< PreviousA 3-D model of the Dyotville glass factory based on photographic and recorded data. Rendering by Chester Cunanan, 2015.Visualizing DyottvilleThe Making of a 3-D ModelChester CunananVisualizing the past can be a challenge, depending on how far back you want to go. Go back a few years and videos are not that hard to come by. Go back a few decades and you can still find some photos. Go back a century or two, there may be paintings and drawings available. But even though an image may be worth a thousand words, it does not necessarily immerse the viewer. Such was the challenge of the Dyottville Glass Works. How could we create a holistic experience that integrated not only immersion on a visual scale, but also immersion via touch and exploration?River Chronicles | Vol 2 | 2017 | 61We certainly had drawings and paintings, we even had a few photographs, but with the only physical evidence being the partial foundation, it was a challenge to envision the scope of the factory. It was even harder to be immersed in the way the rooms were laid out, how the furnaces were placed, or even really to imagine how the place grew and expanded from a dyehouse into an active glassmaking operation. But we did not have only images on our side. We had briefly courted with 3-D printing technology in our recreation of small artifacts, and the time was right for us to take a technological leap and use 3-D printing to recreate a scale model of the Dyottville Glass Works.Luckily, we had a bit more than images to help with the work. We had the Hexamer maps of the factory, which showed, among other things, the measurements and compositions of the materials used in its construction. We also had a LIDAR1 scan of the foundations, which gave us a rough composite of how the groundwork would be laid out in our model. Using a few photos of the sides of the factory as a reference, we were able to obtain an idea of the material density and textural makeup of the outer walls, which helped confirm the placement of some of the windows and doors. With the stage set, all that remained was to combine each of these elements into one cohesive model.Like any form of building construction, the first step was laying down the foundation upon which to build the rest of the materials. For these, we used a combination of the LIDAR scans and the Hexamer maps. The LIDAR scan provided a rough, partial estimate as to the general layout of certain exterior rooms and part of the interior. The Hexamer maps provided spatial dimensions for the walls, the interiors, and the composition of some of the materials used. These two pieces were vital in determining the general composition and layout of the circa-1858 and circa-1889 to 1902 factories, but did not provide any information about the circa-1816 structure, the oldest incarnation of the factory. For that, we had to turn to our own internal research and the 1831 Kennedy watercolor (see pages 20 and 21), from which we were able to surmise that the oldest part of the factory was the converted dyehouse that stood in the center of the glasshouse on the far left in the image. From there, we were able to roughly map out the foundations and how they connected with each other.Once we had this initial information pieced together, we used Autodesk Maya to recreate the rough foundations. Each structural feature was created as an individual piece to allow adaptation and change should new information present itself. That meant that each window and door, every wall, and each piece of roof had to be its own object within the program. It also meant we could “mix and match” details, as necessary—allowing us to simulate both the circa-1858 and circa-1889 to 1902 roof changes, along with the original circa-1816 glasshouse by simply removing the exterior features and adding the original factory elements.62 | Vol 2 | 2017 | River ChroniclesNext we revisited the photographs available to us and extrapolated, based on the materials available at the time, the wear work on the brick and the general pattern of use and decay that most buildings exhibit. We also took a look at similar structures either documented in text or image. From these, we narrowed down a selection of varying materials that we edited and composed to more closely mimic what we thought the factory textures would be. Again these textures were applied on a modular basis, so each factor could easily be altered and changed. Different rainfall pattern? Oak instead of pine? A different bond style of brick? All these elements are easily changed, allowing us to adapt the 3-D model of the factory, should the need arise.Of course, a 3-D recreation is still just a flat image. Our next step was to take the 3-D model and convert it for 3-D printing. This process, of course, presented a unique set of challenges.3-D printing is a method of dimensional printing by which heated plastic filament is extruded in a thin line/layer over a pattern that the printer reads from the computer. As these layers build up, a whole shape is realized, with each subsequent layer fusing with the one before it. In order for objects to be 3-D printed, they need to be “solids.” In 3-D printing, solid objects are defined as elements that have no holes in them. They also need to be thick enough for the filament being extruded to cool and harden in time to support the next layer. That meant that