March 17, 2006

Funded by the National Science Foundation Office of Polar Programs

Location: Latitude 63° 20.3' S, Longitude 52° 22.4' W

Air Temperature: 1.3°C

Flexibility

You have to be flexible on a ship at sea: the nearest Radio Shack is several thousand miles away. Though everyone tries to have spare parts available, it is impossible to stock everything or predict every potential failure. A broken part could render an expensive scientific instrument inoperable, or at very least, inefficient. Such was the prospect yesterday when a laser switch stopped functioning in the Multi-Sensor Core Logger (MSCL).

In brief, the MSCL is proof that man's principle advantage over beast is the consistent quest for laziest solution - the easy way to do things. The MSCL combines a number of instruments for core analysis in to one highly-automated machine, allowing one or two people to run a set of measurements that might otherwise require a small army. The instruments include a gamma-ray impedance detector (for density) and a high-resolution camera (for core imaging). The instruments are next to a track of rails on which the core sections are placed. Each section is pushed past the instruments by a motorized belt-drive pusher. A laser beam-break switch is used to sense the beginning and end of each core section, so that the measurements may be accurately correlated to position in the core section and thus, to position below ground. Fortunately, all of the instruments have been working well. Unfortunately, the sensor part of the laser switch completely stopped functioning yesterday.

After a time spent checking things on the software side, I called in the big guns, in the form of Dan Elsberg. Dan is an electronics tech, and knows circuits much better than I do. He pulled out diagrams and cable pinouts, and we began to troubleshoot the hardware-software interface. Everything worked right except for the laser sensor - it did not produce the expected electrical change to drive the rest of the logic circuit and signal to the computer that the beam was "unbroken." Without this signal, the software never has a starting point, because the beam is supposed to only be broken when something - usually a core section - is in the light path. Figuring out the problem was a good start, but only part of a solution. We needed to find a replacement . . . Thus began our parts odyssey.

First, to the parts bins in the electrical shop - nothing. Next, to the aptly named "hobby room." This room is a large walk-in closet filled to capacity with drawers and shelves lined with bags, boxes, and bins containing parts for all sorts of instruments, the computer systems, sonar, radios, TVs, and other assorted systems that we run. But alas, no luck. We did not find anything close to the desired phototransistor.

At this point, we decided to resort to cannibalism. For a while, we discussed stealing parts from an old TV-VCR combo. But first, we took apart a broken palm pilot, several floppy disk drives, as well as raiding a box of printer parts. Nothing. Then, a little creature came to mind - the mouse connected to each computer. In the process of computer upgrade and replacement, we have managed to end up with a large collection of excess computer mice - especially the fancy new optical ones. We took to one of these with screwdriver and soldering iron, to see what we might find. Disassembly showed promise, as the mouse had two light sensors inside of it: one for motion, and one for the little scroll wheel between the mouse buttons. The motion sensor was built in to a complicated microchip that would never be useable without schematics. But, the scroll-wheel sensor looked like it just might work. Using a voltmeter and a flashlight, Dan checked to see if the sensor worked as hoped. It did! But, when we tested the sensor with the original laser, a letdown. Laser light consists of photons at a single wavelength (from a practical standpoint, color). Though sensitive to some of the many wavelengths from a flashlight, the new sensor was not sensitive to the specific wavelength of the laser.

This was a roadblock, but we realized that the sensor might work as needed with a different light source. Though finding one should have been simple, nothing except the flashlight would work because the light needed to be focused properly. So, we decided to try running the system with a flashlight. This worked for a while, but later in the day, the system was switched to a microscope light fixture. Light from this source is more focused, and unlike the flashlight, will not require new batteries every few hours.

The attached pictures may not be aesthetically pleasing to most. To a tinkering soul however, the solution borders on art.

Most importantly, the logger works!

-Isaiah Norton

Do you have questions? Comments?
Contact us