Ross’s Robot Blog
Beep Boop
Project 1: Digital Surrealism
Project Statement
The Undertaper™ is a surrealist, found object piece showing my own vision of what the head of the being who measures you up after your death would look like, complete with a tape measure for a tongue to make their job just a bit easier. The final piece draws inspiration in look from plague doctors’ masks, taxidermy skulls I grew up seeing at my uncle’s lodge, and Khonsu from Marvel.
This piece is made up of 3 found objects (office tape dispenser, packing tape dispenser, and a tailor’s tape measurer) and two 3D printed parts that I designed to interconnect them. I started working on this piece by taking long walks and imagining what funny or silly things could be dispensed from an office tape dispenser. I originally thought about gum but realized that this wouldn’t really be changing the innate use of the tape dispenser, just using it for something else. I eventually settled on a tape measure for a couple of reasons: 1) the word tape is in both 2) the tape measure is never truly dispensed but has a fun reeling back in functionality. I realized then that the two combined objects formed the shape of a jaw and tongue, so I set to work to find the found object that would be my skull. I settled on the packing tape dispenser because it was the perfect shape and tied nicely in with my tape-based theme.
After finding my found objects, my process involved a lot of dimensioning, sketches, and CAD representations of them for me to design around. When designing something I often focus on functionality first, saving the refinement and details not inherent to functionality till the end. I went through many iterations both in CAD and in my own head of the parts that would connect my different found objects into the skull of my specter of death. While iterations are great, I worked hard on having a functional prototype early enough in my project timeline in order to make corrections on both parts due to the long printing times that come with using 3D prints.
Overall, I’m extremely proud of the work I was able to create especially how the final assembly hinges and moves. I believe this work could be improved by further detailing of the skull and refinement of the 3D printed parts to remove their bulk, but I’m thrilled to be able to share my vision of this afterlife being I made with you all.
9/1-9/2: Struggling for inspiration
I struggled a lot with the start of this project. It was hard to turn off the part of my brain that tries to make useful improvements to the objects I was looking at. I’ve also found that I struggle to ideate with sketching if I don’t have a clear picture of the object in my mind or for reference, so the sketching exercise in class made this project feel even more cumbersome.
I had a break in my day so I went wandering around north campus to look for inspiration and so I could iterate on ideas in my head. I still had nothing to show for it 30 minutes later when I opened up a cabinet and saw a black office tape dispenser. I immediately thought I could find something funny to replace the tape with, thereby changing it’s function, and chose that as my found object.
9/1-9/2: Sketching
9/3: Rough Dimensioning of tape measurer and dispensers
When designing a frame, extension, or augmentation of real world components I often have trouble conceptualizing the addition I want to make unless I have accurate dimensions and representations of the objects in sketches/CAD.
I started by measuring my simplest found object, the tape measurer. I had to make an executive call on dimensioning in imperial or metric, but chose metric due to finding that scotch tape dimensions are in metric. I found that most of my measurements were within .02 mm of round numbers.
I documented all of my measurements on grid paper and then began to CAD up the tape measure holder that fits, on an axle, within my office tape dispenser. My initial CAD of the holder is shown below. I’ve also included my initial sketches and dimensions on the next page
9/3: Rough Dimensioning out parts
9/6: Modeling Tape Measurer holder
Because the tape measurer was relatively easy to model I began modeling the tape measure holder. I wanted to keep the axle behavior that made original tape dispenser interesting so I designed the holder to be a type of cradle holds the tape measurer. The axles on either side should allow for a small rotational motion reminiscent of the tape that once sat on an axle in the dispenser.
I went through a couple of iterations on this initial design, not because I didn’t have a clear idea or accurate dimensions, but because I was experimenting with Fusions software capabilities, limitations, or tricks. One silly one that messed up my initial design was that the symmetry extrude was doubling the length of the extrusion without me realizing it. I’m glad I caught this early.
In the future I think I’m going to polish this design a bit by adding fillets, but I’m gonna leave it open ended incase I need to make changes.
