.

Altogether, I had three 3D Printed PLA pieces to cast- A 24″ tall Luna Park Tower (printed in sections), Mat Fraser and Julie Atlas Muz 18″ tall figure (printed on Makerbot’s Z-18 printer), and the Zee figurine. I used some cardboard and hot glue to piece together a reservoir for each print before going to make molds. I also used some brown wax to smooth some rough areas of the prints.

It was a beautiful day in the shade of the Studio's smokestack.

It was a beautiful day in the shade of the Studio’s smokestack.

I made some small sprues as air vents for the fine details on the tower.

I made some small sprues as air vents for the fine details on the tower.

I also cut wire mesh to reinforce the molds with.

I also cut wire mesh to reinforce the molds with.

I just used a cardboard box for a form for Zee.

I just used a cardboard box for a form for Zee. The bigger molds used tar paper as a form to cast the plaster around the 3D prints.

The classic mold material is a 50-50 mix of Plaster and Silica with wire lath as something to strengthen it. I did a first coat a little wet as a splash coat to make sure I really have the whole surface.

The classic mold material is a 50-50 mix of plaster and silica, with wire lath as something to strengthen it. I did a first coat a little wet as a splash coat to make sure I really have the whole surface.

The mold is filled with mix and sets up.

The mold is filled with mix and sets up.

Because of the silica, a mask is very important!

Because of the plaster and silica, a mask is very important!

It was a very long day of mixing plaster in 5 gallon buckets. The big molds were extremely heavy so I set them by the furnace to dry out some. I was sore that night!

It was a very long day of mixing plaster in 5 gallon buckets. The big molds were extremely heavy so I set them by the furnace to dry out some. I was sore that night!

They were so heavin, in fact that we had to tie them onto the fork lift to lower them in the oven!

They were so heavy that we had to tie them onto the fork lift to lower them in the oven!

Hank is an expert with the fork lift!

Hank is an expert with the fork lift!

The oven they gave me was about 3 feet deep, so we gently lowered the molds in.

The oven they gave me was about 3 feet deep, so we gently lowered the molds in.

Here you can see the bottoms of the 3d prints in the molds. The next step is to burn them out overnight.

Here you can see the bottoms of the 3d prints in the molds. The next step is to burn them out overnight.

There's the bottom of the  Zee figurine.

There’s the bottom of the Zee figurine.

Overnight, the oven went up to 1000 degrees, then back down. It was about 24 hours of waiting until I could open the oven, but the molds burned out very nicely.

Overnight, the oven went up to 1000 degrees, then back down. It was about 24 hours of waiting until I could open the oven, but the molds burned out very nicely.

The Pla burned out extremely cleanly. Yu can see there are some cracks in the molds as the  PLA does expand and contract some as it heats so for the future, I can print with less infill and play with some other mold material formulas to try to minimize this cracking.

The PLA burned out extremely cleanly. You can see there are some cracks in the molds as the PLA does expand and contract some as it heats. For the future, I will print with less infill and play with some other mold material formulas to try to minimize this cracking.

We lost some mold material between Zee's legs, but the mold burned out very cleanly.

We lost some mold material between Zee’s legs, but the mold burned out very cleanly.

The molds were packed with cullet- chunks of the glass they use at Wheaton. Its a Spruce Pine Soda Lime glass. It would be nice to do some castings with optical crystal, but these were just tests so I was focused on what was available and cheap.

The molds were packed with cullet- chunks of the glass they use at Wheaton. Its a Spruce Pine soda lime glass. It would be nice to do some castings with optical crystal, but these were just tests so I was focused on what was available and cheap.

The GB4 computer is the standard oven controller and has been for over 30 years in the glass and ceramic world. I would LOVE to see someone make an internet connected Arduino oven controller with a mobile phone web interface. These things are archaic! If you're an Arduino Hacker interested in exploring this, please contact me.

The GB4 computer is the standard oven controller computer and has been for over 30 years in the glass and ceramic world. I would LOVE to see someone make an internet connected Arduino oven controller with a mobile phone web interface. These things are archaic! If you’re an Arduino Hacker interested in exploring this, please contact me! I have given this much thought and would like to work on bringing glass studios into the 21st century.

Because of the mass of the molds, they have to be slowly heated and cooled. It took another 12 hours or so to bring the molds back up and begin to melt the glass.

Because of the mass of the molds, they have to be slowly heated and cooled. It took another 12 hours or so to bring the molds back up and begin to melt the glass.

We first brought the oven to about 1200F and as the glass melted down in the molds I kept dropping more in.

We first brought the oven to about 1200F and as the glass melted down in the molds I kept adding more chunks in.

One design flaw with these molds was that they were so deep. It would be more ideal if the molds were shallower and the glass did not have to flow down so far. I ended up bringing the ovens up even hotter in an attempt to get the glass to flow all the way in.

One design flaw with these molds was that they were so deep. It would be more ideal if the molds were shallower and the glass did not have to flow down so far. I ended up bringing the ovens up even hotter in an attempt to get the glass to flow all the way in.

