Just noticed the following problem within a day of updating the firmware to version 5.3.0. The problem is that after every print job, the screen splits in half, and the only solution is cycling the power to the printer.
Up until 2015 or so, there were very few plastics and resins for their respective types of 3D printers. Since then, there has been much work done to develop new an interesting resins and plastics for specialized uses, e.g., plastic filament with metal filings, carbon fiber, or even wood chips and resins that are electrically conductive.
Lately, we have been exploring the use of flexible resins and plastic filament. Both the printer manufacturers and 3rd party consumables manufacturers have been developing resins and filaments that remain somewhat flexible after printing. Think of a flexible yet sturdy phone case that is rubbery enough to slip onto a phone but strong enough to withstand repeated wear and tear and drops.
For the Form2, Formlabs has a black flexible resin. Some of the downsides are that the design needs to be printed with thick support attachment points due to the flexible material that may not hold the part as rigid as needed for alignment between layers while printing. Further, the item cannot be very thin for the same reason. For example, printing a phone case fairly vertically will cause small variations in the orientation of the part when it prints the next layer, and this can cause a waviness between layers (pictures to come). I’ve also tried using 3rd party resin because they are available in a few colors or even translucent. The first print resulted in an awesome piece (see below) – a dendritic mathematical growth that is quite coral-like. However, the resin from ApplyLabWork failed miserably as it caused the resin to adhere better to the silicone base layer (PDMS) than the build platform and caused irreparable damage to the resin tank.
For the Ultimaker 3 printer, I have been testing NinjaFlex filament. It is a flexible filament that is not ordinarily compatible with bowden printers like the Ultimaker, but one can buy super-slick bowden tubes to reduce the friction between the filament and the tube as the extrusion motor forces the filament through the tube and into the printhead where it melts and eventually extruded. The first several print jobs failed; however, I found that I can adjust the extruder motor to maximize the tension on the filament and then the filament extrudes more evenly. In the first test, I only tested its ability to be extruded and adhere to a print already in progress. Fortuitously, I needed a golden filament to print an award label for a chili cookoff tag so I used the gold flexible filament to print the lettering. The print came out surprisingly well considering I could not get the filament to extrude at all until I increased the extruder motor tension on the filament to its maximum. Next, I tried a skeleton from Thingiverse. I used 50% infill but am wondering if 100% would do better. Here is the outcome:
Here is a time lapse video from the Ultimaker’s internal camera for the print of a custom stage for a microscope with Professor Christine Helms in the Physics Department.
Professor Helms created the initial design in COMSOL Multiphysics while some tweaks were later done in Sketchup. The printing was done using Cura software with 6 shells, 1.2mm top and bottom, 80% infill, and support structures for three overhangs. It took approximately 800g filament and over 2 days to print.
As part of the century celebration on campus, the Center for Teaching, Learning, and Technology; Digital Scholarship Lab; and Boatwright Library collaborated on the printing of a 2 foot by 3 foot terrain map of campus as it looked in 1914 with 7 major buildings on campus and the roads and terrain as they existed at the time. The image below shows the 12 pieces that were designed for individual printing on the new 3D printer
The photo below shows the final and printed version (the two blocks in the upper, right-hand corner had not yet been finished so they have a pastel look to them).
As mentioned previously, SketchUp 2014 has a new scheme for installing plugins. Trimble SketchUp now hosts a repository for SketchUp plugins that one can access and install directly from SketchUp Make or Pro. There is an Extension Warehouse that can be called by clicking on the button (below) located on the toolbar. 
Here are some recommended extensions as an update to those previously assembled here.
PLEASE NOTE: As of publishing, the Extension Warehouse seems to require an account (free) to download extensions, both inside SketchUp and from a web browser. The unfortunate part is that there do not seem to be any error messages notifying you when trying to download a file when not logged in or if your session times out – it simply doesn’t download.
In order to make the process easier for installing plugins for those in the University of Richmond community who will be using SketchUp through engagement with the CTLT and/or TLC, you may wish to follow the steps below.
There are most probably additional steps related to setting plugins folders permissions for multi-user access and possibly editing plist files for Macs, but there seems to be a new wrinkle in SketchUp 2014 plugins that may cause issues. It seems that all plugins are installed as user plugins (installed in user’s home folder) rather than system-wide plugins. On a Mac, this means the plugins get installed in /Users/[username]/Library/Application Support/SketchUp 2014/SketchUp/Plugins rather than /Library/Application Support/SketchUp 2014/SketchUp/Plugins . In fact, there is no /Library/Application Support/SketchUp 2014/ folder even created. However, in the case above where we want to install the Plugin Loader plugin, I was able to get it to work for at least myself by creating the /Library/Application Support/SketchUp 2014/SketchUp/Plugins folder structure and then depositing the as_pluginloader.rb file and as_pluginloader folder (that are included in the as_pluginloader.rbz compressed file) into the /Library/Application Support/SketchUp 2014/SketchUp/Plugins folder.
The following tutorial contains the remaining steps for creating a cube designed for 3D printing in which the cube consists of 3 interlocking components each made of smaller cubes (click here for the 1st part of the tutorial).
The following tutorial contains the bulk of the steps for creating a cube designed for 3D printing in which the cube consists of 3 interlocking components each made of smaller cubes.