9/11: Modeling Tape Dispenser and Lower Hinge Struggle
My original ideas and concept for my tape dispenser skull left ambiguous an important part of my design, how the hinging “skull” is attached to the my lower jaw. I left this ambiguous because I simply did not have a good answer in my mind at the time and I was worried if I let this hold up my progress I would have to sprint to make up time. I’ve watched a lot of Adam Savage’s Tested and am well aware that it is better to push in a direction, realize you are wrong, and course correct, than do nothing at all.
I started first by assuming that the part that sits in the office tape dispenser (OTD) and has my lower hinge for the skull is separate from my tape measurer cradle piece I designed previously. I decided that my piece will sit on the back top surface of my OTD with an extrusion down into the OTD cavity for support. I thought I might be able to fit my cradle and this piece together like a puzzle that held in place but were easy to seperate from one another. However, due to the strange and hard to dimension geometry of the main cavity of my OTD I was struggling with the dimension and shape of this part.
Scrapping that idea I then resolved to making my lower hinge and my cradle the same piece. This simplified my objects design at the cost of the size of the individual parts, but at this point I was happy to make that compromise. Because I had thought ahead and dimensioned out my OTD I was able to mock up a model of this and accurately dimension an extruded platform off of my original cradle to match.
9/12: Upper Hinge Modeling
To connect the packing tape dispenser to the lower part of my assembly, I decided to use a 3D printed hinge with a M3x40 dowel pin that I had on hand. I chose a hinge because I wanted a opening and closing action with the skull. I initially was going to attempt to 3d print a circular object that would connect to the office tape dispenser and fit around the packing tape dispenser in a similar way to packing tape would. However, because of the packing tape dispensers irregular shape, I found dimensioning of this part that holds the tape to be more of a pain than I was willing to deal with, especially if I’d need to 3d print a couple iterations before it fit “right”.
I settled on designing a piece that my packing tape holder could press fit into. This approach could likely result in multiple iterations, but due to its smaller size and the ease of acquiring the necessary measurements. For my initial prototype I decided to make a simple block with a slot for the bottom of the tape dispenser to press fit into and a hinge.
9/13: CAD Assembly
Once I had all of my original part concepts done I decided to assemble them in CAD. My initial thoughts were that this should be an easy task, but I was not expecting Fusion to fight me so much on assembly. In the end I was able to figure out the best joint mates for my desired purposes with the help of tutorials but I found this supposedly easy to use system a bit of a time sink.
Despite this, I’m super happy with how well things look assembled. I did make a couple more iterations and adjustments to my design once I could see how all the parts fit together. All the trouble getting things to fit correctly in my CAD assembly was worth it!
9/14-9/19: Prototyping at Home and Initial Assembly
Due to the large amount of prints and iterations I thought I might need to make in order to get everything to fit correctly with tolerancing and aesthetics I decided to use my home 3D printer, a first-generation Creality Ender 5. I started by printing the upper hinge that attaches to the packing tape dispenser because 1) this part was small and could print in just over 90 minutes and 2) both the pin holes on the hinge and the press fit cavity for the packing tape dispenser would need iterations for proper fit and tolerances.
Once I finished printing the part I was happily surprised to find that the 3mm dowel pin seemed to fit fine, though the outer diameter of the hinge did crack slightly over time so I bought some krazy glue and iterated a bit on the thickness of this outer hinge. The press fit of the packing tape dispenser was not a press fit at all so I had to redesign the cavity to better fit my object.
I resolved to print this second iteration of the upper hinge with my lower hinge/cradle piece now that I knew that my hinge design was toleranced correctly. The original upper hinge and the assembly with the first two iterations are shown below.
9/20-9/21:
Final MVP Assembly
Now that I had my printed parts and had done my press fit of my packing tape dispenser. I was ready to start my final assembly of what I’m calling my minimum viable product (MVP). This includes my office tape measurer base, the 3D printed cradle/lower hinge, the 3D printed upper hinge, a M3x40 dowel pin, and the packing tape dispenser. For a finishing touch I spray painted the packing tape dispenser white with a paint/primer combo to give it a more skull like appearance.
9/20-9/21:
Pre-Submission Critique
Though I am calling this iteration of my piece my MVP, I am super happy with the current iteration of the design. It hits every goal I had in my initial imagination of a skull that opens and closes with a tape measure as a tongue, and meets the classes project criteria. Honestly, It makes me giggle every time I open the mouth and am able to pull out the tongue.