It was freakin' hot in there!

It was freakin’ hot in there!

Finally, the glass stopped flowing and I knew the molds were full.

Finally, the glass stopped flowing and I knew the molds were full.

The kilns need to be vented to crash them down to annealing temperature as fast as possible, otherwise the glass can "scorch" and devitrify, causing it to have a more milky appearance.

The kilns need to be vented to crash them down to annealing temperature as fast as possible, otherwise the glass can “scorch” and devitrify, causing it to have a more milky appearance.

I just turned off the oven and left it open for a few minutes at a time. The goal is to let all that heat dissipate and bring it back to ~960 asap. The problem was there was so much mass in those molds.

I just turned off the oven and left it open for a few minutes at a time. The goal is to let all that heat dissipate and bring it back to ~960 asap. The problem was there was so much mass in those molds and the refractory brick holds the heat. It was a couple hours of venting before I could start the annealing cycle on the computer.

The pieces took about ten days to cool down. Annealing glass involves soaking it at particular strain temperatures, then slowly cooling it down to room temp. I left and returned in two weeks to find this.

The pieces took about ten days to cool down. Annealing glass involves soaking it at particular strain points to even out the temperature, then slowly cooling it down to room temp. I left and returned in two weeks to find this.

The molds were so heavy and by now fragile that I just dicested them right in the oven.

The molds were so heavy (and by now fragile) that I just divested them right in the oven.

Luna Park Tower.. in glass!

Luna Park Tower.. in glass!

I was slightly nervous the oven might close on me- that thing was deep!

I was slightly nervous the oven might close on me- that thing was deep!

A couple process shots of divesting Zee.

A couple process shots of divesting Zee.

The mold materiam bloke off pretty easily.

The mold material bloke off pretty easily.

revealing what's underneath...

revealing what’s underneath…

Devil Zee!

Devil Zee!

I used a diamond saw to cut off the reservoir

I used a diamond saw to cut off the reservoir

The final Zee.

The final Zee.

Luna Park tower required a lot of picking and scrubbing to get all the mold material off. I also had to be careful not to clog the sink.

Luna Park tower required a lot of picking and scrubbing to get all the mold material off. I also had to be careful not to clog the sink.

Both final castings. I have a couple parts that broke off on the Luna Park Tower that I'd like to glue back on. but overall a good first experiment.

Both final castings. I have a couple parts that broke off on the Luna Park Tower that I’d like to glue back on. but overall a good first experiment. Above you can see Mat and Julie’s legs- they proved to be to skinny for the glass to flow into.

So what did I learn?

First, the work I will make in my fellowship will be shallower castings 9-12″deep at the most. I will be making relief panels cast from 12x12x12″ 3d printed sections. I think these high relief castings will be easier to produce and yield better results. Though I like the jade-like appearance of the glass in these, I would prefer more of a translucent glass appearance for my final product. The shallower molds will be easier to cast and cool so as to avoid devitrification. I also will do some more experiments with other mold formulas. Perhaps some other mold materials will crack less as the PLA expands during burnout. Finally, I’d like to explore casting with hot billets of glass as opposed to cold chunks of cullet. Basically, hot casting ingots of molten glass from a furnace so they skin up and stop moving, but are still quite hot when they are dropped into the molds. These ingots are then dropped in the molds and will more easily be heated to flow in. This should also yield a clearer quality to the glass.

So that’s it. I hope you enjoyed. I’m looking forward to using the process to make some work during this year’s fellowship!

New Luna Park Photos By Raymond Adams

Raymond Adams (http://www.raymondadams.net/) is a photographer friend who I met out in Coney Island. He’s been shooting in the neighborhood for while now and was a Kickstarter backer of Coney Island Scan-A-Rama.  He recently came out to meet me in Coney and shot some nice portraits of me in the installation of Luna Park at the Coney Island Museum. Thanks Ray!11054745_10204541944282739_71594317_o 11034546_10204544275381015_1535181905_o 11025317_10204495361198191_1899424994_o

Announcing The Great Fredini as a Recipient of Creative Glass Center of America’s 2015 Artist Fellowship!

I’m pleased to announce that I have been selected as a 2015 Fellowship recipient at Wheaton Arts’ Creative Glass Center of America to pursue a body of work based on techniques I have been developing to cast glass from 3D printed positives. I plan to use the fellowship pursue works on themes of Heaven and Hell. These ideas evolved out my interest in Coney Island through Dreamland’s famed attractions of Creation and Hell Gate, but have deeper roots in art history- specifically Ghiberti’s Gate of Paradise, Rodin’s Gates of Hell and Gustav Dore’s Paradise Lost. The themes of Heaven and Hell resonate to me in today’s era of climate change and environmental collapse, so I plan on exploring digital sculpting on top of 3D scanned imagery to create sculptures that can be 3D printed and cast in glass. Below is one work in progress from this new series. I’m really excited about exploring this work and hope to have some news about it in the future!

Work in Progress

Work in Progress