Solid Modeling

There are many, many CAD and/or solid modeling packages out there. Higher-end packages include Pro-E, CATIA, and IDEA-S. Let's not forget some of the advanced packages from AutoDesk. I've used IDEA-S while an undergrad at Penn State, where you can see an image from one of my endeavors at right. I know of a good friend who has grown and loved Pro-E, and I've witnessed some of his work. (I was this close to obtaining use to the only known copy of CATIA at PSU in order to assist in my undergraduate thesis, but the fellows in the computer department were unable to get it properly working in time. I only got to see a couple of glimpses of it.)

Now, some of these have analysis packages included with it. IDEA-S in particular had a stress and vibration analysis package, which I used to test the harmonic frequencies of that truss shown here. A vertical vibrating load was applied to the end of the truss that was clamped to the vibrational source, and the opposite end was free. Other members of my stress group actually built a model and hooked up one of those oscillators that look a little like a juicer to the end of it. The results amazingly matched the model at higher harmonics. Lower harmonics differed from model to experiment, likely because we used a custom-made clamp that wasn't quite securing, and so the end acted more likely a simple-support instead of a clamped boundary. We had some great marks for the resultant paper given from an incredibly strict T.A. (who, incidentally, was the person who stopped the assailant of that shooting spree at PSU several years back.) To make a long story short, it appears that IDEA-S and its analysis software seemed to work reasonably well. The learning curve was not so steep given what my friend and other professors noted regarding Pro-E. Modeling may be initially difficult to grasp with Pro-E; I have not witnessed any analysis using Pro-E to provide an honest opinion of its analysis packages.

Before proceed briefly about some mid and lower-end software packages, let me note that there are modeling packages that are more built on their analysis tools rather than their limited construction tools. Such an example that I've used (and currently using) is ALGOR. A colleague of mine had used ALGOR in the past and was annoyed generally with the level of bugs in the program. Second, I note that ALGOR does not have the ability to produce annotated drawings, at least not with the version I'm using. The reason we elected to purchase it was (1) it supported models from SolidWorks (described in a moment), and (2) it had a electrostatic solver package (NOTE: it presently does not have a MAGNETIC field solver). SolidWorks has an add-on called CosmosWorks, which obviously supported SolidWorks, but did not have an electrostatic solver. Anyhow, I didn't even attempt to model in ALGOR. The FEM meshing is very ponderous and slow for items of cylindrical geometry (rectangular items were fine, though). Last, the electrostatic solver konks if the units are set, say, below millimeters. Thus, you have to scale the model and invoke scaling laws. Results are O.K., but an oddly-produced mesh creates a lot of artifacts. Stress analysis is a lot smoother. New releases had some improvements with the ES solver and was faster. I'll try to put a screenshot of an analysis soon, avoiding showing their symbol and infringing on any copyrights.

Mid- and lower-end software packages (<$10000) that I strongly know of or have used include AutoCAD, SolidWorks, TrueSpace, and DataCAD. I've also used a software package called SilverScreen at PSU (and even co-wrote a user's manual for it), but I don't believe they are in business anymore. AutoCAD is difficult to master in 3-D and I would not really advise using it solely for this purpose. It is much more suited for "paper space" models, although modeling is fairly accurate. SolidWorks is what my company and I use now. It has a very shallow learning curve, as a technician of ours noticed this as he migrated from AutoCAD to SolidWorks. It can handle solid rendering, annotated drawings, and animations. The aforementioned CosmosWorks package can be additionally purchased for analysis (though not EM as of time of this writing). There are some nuiances, though. First, one has to watch assembly models very carefully, for unlocked objects can move around if you move other objects. Second, I have seen occasions where the third or fourth significant digit can change, perhaps when upgrading the software. Beyond this and some limitation with animation tools, it's a respectable package. TrueSpace is nice for simple models and especially animations, but is not useful for annotated drawings. I used this at NASA Marshall Space Flight Center for presentation purposes, and AutoCAD for 2-D drawings, then moved over to SolidWorks by request of a friend of mine. I haven't touched DataCAD in over six years, but I recall that package is better suited for architectual engineers and the like. A rendered image from SolidWorks is shown at right.

In summary, recommendations for solid modeling packages depend greatly if you wish to perform analysis, just to create blueprints, or just to make presentations. TrueSpace is a quick, easy package for presentations only. SolidWorks w/CosmosWorks seem a reasonably mid-level package for blueprints and analysis, depending on the type of analysis.

At my current profession, I've been heavily involved with the Simulation and Flow Simulation packages offered through SolidWorks. Mostly I work with CFD and thermal analysis, often combined. I also perform stress, frequency, and random vibration and shock analyses. Rarely I investigate nonlinear analyses; our company has a dynamicist who investigates a lot of that stuff. (Forgive this aside, but I always ask why the name of these packages say the word 'Simulation' when they aren't in fact simulating anything? They are 'Analysis' tools! Maybe they don't want to use the 'Anal' word in their dropdown menus or something. :) )

Antimatter Physics

Antimatter sounds like a religious zealot's nightmare, but it isn't. It does not destroy all forms of life when anything touches it; rather, antimatter converts its equal share of matter it comes in contact with to high energy products. I better term for it is 'Mirror Matter'. We have positrons, or anti-electrons, that are just like electrons, but with opposite charge. These, incidentally, not only can be created by linear accelerators, but simply by obtaining a (radioactive) Na-22 or Ge-68 source. I.e. to a large degree, Nature already is making positrons. Antiprotons are opposite to protons. There can be antineutrons, too, but the application potential for these are weak. Annihilation of a positron and an electron creates two 511 gamma rays, and a 1% probability of creating three gramma rays with total energy less than 1.02 MeV. Annihilation of antiprotons and protons created gamma rays plus charged muons and sometimes other residual heavy hadrons.

NASA Marshall is interested in developing antiproton storage vessels because the products from a pbar-p reaction create charged muons that can be expelled to produce thrust. This reaction can offer the highest specific impulse known to man. My undergraduate research project at Penn State University was to apply antimatter microfusion/fission (AIM) to a deep-space mission to the Oort Cloud at 10,000 AU. Here, the pbar-p products hit a trace amount of uranium coating a solid pellet; the ensuing plasma from the fission reaction is compressed to fusion state. Such a project is rather ambitious at this point. My graduate thesis involved a 'test' of the capabilities of antiprotons by having the fission fragments hit LiH propellant, which is not compressed to fusion pressures; instead, it is simply expelled as thrust. The results I obtained was a specific impulse of 3500 sec and a thrust of 42 mN (for 5*10^11 pbars), which can be measured and is actually comparable to ion engine technology. NASA Marshall is developing a trap called HiPAT (High Performance Antiproton Trap) that could feasbily store this number of antiprotons.

Our company looks at positrons (e+) since they are presently much easier to produce, and the resultant products produce no residual radioactive contamination (antiprotons do). In other words, positrons can be used for endo-atmospheric flight. We examined how many positrons can bring a Single-Stage Reusable Vehicle up to M=8 to a final rocket burn; less than 20 mg of e+ is necessary. We've also examined positrons to replace Nuclear-Thermal Propulsion (NTP) to get to Mars; also, less than 20 mg is necessary. Some applications such as propelling a defense missile to longer ranges or sending a 1 metric ton payload to orbit may only take 1 mg, which is in a realm of producibility right now.