December 13, 2009, South Pole Station –
They Also Serve
My turn at Pole will come in late January, when I will be doing the “close out” work on IceTop. This consists of removing equipment from the frozen tanks and inspecting it. I will fix any obvious damage to the best of my ability and send anything I cannot fix “North” for repair during the off-season. I will also pack everything carefully away, produce an inventory, and generally clean things up so that the crew next season can get the fastest possible start.
For now, however, I am playing a support role as the “expert” on the hardware and software of the freeze control system. Being at Pole can be very stressful, with many diverse tasks to complete simultaneously. The effect of the low oxygen content of the air is also a constant annoyance when it comes to concentrated thinking. It is therefore most vexing to be faced with a technical problem that does not yield to immediate analysis. In such a case it is very reassuring to have someone in the “North” who is committed not only to working on the problem but putting it at the top of his or her priority list. The person at Pole can concentrate on the alligators while the person in the north can concentrate (while breathing real air) on the strategy of draining the swamp (or maybe vice versa).
Brave little tailor.
As an example, I recently helped to identify an obscure problem that resulted when the operating system of one of the computers started a second, hidden copy of the main control program. This caused all sorts of strange and seemingly unrelated problems, but finally we were able to deal with that particular alligator for good.
As a help in this process, many of the diagnostic parameters from the freezing tanks are automatically transmitted “North” and displayed on a Web site. This saves a lot of time in many cases -- including simple curiosity about what is going on. The Web site is publicly visible, but the displays assume one knows a lot about the system and are not very “user friendly.” Even explaining most them would take too long for this blog.
There is one exception, which can be found at: http://www.bartol.udel.edu/icetop/freeze/monitor10/history/
With a little explanation, one can literally watch the tanks freeze. The data in these plots come from two sets of temperature sensors, arranged from top to bottom (with about a six-inch spacing) The photo shows a row of sensors installed on a tank before the foam insulation was added. These are actually the sensors on the “near side” -- the side nearest the freeze control unit housing (white box at right). There is another set diametrically opposite, termed the “far side.” Because of the proximity to the heat generated by the freeze control unit, the “near-side” sensors tend to read slightly warmer than the “far-side” sensors. The far-side sensors therefore tend to reflect the water temperature in the tank a little better, so I will focus my discussion on them.
At this time (December 12, 2009), this year’s tanks are just beginning to freeze, so the “screen shot” above shows the time history of the far sensors on tank “06A” as of December 23 of last year. Each sensor is shown as a series of symbols with different colors and shapes. Initially, all of the sensors read about –25°C (roughly the ambient air temperature) when the tank was empty. When the comparatively warm water was added (about 12/04), all of the sensors jumped up.
Note that the red crosses do not get as high as the others because this sensor is actually above the liquid level and is measuring the air temperature at the top of the tank. Initially the tank lid is closed, so the air above the water is kept warmer than the outside air. The tank is left for a few days with the lid closed while the system that removes dissolved air conditions the water. The water temperature also drops slowly toward freezing. Trying to freeze the water when it is too warm produces a layer of poor-quality ice at the surface.
Fastest gun in the South.
About 12/07/08, the tank was ready to start freezing the water. The lid was opened and immediately the air temperature dropped. By this time, a sunshade had been installed to keep this temperature as low as possible. Almost immediately, the green “X” symbols begin to separate from the others and go below zero centigrade as the surface layer of water begins to freeze. As the layer of ice gets deeper, the blue and then the magenta symbols separate off, to be followed in days to come by the other traces. Note that there is a temperature gradient in the ice, but the water remains at a nearly constant temperature near freezing.
When I get there in late January, I will be watching these plots closely. When all of the water is frozen, indicated by all of the traces being well below freezing, we will pour insulation into the top of the tank, close the lid, and then remove the freeze control unit.