 Hello and good afternoon. Thank you all for listening in. Today we're going to discuss the work Alter Engine by artist Tawa Auerbach. As you can see from the slide, this talk is part two of a talk Peter and I gave in 2018. Part one of this project was centered on understanding the artwork during the acquisition process. Part two has been preparing this object for display and endeavoring to reprint a central component titled Forty Mesh. Tawa Auerbach is a New York-based artist and her works include paintings, artists' books, photography, and sculpture. Her practice centers around the basic elements of human existence, and she'll often explore these ideas through repeated visual patterns expressed across multiple mediums or techniques. You can see examples of this in the image in the form of interlocking shapes on the table in front of her. In 2015, Tawa began working on Alter Engine, which plays at this idea of a helix at the core of the universe. So to make Alter Engine, Tawa used modeling software to render the shapes and rapid prototyping to print the components. To our knowledge, this is the first and only major endeavor by Tawa into rapid prototyping. During our acquisition process, we learned of the exhaustive methods it took for her to produce the components to an acceptable level and how leadingly a print remained in pristine condition. In order to frame our case study, which is the reprinting effort of one of the main components, Forty Mesh, circled in the image on the right, I'm going to give a quick review of the artwork and our work with the artist. So we're looking at a bird's eye view of the nine-foot painted aluminum table with the 126 rapid prototype components. These 126 elements come from 52 distinct shapes that are printed multiple times. Some shapes are printed using the same file that sent to different printing companies or are printed in different materials or have different finishes. The artist started the design process for each of the objects by sketching out a 2D image and illustrator, which is a vector-based design software. And once she was pleased with the initial sketch, she would then import this drawing into rhinoceros, which is a computer-aided design or CAD application. And once it was there, she would manipulate the sketch in three-dimensional space to form the objects. Everything was then exported to an STL, just over short for a stereolithography file. These were then sent or uploaded to various vendors for printing. From our artist's interview, we learned that Taba used five different vendors to create the components from the STL files. These five were selected by her after extensive testing from an even broader range of companies, which was done in order to determine what was possible for shapes and resolution, as well as the different manufacturers' strengths and weaknesses, since identical materials and processes from different companies would produce different results. And these are the primary processes that Taba used to produce her components. The majority are made using the first two, SLS and SLA. She used SLS for the unpainted white and black components and SLA due to its higher resolution and availability of hand finishes for the more intricate shapes. And all of the gold components were painted after production by Alchemy and PainWorks in Brooklyn. And here we can see a breakdown of the different materials associated with the printing processes, which leads us to our reprinting effort of 40 mesh, which is highlighted on the right printed in Acura 60 using SLA. And the whole reason we're speaking today is that the prints, in particular the SLA prints made using these UVQ resins such as Acura 60, Sumos Next, and Barrow Clear quickly exhibited dimensional changes, enough that during and shortly after acquisition, a handful of prints were reprinted or had specialty housing produced by the artist to help slow these dimensional changes. Above, you can see in the slide, are custom supports made by Taba using the STL file to create a negative form or impression, which ensures that the prints are stored in the correct shape. At the time of acquisition, this was only done for a very select few with the understanding that the overall improved packing system was necessary for all of the components to increase their longevity. Thanks, Macon. As mentioned in the talk on the acquisition of 3D printing files, they can be used in a variety of ways outside of printing to check condition, informed packing, and other such future utility. Taba utilized Illustrator to create each of the objects in 2D that made them into 3D via Rhino. Once the file was realized, the artist satisfaction, they were exported as both STL and PNG files. The STL file format was exported for upload to the different vendors who would print the objects, and the PNG format was exported as a reference image for the STL file format. This is a common practice for 3D printing from a third party vendor. At acquisition, we received a set of both of these files, as you can see here. When we received these files, we first assessed their condition by opening in two different 3D rendering software applications. Here, we used Rhino on the left and MechLab on the right to confirm that each file opened properly and rendered accurately when compared to the PNG reference file. Taba also provided us with STL files that she used in conceptualizing the piece, which shows the full work assembled in three dimensions, showing position on the table and measurements where each should be placed geographically. After the acquisition process, one of the first things we wanted to do was see how the work was installed and how Taba approached this process. It became very clear how labor-intensive it was and imprecise when going through the process ourselves. Taba was eyeballing a lot of the measurements and distances, utilizing string and other improvisations to get the placement exact. Using the 3D rendering of the pieces, both guide and reference, allowed us to double check the installation of the work. In this comparison, we realized that three of the components were installed incorrectly. Again, the utility of these files is incredibly helpful as reference material and it offers more flexibility than still photographs, as you can change position and really mimic reality to understand the installation process better. Since working with the artist wasn't going to be possible for every installation, we were looking for a way to make this process way more efficient. We discussed a number of ideas before hitting upon the idea of using the CAD render that Taba had provided to print a stencil for the initial layout of the work. The work could be blocked out in Rhino