This is an example. See below for Jim’s audition
Right-click or ctrl-click this link to download.
This is an example. See below for Jim’s audition
The code is up and running at alchemybrewingNM.com, but needs some general features added before it is made available to the public.
A good friend of mine, Patrick Johnson, has been scientifically honing several Belgian-style ales at his Cedar Crest brewing lab for several years. For the past two years, he’s focused his attention on the Dubbel and Trippel, and he thinks that they are nearly ready for the public. I’ve had the opportunity to try many iterations of these beers at semi-regular tastings, and could not agree more: these beers are incredible!
Given a little more time, Patrick hopes to release these beers for public consumption under his Alchemy Brewing label. With the help of designer Jesse Arneson, we’ve created a simple website that will allow Alchemy’s growing fan base to keep up to date as they traverse the business maze on the way to becoming a full-fledged brewery. I encourage anyone that enjoys beers of all types to log on to alchemybrewingnm.com and sign up for the occasional email update from Patrick: He may even invite you to one of his informal tastings
For my part, I’ve constructed a WordPress-based template, which will allow Patrick to easily post recent developments, and send them to his followers.
I have just migrated to a new CMS, and look forward to adding more content about past, current and future projects very soon.
Verification and Validation of engineering problems involves the application of statistical reasoning to established physical models. Under the employment of Dr. William Oberkampf, I developed simple Verification and Validation models of well understood physical phenomena to be used in conjunction with Dr. Oberkampf’s forthcoming book on the subject. The models were all written in Matlab. In addition to showcasing the principles of verification and validation, some of the models took advantage of Matlab’s Symbolic Toolbox and extensive ordinary differential and partial differential equation solvers.
For the purposes of better explaining the project here, I will outline one of the examples: A 1m by 1m by 10cm thick square aluminum plate is perfectly insulated on one side, experiences convective cooling on another side, is loosing heat at a constant rate from the third side, and is gaining heat energy at a constant rate on the fourth side. The steady-state temperature across the plate is fairly straight forward to model using heat-transfer theory if all of the physical parameters are considered to be exact. However, what would happen if the conductivity of the plate was subject to some uncertainty? Likewise, what if the ‘perfect’ insulation were faulty? The answers to these questions are computationally expensive to arrive at, but are invaluable in evaluating the designs of physical systems.
UPDATE (4/25/2011): Dr. Oberkamp’s book on the subject of Verification and Validation has been published by the Cambridge University Press. For more information, please visit the Cambridge website.
The goal of this project was to design, fabricate and test a device which could accurately position a 60-micro-meter-diameter fiber-optic cable. The micro-actuators were fabricated on 6 inch SOI wafers using standard photolithography techniques at the University of New Mexico Manufacturing Training and Technology Center(MTTC).
First, triangular forks were isotropically etched into the silicon using KOH solution. Careful inspection of the triangles allowed for a 10,000% more precise crystallographic alignment than that provided by the wafer flat. Next, the longitudinal trench was patterned and isotropically etched to a depth of 10 microns. Finally, the actual devices were patterned and etched in a Deep Reactive Ion Etcher (DRIE), and released from the substrate in hydrofluoric acid. All three fabrication phases were carefully designed into a single photolithography mask to lower expense.
The devices work by placing a voltage across the two anchor-pads (top and bottom), which causes current to flow through the actuation arms. The arms’ resistance to current flow causes the them to heat up and expand, pushing the shuttle (center of image) to the right.
The image to the left is a three-quarter view of two very thin, very tall walls intersecting a very thick wall. The height of the walls is 70.86 microns, and their width is 3 microns. These walls were originally fabricated patterning two 3-micron wide straight, parallel lines 3-microns apart from one another on a 6-inch silicon wafer with 1 micron of silicon-oxide grown on it’s surface. Material was then removed from either side of each line using a Deep Reactive Ion Etcher (DRIE). When enough material had been removed (when the height of the walls was large enough), the walls buckled to form a converging-diverging channel between them. The buckling occurs because of the residual stresses within the silicon-oxide layer on top of the walls (if the silicon-oxide were not fixed to the silicon wafer, it would expand to a larger size).
The novelty of this structure is that it was originally created using Manhattan (parallel or perpendicular) geometries, and it is now curved. Creating curved structures on this small scale is typically limited to expensive non-batch-processing techniques. Further, the final critical dimensions of the channel which was created (the minimum channel width is around 1-micron) are much smaller than the original critical dimensions that were patterned on the wafer. Creating sub-micron features using inexpensive, time-efficient photolithography techniques could prove useful in a variety of applications.
These structures were fabricated in the University of New Mexico Manufacturing Training and Technology Center (MTTC).