The BOOMERANG Telescope
(Balloon Observations Of Microwave Extragalactic Radiation ANd Geophysics)

A Balloon Borne Design

Our telescope is designed to measure microwaves coming from space. Unfortunately, the water vapor in the atmosphere is very good at absorbing microwaves. One way around this problem is to get the telescope above the atmosphere. Launching a satellite, however, is a long and expensive process. By launching our telescope attached to one of NASA's Long Duration Balloons (LDB), we can get it up to 120,000 feet and above 99 percent of the atmosphere at a fraction of the cost.

Antarctica is a very good place for high altitude ballooning. The air is very dry, and in the austral summer, a unique wind pattern arises at high altitude. These "polar vortex winds" circulate around the South Pole (-90 degrees latitude). When we launch a balloon from McMurdo (-78 degrees latitude), the balloon takes a roughly circular, constant latitude path around the Pole. So after two weeks, the telescope (hopefully) comes back around to McMurdo and we send a command to cut it loose from the balloon. It parachutes down and we go pick it up. Of course the flight path of the telescope gives us an entirely intentional pun on our name.


Many people are familiar with digital cameras which can be purchased at the local electronics store. These cameras use a detector called a Charge Coupled Device (CCD). The CCD's used in such cameras today often have millions of individual pixels. Some of the pixels are sensitive to red light, some to green, and some to blue. When you snap off a picture, the camera uses information from the three types of pixels to create a full color image of the visible light coming from your field of view.

We are interested in studying microwaves instead of visible light, so we need detectors that are sensitive to very different wavelengths than the pixels of a CCD. There are several kinds of detectors that people use to study microwaves, and they are all very tricky to work with. Our experiment uses detectors called "bolometers". A bolometer roughly consists of a small bit of material whose temperature depends on how much light you shine on it. A bolometer also has a sensor to measure the temperature of this material.

Our telescope has 16 bolometers arranged into 8 pixels, a far cry from the millions of pixels in a digital camera! The microwaves we are looking for are very "dim" (low intensity), and on top of that, we are looking for tiny fluctuations in the intensity. This means that our bolometers must be extremely sensitive. Another way of putting it is that we need to be able to detect tiny changes in the temperature of the bolometers, which corresponds to tiny changes in microwave intensity.

The changes in the temperatures of the bolometers are so small, that to be able to see them, the bolometers must be kept very cold. That way, the changes in temperature of the bolometers due to different microwave intensities are easier to see compared to the bolometer's "normal" temperature.


The bolometers in our telescope are kept at a fraction of a degree above absolute zero (the unattainable theoretical temperature at which all molecular motion stops). We usually refer to temperatures using the Kelvin scale, where absolute zero corresponds to 0 Kelvin. For reference, 0K = -273C = -459.4F. In normal operation, our bolometers are kept at a temperature of approximately 0.3K, and changes in their temperature due to incident light are measured relative to that.

There are several stages to the cooling process. First, the outer dewar of the cryostat is evacuated so that there is no heat conduction by gas between the various parts of the cryostat that are at different temperatures. Next we cool everything to 77K with liquid nitrogen. After that, the inner tank is filled with liquid 4He at a temperature of 4K. This cools the detectors along with it. The inner stage is a closed cycle 3He refrigerator that cools the detectors down to 0.3K. Here is a rough sketch of these different stages:

A Picture of the Sky

At the end of the day, we want to make a picture of the microwave intensity over a patch of the sky (actually several pictures at different microwave "colors"). How do we do this with a "camera" that only has 8 pixels? We accomplish this by scanning our telescope horizontally back and forth. Imagine this: you want to see what's on the other side of a fence, but there is only a narrow crack between the vertical slats. You probably will find yourself moving your head from side to side trying to get the "whole picture" of what is on the other side. In the same way, we can rotate our telescope left and right and build up a picture of the sky. The sky is rotating from our perspective, of course, so this is not quite so simple...