The Tip-Tilt Camera

Digital Cameras

Light waves consist of "packets" called photons which are focused by the telescope into the camera. A digital camera records this image as a rectangular array of numbers. The array, or charge-coupled device (CCD), is a detector composed of a grid of tiny sensors called pixels which are each able to sense photons at an efficiency of nearly 100 percent. This is far better than the eye or photographic film. In addition, CCDs have a linear response and a large sensitivity range. That is, the signal produced is proportional to the number of photons detected. Furthermore, the CCD is able to simultaneously distinguish very bright and very dim objects.

A CCD works in the following manner. Photons from the observed source strike the detector and excite electrons. The detector then collects these electrons and records a signal. The number of electrons collected in each part of the detector corresponds to the intensity of the image at that location.

Tip-Tilt Correction

An AO system corrects for the distortion of images caused by turbulence in the atmosphere. There are three parts to an AO system:

1. Wavefront sensor
2. Wavefront corrector (in our case, a mirror that tips and tilts)
3. Wavefront control computer

The wavefront sensor measures the distortions in the wavefronts coming from a bright guide star near the target object. In our system this is done by rapidly taking pictures of a 7th magnitude or brighter star with a digital camera. This is done many times per second. These images are read out by the control computer, which finds the center of the star in each and finds out how many pixels it has moved since the last picture. Since the computer now knows how far the star has moved, it sends a signal for the tip-tilt mirror to move slightly in the opposite direction in order to move the star back. Actually, in most AO systems the mirror is placed in front of the wavefront sensor, creating a feedback loop. In this configuration each step can be smaller because it is only an incremental improvement on the last correction. This is called a closed-loop system, and the SBIG tip-tilt camera operates this way.

The wavefront corrections need to be done in a small fraction of a second due to the rapid changes in the Earth's atmosphere. The track time is how long the wavefront-sensor camera exposes for each picture. It can be as short as 0.001 second for the SBIG system. It then takes some time for the computer to read in the picture, find the center of the star, and send a correction to the mirror. The aggressiveness is how much the tip-tilt mirror moves during each corrective step. Too low an agressiveness will make the the mirror move very little and almost no correction will be made. Too high an aggressiveness will cause the mirror to move too much and the system will oscillate wildly. The bandwidth refers to the frequency that the system is able to perform both the measurement and corrective steps. A higher bandwidth results in better images. The SBIG camera can achieve bandwidths as high as 50 times per second.

Scientific Observations

The SBIG tip-tilt system actually has two digital cameras. The first is used to follow the motion of the guide star and allow AO correction with the tip-tilt mirror. The second is used to take the AO-corrected picture of the scientific target. Both are controlled by the same computer.

On the right are two images of the same binary star taken with our system. Binaries are two stars which orbit about each other. Without AO it is very hard to be sure that there are two stars (upper right), but with AO correction much of this blurriness goes away, and the two stars are easily distinguishable (lower left). We can measure, in arcseconds, the separation of the two stars. If we know how far away the binary is, we can use a little geometry to determine what the physical separation of the two components are.

Go home.