Troubleshooting the Ultimaker

It’s been quite some time since the first post announcing the arrival of our newest communal 3D printer here at ‘Berg. In the first few weeks of its arrival I put it through a series of tests to make sure calibration was accurate. While it crushed many of these, like the torture test that is Benchy, the experiments led to a fair share of struggles. Fixable struggles, but important lessons that I thought I would share with the community.

The problem: changing the filament.

To gear up for a new print, I needed to change the filament to a darker color. It seemed simple enough. Until some filament got left in the tube. And not a lot of filament. Just a tiny thread less than an inch long that we assumed would be pushed out by the new filament. Turns out this does not matter. The new filament will not push out the old and we were left with a mess. In order to clear it out, we went with the “Atomic Method” (Ultimaker’s terms, not ours). This involved disassembling the bowden tube, cutting a piece of filament to the approximate length of the tube, and manually pushing the remains out. Since this was our first time testing out this method, it took about an hour to change a roll of filament. But the good news is, it has been accomplished.

What did we learn? Tack on an extra 30 minutes to any project requesting a different color filament.

The problem: warping

I have to admit, this one surprised me. In hindsight I’m a little embarrassed by that fact. Making sure that prints stick to the build plate is always something I think about when looking at a model. There are many different ways to make this happen: use a glue stick, create supports like rafts, use an alternative surface (rubber mat) to print rather than the glass plate, etc…What I failed to realize, however, is that anything with a large surface area (e.g. more than 2 in x 2 in touching the build plate) would be more likely to warp. Why? Because larger surface areas cool faster than smaller surface areas. Once it warps even just a little, the nozzle, which dispenses the hot filament, catches and often times loses calibration and/or pries part of the print loose making it even more warped than it already was. As I said, embarrassing in hindsight because this makes complete sense!

Warping can also occur in taller prints, but has yet to be experienced in anything we have tried thus far.

What did we learn? Prints with more surface area touching the build plate are more likely to warp and require a higher build plate temp.

The problem: leveling the build plate

This actually isn’t a huge problem and should maybe be labeled a good practice. The printer sits on a table that could be sturdier. When people visit my office, they oftentimes lean on or bump against the table. Both of these things can throw off the calibration of the printer and cause distortion or failed adhesion in a print. Whenever a print starts to warp right off or won’t even stick to the build plate, it’s a good sign that the calibration is off. Luckily, the printer has an easy process for re-leveling the build plate. By aborting the print, letting everything cool, and following the simple instructions, it’s back on track.

What did we learn? Re-level the build plate! Simple and quick fix.

Where is most of this information coming from? The Ultimaker 2 Manual.

Next up on the blog list, a look at future training and events now that the Fall semester is here…

Printing Robots

It has finally arrived… The Ultimaker 2+ is installed and operating in the IT/DL Suite! In the past week we have unboxed, calibrated, and printed a few pieces to test out this beaut. The first print, this lovely little robot we fondly refer to as Paul. Paul is approximately 1 inch in height and maybe half an inch wide. As the Ultimaker mascot, it was only fitting that he be the first attempt at any printing success.

Up next we put the Ultimaker through the 3DBenchy test or, as it is more fondly known, “the jolly 3D printing torture-test.” This little boat asks the printer to try overhangs, symmetry, rounded surfaces, and inclines. All things that could go very, very badly and cause warping or dimensional inaccuracy, WITHOUT SUPPORTS*. Measurements are also taken to make sure that the printer is calibrated well (or more specifically the software used to send the file to the printer). So, here’s how the Ultimaker stacks up:

Accurate measurements (yay!)
Smooth print, no warping
Successful incline build
Small detail accuracy (mostly legible)

The one STRUGGLE was the overhangs. Each of the archways and windows had some drooping or discrepancy. This really isn’t the end of the world. In fact, it could be solved by adding supports to these surfaces (which can be sanded/melted out later). Overall, 3DBenchy was a success!

 So what does this mean? More testing! Up next we’ll check out duration prints (so far we’re up to 4 hour prints with success), tall prints, and wider prints. All leading up to, hopefully, working with interested faculty and students (and staff) on projects throughout the Fall semester. This will entail workshops in 3D modeling and training on working with (and often troubleshooting) the 3D printer. Lots of planning! But for now, here’s a quick overview of our newest technology:

The printer is currently set up to print using PLA and provide prints as small as Paul the Robot or as large as the build plate can hold (approximately 8in wide x 8in deep x 11in high). We have four nozzle sizes for more detailed prints or quick prototypes. Most any file type associated with 3D building (obj, stl, etc…) can be uploaded to Cura, the free software program that works with the Ultimaker printer. Decisions regarding supports and the density of the print can be made in Cura. This allows one to see how the object will be built layer by layer. These prints are then loaded onto an SD card to begin the print. That’s a pretty quick overview of what it does, but as I learn more I will continue to share.

If you’re at all interested in working with 3D modeling and printing next semester, please stay tuned! I will be working on training throughout the summer and would love to collaborate.

*Supports are small removable structural additions that are made by the 3D printing program


You can also see this post at

Why teach (why learn) 3d modeling?