we had to take our real-world values and convert them into centimeters and, in some cases, millimeters, while retaining the overall thickness between them. It also meant that we had to take the modular elements and either combine or split them apart to create an internal anchoring system that would let the model stand on its own once printed. Finally, we had to position and align the objects on the printing bed so that we would be able to fabricate them in the most efficient space possible.River Chronicles | Vol 2 | 2017 | 63Hexamer map information with material type, dimension, and general layout employed in factory recreation. Dyottville Glass Works, 1878, Hexamer General Surveys, vol. 13, plates 1243–1244. Map Collection, Free Library of Philadelphia.We started this conversion by first determining what elements of the factory model could be combined. We reduced the factory elements into external walls, the inner glasshouse, the glasshouse roof, and a set of external roofs. Each of these elements was selected so that individual pieces could be removed from the physical model to show interior fittings and elements, allowing us to expand the interactivity of the final model in the future. This allowed us to essentially carry over the modularity of the digital model into the physical model. As more information about the Dyottville Glass Works is revealed, we can continue to add onto the 3-D print, allowing us a flexible system for continued growth.We are now in the process of importing the Dyottville Glass Works model into the Microsoft HoloLens virtual-reality program, so that we can integrate “physical” exploration into this immersive experience. Eventually, using VR goggles, the public will be able to “walk” around and within the factory, seeing what workers and visitors to the factory would have seen when it was still standing.Our hope is that this process of multidisciplinary research and execution can create a more holistic approach to public outreach through mixed mediums of digital screen imagery, hands-on model experience, and virtual-reality immersion. We envision an experience where the public can not only browse images and see high-resolution recreations of what buildings and items looked like, but can also physically touch 3-D prints of them, as well as virtually explore the spaces and items. With this combination of technology and immersion, we hope to make more of the past experientially accessible.64 | Vol 2 | 2017 | River Chronicles3-D printed model of the Dyottville factory showing interior. Photograph by Chester Cunanan, 2017.Not all nineteenth-century toys were simple dolls or marbles. Some were designed to educate as well as entertain—like a Prince Rupert’s drop recovered from Dyottville. This 3-inch-long, transparent, pale aqua glass object has a rounded head and a long sinuous tail. It was known as a “philosophical toy”1 because it used the entertaining aspects of the “toy” to help explain scientific principles.2In 1661, Prince Rupert of Bohemia brought the interesting phenomenon of the glass drops before his uncle, King Charles II of England. Mysteriously, the head of the drop withstood repeated hammer blows, but the slightest pressure applied to the tail caused the entire drop to explode into a cloud of glass dust. Fascinated by this “trick,” Charles II sent the drops to the newly formed Royal Society for further study.3 Renowned early scientist Robert Hooke and other members experimented with the tadpole-shaped drops, which were formed by dripping hot glass into cold water. The results so intrigued the society that they were published as an appendix in Merret’s English translation of Antonio Neri’s “The Art of Glass”4 and in Hooke’s “Micrographia.”5 Thereafter, the glass tadpoles became known as Prince Rupert’s drops. Prince Rupert’s DropsA Delicate Strength to Survive the AgesThe Royal Society observed that molten glass rapidly turns into an elastic solid under great stress. Upon hitting the water, a drop’s outer layers cooled and hardened very quickly. As the warm viscous core continued to cool and attempted to contract, it pulled the outer shell into compression, making it very strong. Snapping the delicate tail released the inner tensile stress, causing an explosion of glass dust.6 Benjamin Franklin extolled the wonders of Prince Rupert’s drops and they were sold in scientific catalogs in the late nineteenth and early twentieth centuries.7 Easily created by Dyottville glassblowers, these “philosophical toys” may have been used to teach apprentices the importance of annealing to remove the stress in glass. - Daniel B. Eichinger IIIPhotograph by Chester Cunanan, 2017.Scan the QR code to view a short video of this process, courtesy of The Corning Museum of Glass. Or visit: http://www.cmog.org/video/prince-ruperts-drop66 | Vol 2 | 2017 | River ChroniclesThat must have been the question AECOM archaeologists asked when they pulled the first stationary shaping block out of the water while excavating the 1816 cleanout tunnel. When workers filled in the tunnel during the mid-nineteenth century, they buried no less than eight carved wooden blocks from the glassworks. A glassblower would have used these blocks to cool and shape a “gather” of molten glass on the end of a blowpipe by rotating the glass in the carved recessed shape of the block. This was the first step in the