What I Learned Today

High-speed photography can capture athletes in action. But you need really high-speed photography to capture events like a nuclear explosion.

At the recent National Radio Science Meeting, I first encountered the idea of a rapatronic camera. These cameras have exposure times as short as 10 ns. They were developed in 1940 to capture the rapid expansion of a nuclear explosion, and they were gems of ingenuity. No mechanical shutter at that time could possibly open and close that quickly, so Harold Edgerton came up with a non-mechanical way of controlling the shutter: he put a Kerr cell between two polarizing filters oriented at 90 degrees from each other. Normally, no light would penetrate between the crossed polars. But when voltage is applied to the Kerr cell, it rotates the polarization of the incoming light 90 degrees—permitting it to pass through the second filter. By only activating the Kerr cell for a very short time, you obtain an ultra high-speed shutter.

Here is Edgerton’s circuit diagram (click to enlarge):

Likewise, there was no way to mechanically advance the film fast enough to permit a single camera to take a sequence of high-speed shots, so in these tests they’d set up an array of the cameras, each with a slightly different delay. (Although I immediately wonder if you couldn’t have an electronically controlled refractive material behind the single lens to direct the light across a series of film segments so you wouldn’t have to physically move anything.)

The results are stunning:

At left is an explosion from Operation Tumbler-Snapper (1952), about 1 ms after detonation. The spikes along the bottom edge are evidence of the tower’s guy wires being vaporized by associated gamma rays. At right is an explosion from Operation Hardtack II (1958). This one was suspended from a balloon and the spikes here are the balloon’s mooring cables being vaporized. A beautiful ghostly array of such images is available from a google image search on “rapatronic”.

One thing I haven’t been able to determine is the etymology of “rapatronic”. It may be that Edgerton just coined it (with “rapa” for “rapid” and “tronic” for “electronic”—but that’s just a guess). Please share if anyone knows more!

You can read more about Edgerton and his various innovations aside from the rapatronic camera. Brilliant guy!

I had no idea what a broad range of topics the field of “radio science” covers. I recently attended the National Radio Science Meeting in Boulder, CO, to talk with other researchers about the latest advances in radio astronomy data analysis (e.g., hunting for pulsars). Other topics in the multiple parallel sessions included lightning detection, antenna design, remote sensing of rain, “biophotonics”, “metamaterials”, space plasmas, and “telemetry for monitoring and biosensing”. Intrigued by some of the talk titles, I attended one of the latter sessions.

One goal of this field is to develop and test low-power, efficient radio communications for implantable medical devices (IMD). One envisioned application is for people in very rural areas who don’t have regular access to a doctor. Internal sensors could monitor blood pressure and various nutrient levels, then report them to an external base station they could visually check. As one presenter imagined, “Low potassium? Push a button and find out what you should eat for the next week!”

The devices are still under development, and in the initial work they’re focusing on the ability to monitor blood pressure. They aren’t yet up to human trials. Researchers from Texas A&M and Mississippi State University described how they’d started with rats. They showed pictures of the rat surgeries needed to implant the tiny antennas and then described the experiments, which aimed to evaluate whether the simulated response from the antennas was the same as what was observed when it propagated through rat muscle, fat, and skin. Unfortunately, the presenter noted, they’d been forced to euthanize all of the rats after the tests, because they hadn’t coated the antennas with a “biocompatible material” and therefore by animal testing rules they could not let the animals live. (It seems odd to me that this oversight would not have been caught during the protocol review process!) At any rate, the results showed a not very good match between the simulated response and what they actually got, which they attributed to differences between human skin (in the simulation) and rat skin (in reality).

As a side note, I kept wondering if these tests really qualified for the “in vivo” term the presenters applied, since the rats went to sleep for the surgery and (presumably) never woke up. The point at which they were euthanized was never specified. I started wondering whether live fat/muscle/skin tissue has different dielectric properties than dead tissue, which I assume it must, since circulating blood probably affects any signal propagation. This particular experiment seemed perfectly designed to test both cases. But I wasn’t quite up to asking this question after the talk.

The next presenter (from the same group) continued on to describe their subsequent experiments with larger animals (pigs). Pig skin apparently is a much better match to human skin (insert obligatory “white meat” joke here), and they got an excellent match with their simulation. In this case, they used a proper coating and the pigs were permitted to live. The presenter also commented on how very expensive these particular bred-for-experimentation pigs are (about $10,000 each), although I had to wonder whether one must purchase an entire pig to do a radio antenna transmission test, or whether one can give it back afterwards to be used for other experiments, or possibly time-share with other researchers. But again I wasn’t actually able to ask a question, being more sort of transfixed in a rather distasteful fascination and slightly nauseated by all of the graphic surgery images!

These talks didn’t spend much time on other important IMD constraints, like where power for the wireless transmitter comes from and how to dissipate the excess heat generated without cooking the animal (or human) internally. They noted that the devices had a 25 day lifetime if in continuous use, or 1.7 months if only transmitting periodically, so I’m guessing that limit was based on some nonrenewable power source being exhausted.

Overall, the envisioned future of such devices is certainly promising—and I was kind of disappointed to see how premature such investigations apparently are (if this represents the state of the art). I would also have liked to hear more about the kind of technology used for the sensors that collect the data to be sent by the antennas!

As kids, I think the first encounter most of us have with the idea of an “engineer” is “the person who drives the train”. By the time I started my undergraduate studies, however, I knew that the College of Engineering wasn’t just about driving trains. But I always wondered how a purely functional role could have the same name as what I now saw as an engineer: someone with a very active role in design and problem-solving. Or as wikipedia puts it, “An engineer is a professional practitioner of engineering, concerned with applying scientific knowledge, mathematics and ingenuity to develop solutions for technical problems.”