However, there are many places I believe I could improve on this piece. Due to time constraints and feeling under the weather I opted to superglue the slightly cracked hinges on my upper hinge piece instead of making them more robust and reprinting. I also would have liked to spend more time on the skull portion, maybe by giving it an aged look with some airbrushing. I am happy with the aesthetic and I feel like the tongue and upper skull draw the eye, but I do believe more refinement could be done where the lower hinge and upper hinge pieces mesh together, perhaps through the use of splines to cut out blocks of material. Right now I feel like it has a toilet vibe that I’m hyper-aware of due to evaluating it so much. I did realize this issue with aesthetics a couple of days before submission but already having issues with hinge design due to thinness of materials I decided this approach would be too much trial and error for finishing in the time allotted.
Overall I’m very happy with the state of my piece. I love the mechanical actions of the skull and tongue, I’m okay with the aesthetic but definitely see room for refinement, and I would really like to rethink the hinges if given a chance.
Project 2: Automaton
Project Statement
Babs is a simple automaton of a cow jumping over the moon.The motion is controlled by a cam that sits on a main drive shaft driven by a 1:1 pulley. My main inspiration for this work was my fiance’s tattoo of a cow with christmas lights wrapped around it. Her best friends all got variations on the same tattoo based on a prop cow from a production of Into the Woods that lived in their house senior year.
This piece mixes mediums of wood and 3D prints. 3D printed parts were made with white PLA. My pulley wheels, cam, and piston coupler were all 3D printed. The cow and moon were made of balsa wood provided by the class. I laser cut the cow basing it on a creative commons cow profile I found, but then manipulated it in Adobe Illustrator to have spots matching pictures of cows I had online. The moon profile was lasercut with guidelines, but was hand cut out and painted.
I’m really proud of the smoothness of my cam action and the jumping cow. It makes me happy everytime it runs. The lights blinking add a special touch. The belt driven shaft works, but the flexibility of materials used scares me due to the tension from the belt. I think I could better reinforce this and save some worry. I also love the lights but think it would be better if the motor and lights were on the same circuit.
Finally my thorn was the moon. I personally didn’t love the paint job on the moon and I am also disappointed that I couldn’t get the bevel gear setup to work in the time allotted. In the future I think I would have gone for actual machined gears or more of a pinwheel approach. I also think having a variable speed drive for the cow would be nice to make the jumping slower.
10/2-10/5: Starting Ideas, CAD, and Being Realistic about Time
Starting this project, my initial idea was to incorporate a mechanical differential into my automaton. I’ve been hyper-fixated on cars and automotive engineering the last few months so I thought this project might be a great excuse to put some of that into practice. My initial idea was to have an automaton that was doing bicep curls with each arm. The gimmick would be that I could put different weighted objects in each hand and the rate at which the automaton would curl the weights would change thanks to the mechanical differential. It’s a pretty cool idea.
However, as I looked at the timeline and my own bandwidth due to work deadlines as well as IROS taking the majority of my efforts this week, I decided to simplify my design and do something else. I chose to do a cow jumping over the moon as a gift for my fiance who just got a cow tattoo she really likes. I would keep the spirit of the differential by using bevel gears to make the moon rotate perpendicular to my main drive shaft.
Once I made up my mind I began to CAD out the box I would need and looked for a cow stock photo I could CO₂ cut out of balsa wood. I began to outline a BOM including the dowel rods I would use for my drive shafts, rubber bands for belts, bearings to reduce friction, etc.
10/6: CAD Assembly and Starting on the Box
I wasn’t sure what dimensions I needed to have for my pulleys, gears, etc. so I began by starting with the box. I knew I was restricted between 8x8 to 10x10 so I settled for 8x8 as my starting point. Once I had the box dimensioned correctly to match the available plywood, I began to CAD up pieces I had ordered on Amazon. Having the confines of the box definitely helped and within the next hour I had my initial draft CAD of my pulley’s, bearings, drive shaft, CAM, and motor mount.
I then began to assemble these into box, looking at and modifying the parts dimensions to better fit the space they occupied. At the beginning of the last class that day I also watched Alyssa assemble the box, so I also cut out my box sides in the shop.
