Wednesday, February 12, 2014

Short one, more of a question on Lithography style 3-D printing

Would it be possible to make a photo-curable resin, that would behave differently depending on what frequency of light it was exposed to during the curing process?  Even more ideally could you tune more than just the color, maybe even the mechanical/chemical properties.

Stereo Lythography onto Existing Items

So I finally remembered an old idea and I'm kind of excited so I'm going to go reasonably technical.

  Current 3-D printers are designed to start from nothing and by extruding material, sintering metal, curing a photo reactive resin with light, etc....  On their own these various technologies for rapid prototyping can create a range of wonderful items, a big trade off for these existing rapid prototyping technologies is how long it takes to make an item and the cost of the materials used.*  Make Magazine recently posted about a software package that reworks item designs so that the majority of the item can be made from, relatively, cheap lego bricks, with only the most complex components actually made out of printed parts.  This is a fantastic step in decreasing the prototyping time of an item, but I feel more sophisticated and flexible solutions can be developed.

Imagine you are a tourist destination store or amusement park, your profits greatly benefit from being able to provide unique items for people to commemorate their visit.  While a standard 3-D printer would be excellent at providing the unique aspect, the necessary turn around time could drastically reduce sales, especially if the customization is rather minimal.  You need a 3-D printer that is smart enough to add material to an existing item, you also would need it to be easily handled so that any of your employees would be able to fabricate the memento without supervision.  What I would like to suggest is a printer that integrates the requisite technologies to do just that.

Figure 1
Step 1:The base item being inserted
 into the printer to be scanned
This proposed 3-D printer would need a 3-D scanner capable of accurately mapping and positioning whatever item was placed inside the printer, it would also need a grip system capable of holding whatever items were placed inside the printer and keep the object stationary enough that the object would not move during the print process.  Of the existing 3-D printing technologies I believe that stereo-lithography is most likely to be able to integrate this concept in the near term**.  Once an item is secured inside this new lithographic printer it would be scanned and mapped by the inbuilt 3-D scanner, this scanning would ensure that the computer has a very clear idea of where it would need to add material.  As each layer of material was cured the object that was being added to would be lowered further into the bath of resin, layer by thin layer the starting object would be transformed.  A formless basic action figure becomes a custom hero, showing all of the traits chosen by a family to commemorate their visit to Action Universe.

Figure 2:
Step 2-10,000: Layers of resin are cured onto the
selected item
I believe that an extrusion style 3-D printer could integrate this approach, but the system would require far more mechanical precision than a stereo lithographic printer.  For technologies like laser sintering I am more dubious that you could directly print onto an existing item and have it make sense, in theory a manufacturer could first print on a small containment wall, then deposit a layer of the material that they actually want to add, fuse the target material, add another layer to the containment wall, and repeat the process, but it feels like it would be less likely to actually save time in manufacturing.


Originally I had planned on using this post to improve my chances for applying to work at FormLabs, they have a 3-D printing system that uses stereo lithography, but I realized in writing this article that the approach I am suggesting for adding printed material to an existing object, would be more difficult to impliment considering the FormLabs printer uses a shallow resin reservoir design, Figures 4 and 5 show scenarios where it wouldn't be feasible to print directly onto an object using Form Labs style of printer and where it would be, respectively  In Figure 4, someone wants to print onto an existing object, but the desired start point for the print cannot be covered by the photo curable resin, consequently they will not be able to print directly onto the item.
Figure 4

Figure 5
In Figure 5 the user has selected an object with a flat surface that can lie flush with the resin reservoir, this would allow for printing onto the existing item.

In the majority of cases where users are simply trying to create something akin to a hood-ornament the could use 3-D scanners to create a mount that would then allow for the addition of other printed objects.  This concept makes the most sense for fully covering an existing item, like a Lego model.


*a kilogram of plastic for an extruding printer= $35/kg
Photo-curable resin = $150/L
Sintered Metal=$8/cm^3 (I couldn't find the price of pure feed stock, so I am providing the price per printed cubic cm (hey better than nothing))

**I haven't had a chance to work with stereo-lithography and so I am  making some light assumptions on the flexibility of the technology, if it turns out I am wrong please let me know.
*** the idea of priting custom toys for amusement park guests is not my own, I recall reading about it in both the book Makers, by Corey Doctorow and this article on Disney Parks unique princess toys.

Thursday, February 6, 2014

Fixing Traffic Circles

According to traffic safety records replacing traditional intersections with a traffic circles will have a measurable reduction in property damage and accidents, what these statistics ignore is how frustrating it is to help people navigate a traffic circle.  In the personal experience of the author, knowing which road to exit onto at your first encounter of a neighborhood's traffic circle, effectively describing where to go may cause confusion.  For this article we will look into two potential standardized ways for labeling traffic circles to aid in communication.  The first version of this concept takes inspiration for the how runways are designated, every 10 degree point along the circle of the roundabout has a new value of 1/10th the angle, for example if one of the turn offs from the round-about exits in an easterly direction, it will be 90 degrees off of the true north orientation and be labeled exit 9.  For an exit that turns off towards the southwest, you would technically have the angle be 215 degrees, but for ease of communication it is preferable to use whole numbers, it would be up to the discretion of the city planner whether the number is rounded either down to become exit 21, or rounded up to 22 as shown in the example image below.
Runway Style Labeling of Roundabout Exits
Alternatively the traffic circle could be labeled more simply where each exit is named alphabetically, with A being designated as the first exit clockwise from geographic north.  The advantage of this labeling approach over the runway style labeling is that drivers are more immediately warned that their chosen exit is coming up.  Drivers getting off at exit C know that at exit B they need to make sure they are getting in position to turn off.  The challenge that might face this approach would stem for the need to relabel exits if a new exit is added to the roundabout, but from a practicality standpoint this is a relatively negligible concern.  
Example of the Alphabetic Labeling Approach

Both of these concepts are simply the author's suggestion for how we might improve roundabouts to aid in communications.  Suggestions for improving these concepts or alternative approaches are always welcome.