This week the Digital Learning team graciously invited me to speak at the last Tech Tuesday of the Fall 2016 schedule. For this hour long conversation, I opted to discuss various 3D modeling softwares covering modeling, photogrammetry, and hosting platforms (see previous post).  With Play-Doh in hand, I set out that morning still unsure of what I hoped to achieve in that hour.  I had a presentation that gave a birds-eye view of what was possible, how much it would cost, what the learning curve looked like, and how it could be used in higher education via the classroom or personal research.  Meanwhile, my brain was spinning from a wonderful twitter storm of 3D modeling pedagogy that began with this tweet by John Stewart:

The conversation that followed the presentation was fantastic.  It quickly moved to 3D printing, an avenue I mentioned only once at the start of the presentation.  Like John Stewart, many faculty wanted to know how they could incorporate this kind of work into their classrooms, what the value or affordances were, and wether or not they could feasibly do so in just a semester.  Perhaps my favorite part of this discussion was Patrick Williams, professor in neuroscience here at Muhlenberg College, giving his personal experience with 3D printing some voice saying he was a long-time skeptic of printing until he purchased his own and began using it for his research. His original skepticism or concern is one I have heard often, both at Muhlenberg College and the Wired! Lab at Duke University.  I have found that the concerns stem from 2 different places:  the newness of developing technologies and the affordances of the technologies.

In Inventing the MediumJanet Murray writes, “we are dealing with an immature medium, which is more diffuse and has much cruder building blocks at its disposal than a mature medium like print.”  While 3D modeling softwares have been around for a few decades now, they developed in particular areas outside of education: 3D animation, architecture/spatial design, and engineering.  What I appreciate about Murray’s statement is the recognition that this technology is new and using it means making up a lot as you go along.  The building blocks for 3D modeling in education do not go back decades.  They are not all tried and true.  Using this technology in the classroom for a new assignment or research method means developing new pedagogy, curriculum, training, expectations, analysis, and thinking.  The tested 3D scholarship frequently mimics the real world applications of the softwares using them for exactly the way they were designed to be used.  Think AutoCad.  It is easy to see how 3D modeling could be used in a design or construction course.  The building blocks are there. But what about physics? Religion Studies? Media and Communications? Just because the technology has not been used does not mean it could not or does not lend value to a particular field of study.  As long as a project is well conceived and provides affordances that another medium (particularly video, text, etc…) does not offer.

It is difficult to imagine what one can do with new, and often times unstable, software.  This frequently causes anxiety with faculty and students.   When asked what to do in the face of potential failure my answer is always “that’s okay.”  While some may perceive this as crass or rude, it is not ill intended.  Rather, my personal belief is that these “failures” are never a waste of time.  Something was learned along the way.  Maybe the project did not end up exactly the way it was imagined.  Why? What happened along the way? What was created? How could the process be changed?  Whenever these technologies find their way into a course, be that in one assignment or as the foundation of a course, things are bound to change (see Caroline Bruzelius’ short essay for more).  While the fear of failure is justified, it should never overwhelm the possibilities of a well-developed idea.

While I do not want to make this post go on any longer, I do want to take a moment to explain a bit more about what I mean when I say “well-developed” or “provides affordances.”  Technologies are not used for technologies sake.  They are used when they can benefit research/scholarship.  Again I turn to Caroling Bruzelius and her TEDxDuke talk about using 3D modeling in her course:

Her discussion of frozen plans and the need for a 3D model to represent the many architectural plans, is a great example of the affordance of the technology in this particular case.  “In this particular case” is also a key phrase.  Not every project should be duplicated.  Reconstructing a grocery store that exists today and is a carbon copy of many others in the chain of stores is probably not a priority.  The affordances of the 3D modeling, being able to explore the inside, understand the architecture, visualize the changes, etc…could be gained by a visit to that store or another of its kind, modern photography, or video footage.  This is far different than reconstructing a statue or building that no longer exists.  In the case of a lost building the affordances become far more relevant.  The 3D model can take the written accounts, sketches, plans, whatever evidence is left of this building and create a simulation backed by archival research.  It completely changes the way an audience would understand and interact with the object.  This is just one example of how each project is unique and has to be developed and thought through carefully.

While there is so much more that can be discussed in regards to 3D modeling and the classroom (or 3D printing or virtual reality), I will end this particular ramble and invite any conversation be it interest in incorporating 3D modeling into their research or classroom pedagogy or just general curiosity, feel free to reach out or add a comment!


3D Modeling @ Muhlenberg

Our last TechTuesday is coming up on December 6th! We hope that you can join us for a conversation about ‘how to’ and ‘why bother’ using 3D modeling in education. The session will be held in Seegers 109-110 from 8:30 to 9:30 with coffee, pastries, great folk, and Play-Doh. That’s right, I said Play-Doh. So now that I have left you with that little teaser, here’s a bit more:

3D modeling has been used to create prosthetics, drones, artworks, and many different kinds of models. Interior designers and architects use programs like AutoCad to create buildings and assemble spaces. Students use 3D modeling and printing to learn more about bone structures, replacing the often times expensive kits with a more affordable alternative. 3D animation and gaming platforms are also free and accessible to anyone who may want to use them, no longer locked behind expensive softwares.

On Tuesday we will take a look at the software listed below, going through a fairly high level view of what each does.  The conversation will cover modeling, texturing, rendering, photogrammetry, and hosting.  Additionally, each software platform will be paired with an example of digital scholarship and/or course project.

Check back early next week for an in-depth post about 3D modeling in education.

SketchUp: Additive 3D modeling/construction
Cheetah3D: 3D modeling and rendering platform *(not free)
Unity3D: Gaming platform
123DCatch: Photogrammetry software
SketchFab: 3D model hosting
TinkerCad: Simple 3D design for 3D printing

Educational Resources
3D Printing

3D Data Visualization