Overall, I was happy with the progress for the day and the ideas had begun to take shape into actual mechanisms and objects.
10/11-10/18: Cutting Out Cow and moon. Problems with Prints
For the project criteria, I decided that the moon in my design would be my handmade piece of my project. I cut guide lines in the balsa wood using the CO2 laser and then used the bandsaw to cut out the moon. I struggled to decide what part I would make by hand but settled on the moon after encouragement from Alyssa that moons don’t need to be perfectly round.
I also decided to laser cut my cow. Finding the right cow outline for my project was a struggle. I found a couple stock photos that costed money but gave the profiles I wanted with cow spots, but I settled for a drawing of just the outline. I then drew in the pattern on the cow using stock cow spot and black spot shapes that I warped and manipulated to match pictures of live cows from the side that I hoped to emulate. I’m pretty happy with the result.
I also worked on 3D printing different components and building my box. My original plan of having miter gears and a pulley system did not work out due to issues I was having with 3D printing and tolerances. I eventually settled on just having the cow “jump” up and down using a cam I designed.
10/11-10/18: Cutting Out Cow and moon. Problems with Prints
10/18-10/20: Getting things moving
After some finagling I was able to get my cam setup to work. The rubber bands I bought have good tension as belts but I slightly messed up my distance between pulleys, so there is a small moment on the motor and pulley attached to it. I need to add some polish, but otherwise things are good.
10/23-10/25
I spent the last 2-3 days of the project adding polish to my design. To fix some problems I had with the pulley pulling the drive shaft I took an extra shaft pulley I made as a backup and used that to hold the main shaft in a constant position. I also made backups of all pulley components in the event of one breaking over the final stretch due to belt tension not being perfect. Overall I’m happy with the result, however, I do have things I would improve or change. In future iterations I would have put more polish on the actual box and found a belt that was better tensioned on the pulleys. I also would add a brace to keep the motor from flexing. An easy improvement I’d like to add would be wiring the lights and the motor into the same battery box, but I didn’t have the bandwidth to really get into electronics this time.
Project 3: Emotive Object
Project Statement
Dancing Groot is an emotive object that dances and hopefully makes people smile. The system works by having a servo motor that is controlled to oscillate back and forth between pre-determined set points giving the “dancing” look. I wanted to replicate the post-credit scene of Guardians of the Galaxy by having Groot stop dancing when someone is “watching” so I have him stop dancing once the ultrasonic sensor detects motion a desired distance away. To make the dancing motion smooth I used linear interpolation. I found I could vary the “dancing rate” by changing the amount of time the loop sleeps and by modifying the number of samples.
The piece mixes mediums of cardboard and balsa wood. Because of the project syllabus requirement of having a project that is 150 mm^3 I decided to make the base out of balsa wood. I had limited experience with making 3D shapes out of laser cut materials so this was a challenge to have to think simultaneously in 3D and 2D. To give my groot a more organic feel, I decided to make him out cardboard but wrap him in brown paper bag material. I connected him to the servo by hot gluing a circular piece of cardboard I cut out to the servo motor and then I used a brass brad pin to secure groot to the circular piece, giving him a bit more sway back and forth.
My favorite part of this project was the baby groot. The eyes my fiance picked out for him really give him an extra level of cuteness. I’m also really happy with how the box generally turned out and that my dimensioning only needed slight modifications between my initial concept and the final one. However, I do wish the box had a bit more polish with better laser engraving of the tape deck on the face so it was more noticeable and I wish I had added a hinge so I could open and close the front without relying on tape. I also want to improve the dancing action by having it correspond to a songs BPM. Abhishek brought up the idea of an FFT and a microphone which is great but I also don’t want him to dance to people talking so I’d have to get into signal processing at a deeper level than I have ever done. My least favorite part of this project was the ultrasonic sensor. I think my implementation was pretty good but I had trouble with it performing consistently or sensing imaginary objects due to noise. I would likely replace this with a better version or look deeper at my code in future versions. Overall I’m really happy with the initial project and I think some polish in the laser cutting, code, and maybe a coat of paint could take this to the next level.