using the original rendering provided by the artist and creating stencil blocks that can be placed on the table to provide a rough layout guideline as a first step for the installation. This would speed up the process immensely by allowing the objects to be initially placed and roughly brought into line at a quicker pace so that more time could be focused on the precise movements done later by the artist. Here you can see a crude rendering of that idea. It would allow us to roughly place the objects and thus cut down time. I used the SEL file at the tabletop to create the Adobe Illustrator file of six sections of the table. We then worked with Talus to talk through the materials as we needed something rigid enough to handle and hold up over time but something light enough to ensure we wouldn't risk damaging the sensitive surface of the table. Here is a test install of one of the stencils in Queens. We ended up using BB Flute Heritage corrugated board. It satisfied our requirements and was cut using the digital files we made with a laser cutter. In 2019, Curator Michelle Quo encoded the piece in the exhibition New Order which was a fantastic opportunity to test this new template out. The 3D renter of the entire piece became our guide for this reinstallation. Here we are assembling the stencil on the table and here it is fully laid out on the tabletop. We then laid out the objects. During install we marked center points and end points of components in the windows to make the following install even more efficient. Here you can see those reference points here so the letters are what objects need to be placed there refers back to the component number and the little carats that you see drawn are the center points of each box. With the exception of two central components at the front and back of the table the installation was complete and then we had to very carefully lift the stencils off to finished install and here's where the double layer of the BB Flute was important to give us some rigidity as we lifted these stencils off. This immensely cut down the install time. In the test install it took us about five to six hours to install the piece. With the stencil it took us about just under three hours to fully install everything. We even think this will actually be cut down to even fewer time given now that we have practiced with this install method. So now I'll turn it back over to Megan who's going to discuss the reprinting process we undertook with 4D Mesh. And now we'll dig deeper into the star of today's talk, 4D Mesh. You can see the artist render on the left and it circled and read on the right. Here's another view of 4D Mesh shortly after acquisition compared to the render on the right and you can see how collapsed the shape had become. The image was taken in 2016 and the print was made just a year prior in 2015. And even though it's not readily apparent the color of the print had also started to shift towards yellow. And as Peter has detailed, CAD affords the ability to show different views and perspectives which really allowed us to thoroughly understand the change in condition of 4D Mesh. We discussed the condition during acquisition but given the complexity of 4D Mesh and the difficulties stated by Talba in successfully printing the component we all agreed that reprinting and or housing this piece should be purposely postponed until the artwork was slated for exhibition. And as I've mentioned before, we discussed her process for creating the work, the manufacture of the components, the materials she used and future iterations of installation and exhibition. As reprinting was necessary aspect of future installations of the work Peter and I had also prepared a small reprinting test for one of her components to discuss during the artist's interview. So in 2019 when this work was chosen by Michelle Quo for inclusion in the new order exhibition we really felt optimistic for a successful exhibition of UltraEngine given all the work that he had done with the artist to understand the artwork, its installation and possible reprinting. For 4D Mesh we knew the material, the company, we had the STL file. We also had an active relationship with the original vendor from the research phase of the acquisition. I had previously reached out to GPI for samples and for thoughts on their prints with Talba. They were really open and accommodating. They sent me several samples and test pieces of 4D Mesh to have on file. And I say this just to give some context in the level of communication I thought we had prior to reprinting. 4D Mesh was printed by GPI as I've mentioned in Lake Bluff, Illinois. Talba used them to print 14 of the largest and most physically complex works with SLA using Acura 60 and Stamos Next. So we called the company and asked them if they could print us two copies of 4D Mesh just in case one broke in transit. And the company rep who I was familiar with was still familiar with the artist and the print and they said they had samples on file from the last printing. So they said no problem and they gave us an estimated schedule about two to three weeks. But our first hint of trouble was that it took the printer about a month longer to print the components than was agreed upon without any clarity on what was causing the delays despite really persistent follow-ups. And when the components finally did arrive, we impact them and we initially breathe the sigh of relief because they are intact. Hopefully you can see the two on the left of the reprints. You can see that they're closer to that height of the digital file than the one on the right which is the original. But on closer inspection we started to notice these shiny globs and in some cases these globs are around hot pink marks scattered throughout the matrix. And we reached out to the printer to figure out what this globular material was and to ask why these prints were so different in quality in comparison to the original they pre-printed. But despite following up over email and on the phone there were weeks of we're looking into this and we'll get back to you. So about six weeks after these prints arrived we were finally able to sit down and talk on the phone with the company rep. And we learned that the person who oversaw the earlier prints were done with Tava had left the company and that the texts weren't really able to reduce the print of the same quality. These pink marks were to indicate to the post finishing technician where breaks had occurred and the globs were areas of repair. So we asked GPI if they would be willing to try again and they said no and we asked them if they would refund us the cost of one of the prints which they eventually agreed to. At this point the exhibition