Recently I picked up some books at the library on steam locomotives and other fascinating train topics. And suddenly, I realized why the train-driver could also lay claim to the title of engineer. Historically, at least, the railroad engineer did not just drive the train. He (these were generally men) also had to be a top-notch engineer, familiar with all of the inner workings of his engine, as he was continually required to maintain, lubricate, tend to, and sometimes even repair the engine. One book noted that the “iron horse” was just as temperamental and required as much attention and grooming as the organic horse it had replaced.

Today, it seems we have a very different view of machines and their users or operators. Most of us who drive cars do not expect to have intimate knowledge of how they work; instead, we hire car mechanics to deal with the details and fix problems. No doubt today’s train engineers are also much more removed from their engines than those of the 1800′s. Computers likewise (or our attitudes about them) have evolved as well, so that users need not understand operating systems and file systems and network protocols, disk scheduling and memory allocation and pipelining, kernels and shells and scripts. Instead, one can hire an IT expert to maintain the machine and fix it when it breaks.

Is this trend a result of the growing sophistication and complexity of these machines, or a shift in our social attitudes towards the desirability of being involved in details? Or both? If it’s an evolution in the machine, are there other machines out there in their infancy and still requiring that the engine be a part of the engineer?

I can see the allure of both the low and high levels on this abstraction spectrum, for computers. I personally enjoy tinkering with the configuration of my computers and knowing what’s going on under the hood, even up to spending hours buried in an ancient Linux machine to get a wireless card working—but sometimes that loses its appeal when I just want things to work. And in truth, when the machine is reliable enough that I don’t need to be checking and tinkering constantly (as with my Mac), I’ve found that I stop doing it. Yet I still feel a tug of curiosity about other machines as well—I’d love to take a basic automotive mechanics course, and learn more about how trains work, and I’m fascinated by pulleys and linkages and astoundingly clever machines of all kinds. But then, I’m an engineer by inclination—or perhaps, I have a wish to be (in a friend’s charming coinage) an “engine-er”: someone who knows and tends and cultivates the engine, akin to a farm-er or a sail-or.

I always thought resistors had a certain chic appeal, with their banded color-codes. And so, during a recent trip to Radio Shack, I decided to pick up materials for making some ultra-geeky jewelry. It turns out that Radio Shack does still sell racks of electronic components, including resistors and capacitors.

Since I’ve never made any jewelry before, I appealed to a friend for a jewelry-making lesson. She’s made lots of her own necklaces, earrings, and bracelets, and showed up with books and magazines full of instructions and inspirations. She also brought lots of useful tools, mainly in the form of several different kinds of pliers and cutters. After playing around with the components a little to decide how to arrange them, we got to work.

I learned how to shape the wires coming out of the ends of the components into little rings so they could be connected. I gripped the wire with the rounded pliers and used another pair to tightly wrap the wire into a circular shape. After snipping the end of the wire, I used the pliers again to bend the sharp, free end in towards the component so it wouldn’t poke or snag. While the tools are great, I can see that it takes some experience to get used to using them most effectively (I kept wanting to put them down and work with my less precise but more familiar fingers instead).

My jewelry-crafting friend came up with the clever triangular arrangement and donated the hematite cubes that serve as the join point. The cubes didn’t have a hole large enough to pass both wires through, so we put one through and twisted the second one around the first, which was the most technically difficult part of this assembly.

We determined that the wires attached to the resistors and capacitors served quite well as jewelry wire, although apparently it is somewhat more flexible than is typical. However, I wouldn’t want to stick it in my ear (resistance wire is made from a nickel/chromium alloy which could irritate the skin of those with a nickel allergy), so we connected the earrings to sterling silver earwires.

Those are 1-μF capacitors. I wonder how many of my readers can specify the resistance of the resistors? (As a reference, here’s the resistor color code.) And in case you’re curious, here’s more info about how resistors actually work or how capacitors actually work.

Other cool ideas:

Tonight I was lucky enough to get to watch the International Space Station fly by: a bright unwinking point arcing upward and then, just past zenith, disappearing as it passed into the Earth’s shadow. This sight somehow never fails to stir something inside me. It is one thing to read news articles about spacecraft in orbit, and quite another to see them with your own eyes. Beyond that, the ISS marks our ability specifically to maintain a human presence in space—quite a bold and amazing feat, no matter how many years have passed since the Apollo missions.

One way that we get humans (and equipment, supplies, etc.) from the ground and up to the ISS is via the Space Shuttle. The Shuttle is a marvel of engineering—perhaps too much of a marvel, rendering it less reliable than we might hope—and it is in the sunset of its career. 2010 will see the final Shuttle flights (just five more are planned), and then the Shuttle will be retired while effort and funds are focused instead on developing its successor, the Orion spacecraft. During the 5+-year gap between the Shuttle’s retirement and the first crewed Orion flights, we will rely on the Russian Soyuz spacecraft to reach and resupply the ISS.

There are many fascinating engineering details about the Shuttle and how it works. I recently came across one that really surprised me. It turns out that when the Shuttle de-orbits and descends through the Earth’s atmosphere to land, it manages that whole long glide and deceleration without any engine power. Effectively, it aerobrakes (taking advantage of our thick atmosphere), and attitude control is maintained through thrusters and hydraulically actuated surfaces. This isn’t enough to slow it all the way down, so the Shuttle also executes several S-shaped swoops to the left and right, dissipating speed horizontally instead of vertically.

You can watch a Shuttle landing from the cockpit view (STS-98): fascinating, although I guess the video starts after the four swooping banks have already been done. The dramatic right turn you see in the video is the commander’s final lineup with the runway—a final turn that also reduces speed and altitude.

The next shuttle to launch will be Endeavor (STS-130), on February 7. It will be the 32nd shuttle mission to the ISS.