In July 2014, I was commissioned by a company called Comradity from Stamford, CT, to create a device that would allow for individuals to be 3D imaged so that they could create figurines of people with a 3D printer. By the end of this project, I worked with Comradity to develop a marketing video that demonstrated the potential of the machine my partner and I created. The aim of this video was to introduce potential business partners to the capabilities of this technology, so that individuals could realize potential markets to venture into with Comradity.
Here is the video we came up with! Comradity was really happy with it- so I was thrilled, but if you ask me I do not think I come off as the most enthusiastic person in front of a camera…
Rewinding a year back a year to the start of my build-log of the design, manufacturing, and testing processes that we essential for the proper functionality of such a technical device!
To get started I made a mockup of the pedestal base in SolidWorks, this way I could figure out what problems I would have before anything was physically made thus saving time and money. An added bonus was that once I figured out the components I would need from McMasterCarr, like bearings or motors, they had.sldprt files available for download that I could 3D print. This meant I could further test the fit and performance of pieces, albet while the machine did not bear any major load, before I committed to buying real parts from the company. Furthermore, these models were imported into my SolidWorks assembly, saving me time in the design process.
After having a solid concept of this product on the computer, I got to work manufacturing parts on a Shopbot CNC Router.
Fast forward to the point where I gave it a simple paint job! There was a good amount of trial and error in picking an appropriate motor that an Arduino could power… I went back and forth on using a stepper motor driver and some 400+ oz motor or just using a DC 115v motor and controlling speed with some sort of dimmer. Eventually we settled on the latter, and eventually the pictures in this log will start showing the DC motor instead of the Stepper Motor. Ultimately this decision came back to the technical feasibility and practicality. Yes, in the future it might be useful to control rotation very very precisely, but for this machine it was way cheaper to not buy a high current stepper driver. An added bonus is that the dimmer that ultimately controlled the DC motor let us fine tune the current sent to the motor, so that no matter the weight of the individual, the platform would still rotate at 3RPM.
Eventually, because of hand drilled holes and awkwardly making certain things together, the machine’s tolerances became a lot looser. The biggest negative impact from this was that the chain drive I designed would have a nearly constantly changing tension in the chain. At this point in the project I could no longer get access to the Shopbot and I didn’t build my X-Carve CNC Router yet, so parts that were funky needed to be fixed by hand instead of recut. In short, a chain tensioner needed to be created for the scanner! This was made out of various nuts and bolts from around my garage, a spring from a nerf gun piston, and some polyester tubing. Sometimes form follows function from a very far away distance… like a creepy stalker with a 500′ restraining order….
This is what the finished standing platform looked like! I’m really happy with how I added the skateboard grip tape to the top surface, people saw this texture and realized this platform was something that was built to be stepped on because of it. There was a cover for the DC motor at one point, but it got lost (D’oh!).
Another part of this scanner was developing a tower to move a Microsoft Kinect up and down to capture the full body of an individual. This was critical, because the Kinect does not have the best field of view, if you had the scanner farther than 25″ from the individual there’s a drastic loss in scan resolution. The mock of of this tower was created with some 20mm aluminum extrusion, some pine lumber, a bit of MDF for the top and base, some I-bolts for pulleys, skateboard bearings, various fasteners, and an awesome 3D printed kinect slide that is featured below.
This video is a test of the tower with a very basic script on the Arduino UNO that controls the scanner. Component wise, the tower has a A4988 Stepper Motor Driver made by Pololu and a 235oz NEMA 23 Stepper motor I found in a surplus scrap yard! The scanner ended up needed a 2:1 pulley so that the stepper could actually lift it, and this made it super slow. Luckily this low speed ended up being beneficial, because when the Kinect was running and Skanect was capturing a 3D point-cloud it could easily lose tracking of the person if the camera or the platform moved too fast.
Okay so for the next part of this project, I had to focus on making it pretty. To be completely transparent, by this point the project had an approaching deadline so I needed to make something quick. The design considerations I had were basically that the exterior needs to hide the electronics from being damaged or seen, allow fairly straightforward access to the electronics if new firmware needed to be updated. I wanted to hit a sleek aesthetic, and if I were to try to nail that better I would have designed a pattern and vacuum formed a final exterior casing- but shoulda woulda coulda… frankly it didn’t look half bad in the end so check it out!
All that was left was detailing the rest of the tower to match the base that I had made out of polyester sheeting and MDF, 2 cans of black spray paint did the trick nicely. Here is everything all together, the scanner is fully completed.