process of making an object, such as a flask. The shape of the block would help determine the final shape of the item. The stationary blocks found at the factory site were meant to be used by the glassworker while standing, with the block resting on a raised platform or bench. Using the block while standing allowed a glassblower to work faster. In the early nineteenth century, glassblowers were paid by the gross, so speed equaled money. Glassblower Samuel Huffsey noted in his diary that he made 40 cents a gross for vials while working at Dyottville during the 1825–1826 season.1 Another type of wooden block has a long handle attached and looks somewhat like a ladle. Glassblowers use this type while seated at the bench. All types of wooden blocks or molds are kept wet in order to keep the wood from burning. When properly moistened, the gather of glass is actually cushioned by a thin layer of steam. Regardless, with continual use, wooden blocks eventually change shape and burn out. Having been waterlogged and buried for 150 years, some of the Dyottville blocks are amazingly well preserved. Some are charred from repeated use, others are unused and still show the carver’s chisel marks, and some are only fragments. Surprisingly, the three Dyottville blocks undergoing conservation at the Maryland Archaeological Conservation Laboratory are either fir or cedar. Wooden blocks were traditionally made from the wood of fruit trees; fruitwood does not have a resinous sap. Cherry was a favorite and is still used today because it has a tight grain and is easily carved.2 Blocks, like most of a glassblower’s tools, have changed little over time and continue to play an important role in the production of many glass products. - Mary C. Mills(Top Left) The Prince Rupert’s drop and wooden shaping blocks are amazing survivors recovered from the furnace cleanout tunnel. The tunnel regularly filled with water at high tide. AECOM project photograph, 2011. (Top Right) A 3-D scan of the wooden shaping block recovered. For more information see page 69. Render by Brett Harte, 2017.Wow, What Is It?River Chronicles | Vol 2 | 2017 | 67Scan the QR code and watch as the archaeologists pull the stationary block from the muck. Or visit: https://vimeo.com/67913120#t=1m35s68 | Vol 2 | 2017 | River Chronicles3-D scanning, or a process known as digitization, gives us the unique opportunity to bring physical objects into the digital world. Through the use of a noncontact 3-D scanner, these real-life objects are broken down into their core digital structures, known as “points” or “vertices.” Four vertices form the basis for a face or polygon, which accounts for the remaining shape and structure of the digital object. In order to complete a 3-D scan, objects are placed on an automated turntable in front of the 3-D scanner. The scanner uses a laser that washes over the width and height of the object, capturing digital points to then store on a computer. When a single turn is complete, the turntable will rotate the object slightly to allow for the next side, top, or bottom portion to be digitized. When the scanner has captured all sides, the scan is complete and can move on to be textured by utilizing high-resolution photographs of the physical artifact.For the texturing of the wooden stationary blocks excavated from the Dyottville glass factory site, the process was handled in a similar manner to other solid, nontransparent artifact materials, such as ceramic and rock. The unique characteristics of the material itself, however, made the handling of the wood blocks different. While wood blocks are fragile, like all Physical to DigitalPreserving Cultural HeritageBrett Harte and Mark E. PetrovichRiver Chronicles | Vol 2 | 2017 | 69View the 3-D wooden blockScan the QR CodeOr visit: http://diggingi95.com/virtual-gallery/glassworking-blocks/The top row is a photograph of a stationary wooden block, the middle row a 3-D render of the scanned block, and the bottom row a textured 3-D render of the same block. Photograph and rendering by Brett Harte, 2017.artifacts, the blocks were especially problematic due to the limited amount of time that they could be exposed to air. Caution was required because the wood would begin to dry out if the artifact was exposed for an extended period of time. The scanning of the artifacts proceeded in the normal manner, except that the required editing and cleaning of the mesh of the digital model could not be accomplished with the physical artifact present for reference. The wood blocks had to be refrigerated immediately after the scans were captured. Without the ability to examine the artifacts during this process, the texturing had to be done quite differently. Despite the fact that these pieces were not present during texturing, the wooden appearance made texturing the digital recreation easier than an artifact that has a specific design or look, such as a hand-painted piece. The wood texture details, such as grain, were able to be mimicked using high-resolution photographs and digital editing to fit the 3-D model, as needed.The digitization process provides an excellent way to disseminate information about fragile or unwieldy artifacts. The digital models can be used for public-outreach events, as well as in support of the research-based ventures of educators, students, hobbyists, and professionals alike.Next >