The remaining steps that would be necessary for preparing the sculptures for 3D printing are limited to reducing the dimensions of the interior faces in order to permit the pieces to fit together with the printing tolerances and materials used. A tutorial demonstrating these steps will follow.
Professors Ovidiu Lipan and Laura Runyen-Janecky approached the CTLT with the idea of developing a new fly experiment for introductory science students in the Integrated Quantitative Science course in which students would study the locomotion of various ages of a particular fly. We began researching and developing some options for a device to use in the experiment, guided by the fact that there is nothing commercially available for their application and that the university does not have a machine shop available for this.
Among the specifications were things like the device needing to allow simple loading and unloading of the flies, controlling the flies from escaping, allowing reproducibility and multiple trials for the experiment, permitting video recording of the experiment for analysis, backlighting the device to allow better contrast for data acquisition, and being able to mark the device for “start” and “finish” lines.
Over the next couple months, we developed a few prototypes in Trimble SketchUp and discussed the virtual designs. We then proceeded to make physical prototypes using the CTLT 3D printers (Solidoodle version 2 and version 3 printers). During this phase, the CTLT also learned how to improve accuracy of printing through experimenting with PLA plastic filament versus ABS plastic filament (PLA contracts less as it cools after printing which allows for better printing precision, necessary for fitted parts).
After the last prototype proved successful in testing with flies early in the summer, the CTLT then looked at fabrication options for creating 10 of the experimental setups. For printing the pieces through Shapeways, a popular online 3D printing service, and assembling the devices locally, we determined that the cost would be $2310 (see table below). We then calculated the cost for doing the 3D printing ourselves on our 3D printers to be $137 or about 6% of the online cost.
| Item Description | Cost Each | Quantity Per Experiment | Cost Per Experiment | Quantity | Extended Cost | Shapeways Extended Cost |
| Magnets – 1/4×1/4 | $0.65 | 4 | $2.60 | 10 | $26.00 | $26.00 |
| Magnets – 1/4×1/8 | $0.53 | 4 | $2.13 | 10 | $21.30 | $21.30 |
| FlyTrack | $4.00 | 1 | $4.00 | 10 | $40.00 | $1,004.50 |
| FlytTrack Base | $2.00 | 2 | $4.00 | 10 | $40.00 | $1,170.00 |
| magnet holder | $0.02 | 4 | $0.09 | 10 | $0.91 | $73.60 |
| Plexiglass | $0.83 | 1 | $0.83 | 10 | $8.33 | $8.33 |
| S&H | $0.00 | $6.50 | ||||
| Total | $136.54 | $2,310.23 |
These numbers are for the final pieces. However as mentioned above, we experimented with the design and went through about 3 physical prototypes. These prototypes would have each cost about $120 for testing and would have taken 3-5 days for turnaround time, adding that on to the development time that we had used.
We admit that this is not a complete story about costs. Obviously, we are not including the costs of staff time (although the bulk of the CTLT time was done off the clock and the rest is considered aligned with the goals of effectively integrating technology into teaching and learning). Further, this was not an exhaustive cost analysis. We did not research the costs for traditional fabrication for such a device. Of course factory mass production would be too expensive, but there are “traditional” machine shops (without 3D printing) that could probably make these devices incorporating most of the aforementioned specifications. However, in our experience with outsourced research instrument design, we believe the hourly rate of a machinist’s time would be enough to make this unfeasible as well.
We proceeded to print the pieces in the CTLT and acquire the additional parts to assemble the devices. The time required to print the 10 FlyTracks and 20 FlyTrack bases (10 devices) came to 110 hours. Luckily, very little of it required human attention.
The final design is available from Thingiverse to allow other schools and individuals to use and/or evolve the design.
Directly from the founder and CEO of Tinkercad is word that they are being bought by Autodesk, and there are “exciting plans for Tinkercad.” The shutdown of Tindercad has been rolled back, people can again sign up for accounts.”
We like TinkerCad because it most of the basic features from SketchUp but has an even lower learning curve and can create “watertight parts” more easily (the parts need to be mathematically solid if they are to be 3D printed).
One can sign up for a free account at http://tinkercad.com/. Best practice is to save your work and export it from any cloud service like Tinkercad soon after completion (just in case the service changes).
As a collaborate project between University Communications, CTLT, and Boatwright Library, we developed the largest 3D print to be printed on a UR CTLT printer for inclusion in a video that documents the night culture in the library. The cubic prop measured 180mm (over 7 inches) in its largest dimension and paves the way for large academic 3D printed objects. The Solidoodle 3 printer can print an object up to 8 in. x 8 in. x 8 in., so we expect it will get more use for printing everything from large molecular models to student 3D design sculpture.
Communications developed the typography and provided a sample to the CTLT for further refinement and extrusion into a 3D shape. The CTLT imported the design into SketchUp from an Adobe Illustrator intermediary file format. SketchUp allowed us to create the solid object with some hidden support structures to give the letters more rigidity during filming, and we could model the final look and feel of the object for approval. We conducted a couple scaled test prints and iteratively developed the design for optimal printing. We then set up the 3D printer in such a way that it could execute the 50+ hour print job while allowing time lapse photography of the printing process to be included in the video project.
To read the article and see the video, please visit the alumni magazine web site.
