Badeschiff
In water too polluted to swim in just make your own little crystal clear pool submerged in it. I need to build one of these.
Some really awesome photos of a SpaceX hangar over at the Wired Science blog. They provide a neat glimpse of the large behind the scenes effort required to do actual rocket science on a large scale. A lot of precisely engineered metal alloys has a lot to do with it you would gather. SpaceX is doing some critical work in bringing space a little closer to earth for those without a NASA-level budget behind them, which is mostly everyone. Their newest, biggest rocket, the Falcon 9, launches this year.

More here. I so need a hangar.
I’m entrepreneurial so making a business out of what I enjoy doing would be ideal, and am working towards such a goal. Exactly what kind of business would that be though? Perhaps a kit business described in slides over at Adafruit’s blog, who obviously would know a thing or two about this (having done so). It doesn’t sound so hard from those 17 little slides.
NASA may have made these for purely functional reasons but they look incredibly cool. Over 200 custom tools were made for the Hubble Telescope repair mission. The one on the left above is generically called the “pistol grip tool”, which is a self-contained, high-torque drive, the tool features an on-board computer. Spotted here as well.
I’m building a quadrotor UAV, which is like a helicopter except it has, you guessed it, four rotors. The advantage of this is a much simple control scheme for the motors and rotors, both electrically and mechanically. Just by varying the power output of each motor you can do any maneuver, with software figuring out exactly what to do. Below is an initial, simple model rendering using Google SketchUp, my go-to tool for quick and easy modeling. Doing a rough model is great for getting an idea of what components you’ll require and how they’ll fit together and to see how it will look from any vantage point. My goal is for a payload of around a kilogram, enough to get a camera airborne for some photography from a rarer perspective.

For a circuit I’m working on it would be preferable to use a 3.3v supply for a microcontroller, as it interfaces with a 3.3v sensor, and the uC in question needs to run at 12Mhz. The problem: the uC, the Atmel ATtiny88, is not specified to run at 12Mhz off a 3.3v supply. Solution: overclocking, with the hope that it won’t cause any critical issues. We’ll see.
The probability of it not working is dependent on how much you overclock the uC. Below is a run through of the simple calculations to figure out roughly how much the uC would be operating out of it’s guaranteed operating range.
You’re given the following speed grades with their voltages in the datasheet:
My voltage, 3.3v, is in between so we calculate the corresponding frequency that 3.3v could operate at according to the manufacturer, based on a linear relationship between the two data points given (note: note linear outside these points).
Basic line math, you know, slope and stuff:
12Mhz – 8Mhz = 4Mhz
4.5v – 2.7v = 1.8v
4Mhz/1.8v = 2.2 Mhz/v
3.3v – 2.7v = 0.6v
0.6v * 2.2Mhz/v + 8Mhz = 9.32 Mhz
Quite a bit short of the 12Mhz I was looking for. I’d be overclocking it 29% which is significant. But knowing that this project is not a demanding application and that manufacturers (usually) underspec parts it just might work. Or it might seeming work fine in the lab only to fail sporadically in random ways in the field rendering the unit effectively garbage.
I’ve been giving the new Wolfram|Alpha a spin and like what I see. It’s not your regular old search engine and doesn’t serve that purpose well as many have pointed out. And it’s not supposed to as it’s more of a computation engine, and a powerful one at that. From the makers of Mathematica you would expect some sophisticated calculation ability and they deliver.
The best way to see what it can do is to simply play around with it by entering various inputs and seeing what it produces. As an engineer I can anticipate myself using it daily for the routine calculations, technical details, material properties and so on that would take me longer to obtain otherwise, and speed is crucial to doing innovative things.
It’s definitely an innovative technology and product but it’s hard to see what it’s impact will be in the future, whether it will rise beyond the user base of technically-minded people to something nearly everyone would use.

The forth in the series, ‘Terminator: Salvation’, was incredibly action packed and rather exciting, i.e. exactly what a summer movie should be and hopefully there’s more coming this season. I’d give it my two thumbs up to all of my readers around the world, it’s worth the $8 to go to the theater and see it and be entertained. However, I do have some quibbles with the whole series though, major and minor, but I’ll just explain the major one which kind of makes any others moot.
Okay, if a Skynet-like entity launches an attack like seen in the third movie, obliterating much of the world in a nuclear holocaust, that would have specific consequences that would be hard to avoid. First off, whatever technology and resources humans had available to fight Skynet would be just what survived the nukes. No new technologies could be developed due to the whole R&D infrastructure gone, and hardly any new weapons would be produced with factories shut down. And modern war requires a tremendous amount of resources to back it up from supply lines, to ammo, logistical support, etc. This would make fighting any enemy that was able to do all of the above pretty much impossible for any extended period of time, unlike the movies’ long running war. Humans would be quickly annihilated.
Although I just finished up the semester I’m not ready to move on quite yet from the infrared motion tracking project started earlier this year for a senior embedded design class. In fact, most of what I want to do remains to be done.
My plans include:
That’s actually a lot of work so be patient, it could take a few weeks to figure it all out during my summer break. Tentatively I would expect by June 15th to have every little detail worked out. I may regret saying that, but I’m optimistic in general and optimistic knowing that I can get it at least working acceptably, having done so already. So stay tuned as I’ll be doing a lot of work on this and frequently updating my loyal readers on my progress.
I think it just looks cool but technically it’s a “direct numerical simulation of a lifted autoigniting ethylene-air jet flame… volume rendering of formaldehyde (blue/green) and hydroxyl radical (red/white) denoting the location of preignition upstream of flame and lifted flame, respectively.”

Found while browsing the NCCS site while researching the fastest computers in the world which this simulation was run on, Jaguar. It’s no wonder my engineering simulations take so long and look so less impressive.
In an article simply made for me Spectrum makes a list of 25 microchips that shook the world. It’s an interestingly list alone and the stories behind the chips make it even better. I recognize most of them, having worked with a number of them which is actually quite astounding as they are mainly old designs. This is another illustration that good engineering can stand the test of time, as well as once you find something that works well you use it, becoming familiar with using it designs, and use it again and again down the line.
The complete (and unordered I think) list:
The ones I’m most familiar with are the indispensable 555 timer/astable multivibrator and the 741 op-amp. It would be shocking if they didn’t make the list. I’ve used both this year, despite them being on average four decades old. The 555 was just the trick for a XBox rapid fire button. And the 741 is my standard starting point for any analog op-amp needs. In college I became acquainted with ARM processors which make up a majority of the processors in the mobile world and the flexible Xilinx FPGAs.
Overall it’s a very representative list of the different things you can do with an integrated circuit: DSP, timing, communication, memory, analog, imaging, waveforms, speech generation, audio playback, micro and parallel processing.
I’m not sure if they missed any, although they do offer another group that didn’t make the cut. But if I did have to name one it would be the 595 shift register. I use them a lot and see them used widely for the common need to expand serial data to parallel, which comes in handy to expand the output lines for microcontroller projects.