was about a week away and we just had to decide which print we'd use for the exhibition. And although both reprints had areas of pink marker and globby repairs one was significantly better than the other and both reprints were more dimensionally true to the digital file. The original while collapsed was flawless. This is the better of the two reprints. The pink spots are hard to see at a glance and there's fewer maybe around 10 of these globby repairs. This print was deemed unacceptable. The pink and globs which are more around 20 to 30 were much more noticeable and here's the original which I appreciate much more now than I did previously. We didn't realize it at the time but we took a lot for granted in our assumption of what the printer understood about the goals of the project. We assumed that the artist's relationship would carry over to our interactions and it turns out that Tava's relationship was way more specific than just the company. It was this one person who made the print feasible and in the future we're going to endeavor to develop strong relationships of printers who might produce reprints for us and spend more time articulating the visual priorities of the projects. Ultimately we decided to install both the better reprint and the original and choose after seeing them both in the table and context. Here are the two options and we and by we I mean the curator Michelle Quo ended up deciding that the structure and the height of the reprint was more important than some of the printing flaws and here's a snap of it installed. The flaws really quickly received from just fanning a few feet away but the impact of the dimensional improvement or shape of the reprint becomes more focused. So ultimately one of the reprints was a success and it did satisfy the major visual requirements for the artwork. We were able to print a replacement component with a correct shape using the artist's original materials and printing process. And there's a huge bud coming but it's also clear to us that our assumptions around quality, repair, timeline, et cetera these were not understood by the vendor and refabrication or reprinting was a new concept in art or conservation but the newness of rapid prototyping and the speed at which the technology is changing really added some wrinkles to our approach to reprinting 40 mesh. So I'd like to review some of the thoughts we had around ways to make this endeavor more successful such that if we decide to reprint a component in the future we should think about the pros and cons of using the original vendor and reprinting someplace local or someplace where we have an existing relationship. It was really clear to me by the end of my communication with GPI that we were more of an oddity than a form of revenue and they were not interested enough in the art aspect to put the work in to get it right. One thought we had is to make a contract. This isn't really something when it does with our vendors primarily because it falls someplace between conservation and fabrication. We don't really have a method for trying to recoup losses if a product isn't up to our standards primarily because it hardly ever happens and these are both leading to larger issues of just communication. We didn't take the time to ensure that the vendor understood the priorities for the piece and what would be considered acceptable versus unacceptable. In the future for a component this complicated that requires hand finishing after production is to create a visual check-in after printing before the removal of support material to discuss the approach. I would also have discussed finishing tactics ahead of time with the person assigned. We, as I mentioned, eventually learned that Taba had this relationship with a finisher at GPI in charge of some of her prints who was very sensitive to the aesthetic needs of the artwork and who unfortunately no longer worked there. My background is in bronze foundry work and I really understand that a printer cast can often be only as good as the finisher assigned. In the future I would have discussed repair and marking needs and perhaps had left things broken to be mended at MoMA or would have stipulated not using that hot pink marker and brainstorm other ways for a finisher to address the work. In addition to trying to pave a better path forward for future reprints, one of the best things we can do is just try to extend the life of the printed components. When the work was proposed for acquisition one of the first concerns we raised was the housing of the components. You can see the housing we acquired back in 2016. The cavities didn't properly support the artwork and many of the larger fragile prints difficult to remove safely. The boxes also weren't strong enough for extended travel beyond trips from midtown to our storage facility in Queens. Unfortunately a request for loan to SF MoMA for our box retrospective prioritized a complete re-housing project. So here is one of four new crates built just for the rapid prototype components which provide really good dimensional support and also very thoughtful methods for removal and handling. The new 40 mesh print needed very specific support to hopefully slow any collapse or sag in the resin. We really just hope that this storage system helps prolong the life of these components to reduce our need for reprinting as well as allow the work to travel and be exhibited and accessible to a wide audience. We now are approaching another installation of the work at SF MoMA in the coming months which will give us another opportunity to check how the objects are faring in storage and allow us to go and adjust any of the packing, installation procedures and or any other issues that come to light and this future re-installation of the work. What we've learned through our test printings of 40 mesh in particular is that the translation of these components in reprinting is exceedingly difficult if not impossible to get an exact one-to-one relationship. This means that our care of the components to prolong their life is extremely important. We may still need to produce prints in the future but want to limit that need as much as we can. So we want to just take a moment to acknowledge some of the people who helped us through the various aspects of this project to date and well into the future in and outside of the field struggling with similar concerns to help us determine next steps as we are likely only scratching the surface of the work being done in an area with constantly changing materials and technology as we all know contemporary art conservation does. So finally I want to thank you all for your attention and Megan and I would love to hear any questions you have. Thanks again.