home latest pics nebulae clusters galaxies equipment information


E Q U I P M E N T
coldbox



coldbox
Emiel Kempen pointed me to the site of Gary Honis, who showed on his site how to make a coldbox for a DSLR.
A coldbox is designed to deminish the amount of noise by cooling the camera. There is a relation between the amount of noise and the temperature of the chip. I cannot measure the temperature of the chip itself, but I measure inside the coldbox with an electronic thermometer. In the past I cooled my camera with a coldpack and my experience was that it takes about 30 minutes before the cold reaches the chip. In the coldbox a peltier element is used to cool the camera. After one hour the camera is cooled about 10 to 12° C below the ambient temperature. I think it is possible to reach more difference between inside- and outside temperature by perfecting the coldbox, but I 'm just a sellotape engineer.

WARNING: I discovered that my Canon 40D (in the coldbox) stopped functioning at -16° C.
But don't be afraid, after warming up the camera works as usual.
The amount of noise can be meassured by determining the standard deviation of an image. To show this relationship I calculated (with ImagesPlus) the standard deviation of several dark frames of 10 minutes at 1600 ISO at different temperatures. A dark frame is an exposure with the cap on the telescope. The relation is not a straight line, because the circumstances are not always the same. The wind is important, the time between the exposures, the use of liveview, or my impatience to wait long enough, are all factors that can influence the results, but the relation is obvious. The lower the temperature, the lower the standard deviation, the lower the noise in an image.
standard deviation
The temperature of the chip is not the same as the temperature inside the coldbox. Some "heat" is coming from the telescope. I use a focal reducer or IR blocking filter between the camera and the air in the telescope, but still some heat is coming into the camera and the chip. In the winter of 2008/2009 I took dark frames without a coldbox, but not enough to draw a line. I think the net effect is about 6 or 7 degrees. When it is cold (below zero) the effect of a coldbox is less than when it is warmer. The effect is not a straight line.