11/1: Starting Out
I was struggling at the beginning of this project because I couldn’t think of anything that I could make of cardboard that met the requirements of 150^3 mm. Eventually I settled on doing groot dancing from guardians of the galaxy. The basic idea is to have a dancing groot made of cardboard that stops dancing when something is detected by the ultrasonic sensor.
11/5: CAD and Iteration
I started by switching from Groot being in the pot to him sitting on top of the guardians cassette player from the movies so the ultrasonic sensor would look like the wheels in the middle of the tape. I began working on the base but I was struggling to make something that big out of cardboard that had nice dimensions and looked nice so I scrapped that and moved to designing the base out of balsa wood in CAD to be CO2 cut. I based all the dimensions for the tape player etching off of the initial dimensions I found for the ultrasonic sensor. The CAD is a mess but I’m pretty happy with the result.
11/7: Cutting and Building Base
I finished up polish on my sketches for the laser cut base and then did a couple different version of the final base. I had some issues initially with the way to import files onto the printer but once I figure that out with the team at the FRB shop I was in business.
I still haven’t got the lasercutter down and it sometimes doesn’t cut all the way through the material on the first pass so I often have to run the program twice. Once I had two versions of the base cut out I got to work wood gluing the sides together. I decided to leave the front part of the design unglued and just taped it with clear tape so that would be easier to disassemble and adjust wiring in a small space.
11/8: Grootin’ Tootin’
After my base was finished I moved on to making my little groot. I made a couple iterations of him because getting sizing and layering cardboard to be thick/strong enough was a challenge. My fiance helped by picking out the cutest little doll eyes from the store for him and drew a cute little smiley face on him. I also got most of the code going to make him dance back and forth when no one is watching.
11/9: Finishing up
Finally, I put everything together and finished fine tuning the code to get my little guy to react correctly. I thought about spending more time to get the BPM of the dancing to exactly match a certain song but I kind of like him just dancing to the beat of his own drum. Overall, I’m really happy with the little guy.
Final Project: is mystery
Project Statement
Group Project Statement:
Inspired by Dr. Suess’s story The Lorax, The Once-Lit Lamp is an interactive art piece allowing the viewer to step into the world of the iconic character, the Onceler, in the form of a lamp with a touch of magic. As the viewer approaches The Once-Lit Lamp, one can engage in an act of environmental stewardship: upon touch, the lamp awakens and comes to life, eventually setting into motion a delicate mechanism of the growth of a vibrant flower. Symbolic of Mother Nature’s resilience, the flower blooms and sways amongst the scene of truffula trees, mirroring the positive impact of humanities responsible actions on the environment. In doing so, The Once-Lit Lamp invites viewers to reflect on human actions that affect the health of our natural world through the lens of Dr. Suess’s The Lorax. By simply touching and interacting with the lamp, the viewer becomes an advocate of the environment and its beauty.
Personal Additions to Statement:
We took inspiration from Project 1 using a found object (the lamp) as the center of our project. To create the effect of the young tree growing due to the light shining we utilized an arduino reading changes in light from a photo resistor caused by turning the lamp on. The Arduino then commanded a stepper motor to rotate. The stepper motor is connected to a rotary to linear actuator called a scotch yoke made out of 3D printed PLA parts. The base is constructed out of plywood and 2x4’s. The trees were made out of different sized dowel rods, painted to match the patterns from the lorax.The rose of this project
for my contribution was getting the rotary to linear actuator to work and making the tree grow. The bud for me was the noise made by the mechanism and I’d like to make it quieter with a better motor for the task we are giving it. The thorn of this project for me was the base. When I was assembling it everything seemed flat but over time the box had a wobble to it.
Brain Storm
Initial ideas
11/6: Initial Brainstorming meeting with the Team
For our initial meeting each of us on the team brought 1-2 ideas for projects to discuss. Ashley pitched a great idea of having a little lamp character that we would move around or have react to a person. Gongwei brought up the idea of having a flower open and close, so we moved forward with having a lamp that has a flower that grows and blooms as it grows. Our sensor will be a light sensor, and our actions will be the flower growing and the flower blooming.
In terms of responsibility, our group is split well. I will be taking the lead on fabrication and mechanical design, Sammie has electrical system and coding, Gongwei has major power systems and electronics experience, and Ashley has an art and computer science background.