Cloud Computing

31 12 2008

I’m getting a crash course on cloud computing today… I’m moving a client’s ruby on rails site to Amazon’s Ec2.

Ec2 is useful.

My plan is to live stream all experimental data (including audio/video) to an Ec2 cloud, where it can be hosted, RSSed and analyzed. A realtime community interactive experimental workflow. I would love to post data, and have _someone else_ draw an insight from the data produced. And I’m sure it will produce lots of data, once we get rolling.

Ec2 also opens up the possibility of provisioning 10,000 servers for a day… when you really need to crunch some numbers.





Bloom Filter

30 12 2008

File this under tools I bet I’ll need:

Bloom Filter in Ruby

Excerpt:

A Bloom Filter is a probabilistic data structure that is used to test whether an element is a member of a set, or more simply, it’s an incredibly space efficient hash table that is often used as a first line of defense in high performance caches.





New Tools

30 12 2008

Gotta love mcmaster + next day air:

new_tools1

 

All metric tools. What a pleasure working in metric. Ball point hex keys, metric wrenches, metric digital caliper, metric ruler, metric drill bits, metric tape measure, and some files and wire cutters.





Finite Element Analysis

30 12 2008

We need to nail down calculation for chassis strength under the load of the electromagnets. Ideally we would want full finite element analysis. I found an open source package called OpenFEM, which depends on the also open source SciLab. Scilab looks generally useful for this project. 

Please leave a comment if you know SciLab. It really helps to have a guide when learning new software… this stuff is all new to me. 

I did some crude calculations for the force from the coils. Now we need to calculate the strength of the chassis for various materials and designs.

The strongest material for SLA is BlueStone, here is the BlueStone datasheet. Tensile Strength 66 Mpa. Compare that to steel which has a tensile strength around 500 Mpa. Time to get a quote for Direct Metal Deposition from POM

Reading up on tensile strength I noticed the following: Tensile strength is an intensive property and, consequently, does not depend on the size of the test specimen.

 

UPDATE

List of Free Numerical Analysis Software. Treasure trove of free tools.

I’m looking for a way to import an STL file for analysis. 

CalculiX looks like its more directly useful. CalculiX supports STL geometry import export, see CalculiX Docs . 

With CalculiX Finite Element Models can be build, calculated and post-processed. The pre- and post-processor is an interactive 3D-tool using the OpenGL API.

These are geared towards physical modeling:  CalculiXCode_AsterFEniCSImpactSLFFEAYadeZ88





Hercules Extruder 3D

29 12 2008

Still cutting my teeth with sketchup. Here is the hercules extruder in 3D:

hercules_3d

 

 

The sketchup-svg-outline-plugin is great for producing laser cuts for Ponoko. I recommend this sketch-svg tutorial.

Added the trapped nut cutouts:

trapped_nutThis intersect model tutorial was essential for making the trapped nut cutouts.

I just ordered v1 of the extruder from ponoko:

herc_ponoko





burning air

28 12 2008

Before we worry about a vacuum, we can operate the magrid on air. If you look at the transcript for “Should Google go Nuclear” , on page 10 Bussard says of the WB2: “All these tests were run on air or argon”, accompanied by this picture:

burn_air

 

I think this means forming an electron cloud in the center of the magrid, despite presence of air. This sounds like a Non-neutral_plasma . I suppose you would operate the magrid normally… bring the shell to a high potential, turn on the electron beams, bring the magrid up. The streaming electrons ionize the air. The air ions rush to the center. Some of that energy is released as light. Even if it’s possible to burn air, it may take low pressure air, requiring a vacuum vessel and pumps.





scotch tape x-ray vacuum chamber bake-out

28 12 2008

I had an idea for a low cost way to quickly bake-out (dry out) a vacuum chamber.

At one point I read that the fastest way to bake-out a vacuum chamber was with x-rays (I can’t find the link). But assume for a second this is true. 

Now. Consider the fact that we can make x-rays very cheaply with scotch tape in a vacuum. And in our case, we are already operating inside a vacuum! 

Cheap,fast, high quality vacuum chamber bake-out.

It would look something like this:

scotch_tape





debug

26 12 2008

the chain is touching, I don’t think it should.
chain_touching

 

this is why you need Ball-Point Hex Set (part number 5709A18 at mcmasters):

ball_point_hex_wrench





Hercules Extruder

21 12 2008

I’m working on fabricating the Hercules extruder for the reprap. It is simpler then the current extruders, and it just looks sturdier and more reliable.  The lasercuts look like this:hercules_extruder

Haven’t tried these out yet. Here is the sketchup file





Superconducting Cable

21 12 2008

On Dec 4th 2008, I ordered 20 meters of 4mm YBCO superconducting cable from theva. The lead time is 8 weeks, so it hopefully will arrive at the end of January 2009. This should be more than enough cable to complete several basic tests of superconductor construction. I also want to explore the fabrication of a SMES device for high power, high speed lab power supply. We will need to devise circuits to charge and discharge the superconducting cable, I’m not seeing much information on this topic on the internets.  And of course this will require liquid nitrogen. I’ve always wanted to play with liquid nitrogen.








Follow

Get every new post delivered to your Inbox.

Join 633 other followers

%d bloggers like this: