Home

CAMERA COOLING EXPERIMENT #5

This investigation shows that cooling the Canon 350D camera with a Peltier cooler results in much less noise.  The red line in the graph shows noise increasing with each new exposure in an uncooled camera.  The blue line shows what happens to noise when it is placed on a cooling plate.

Experiment A.  (Red Line in graph - No Cooling - The Control Experiment)

This experiment consisted of an uncooled camera being turned off and placed on a desk for several hours to be certain that it had come close to room temperature (26.9 'C).  The experiment started with a series of forty 2-minute exposures with room temperature at 26.9 'C.  There was a 10 second pause between each exposure.  The result is plotted in the red line on the graph below.  The text at the bottom of this page describes all the details of the experiment.

Experiment B. (Pink and Blue lines in Graph - Cooling the Camera)

The Cooling Experiment began with the camera being turned off and placed on a desk for several hours to be certain that it had come close to room temperature (26.5 'C). A series of 15 x 2-minute exposures were then taken with the uncooled camera.  Each exposure was separated by a 10 second delay.   The result is plotted in pink in the above graph and follows the red line closely - which is what you would expect because during this period the camera was not being cooled.

But during the 15th exposure the camera was carefully placed on top of a flat cold plate of a Peltier cooler.  Exposures #15 through #25 were then taken while the camera was sitting on the cold plate and cooling down.  The temperature of the cold plate between the 15th and 26th exposure was observed to be 21 'C .   The graph shows that there was an dramatic effect on the noise in the camera - reducing noise by approx. 400 units.

At the 26th exposure a thin insulating piece of paper that was under the cold plate was removed to drive the temperature of the cold plate even lower.  The temperature of the cold plate was observed to drop some more (and stabilised at approx. 18 'C after exposure #34).    Exposures #26 through #33 were taken while the plate was in the process of cooling down from 21 'C to 18 'C.  During exposures #35 to #40 the cold plate temperature stablised at around 18 'C.

Interpreting the Results.  

Conclusion:  Cooling a Canon 350D results in less camera noise being present in raw images.

a) In Experiment B (see pink and blue lines on the graph) you can see the transition point (at exposure number 15) where the camera was placed on top of the Peltier cooler's cold plate.  The difference in noise level was noticable after just 3 or 4 exposures of the camera on the cold plate;

b) In Experiment A the noise in an uncooled camera continued to grow until it was above 1600 units after the 27th exposure;

c) The measured camera noise in both Experiment A and Experiment B grew at approximately the same rate when uncooled (i.e. for the first 15 exposures), but in Experiment B the noise in the camera dropped sharply when the camera was put on the cold plate;

d) For the cooled camera the measured noise level after 40 exposures was a whopping 40% less than in the uncooled camera.

The graph above is a 3-point moving average of the data in the tables below.

EQUIPMENT & METHOD
Canon EOS 350D (unmodified) In-camera Noise Reduction Off;  Battery and CF card in use;
ISO 1600; 120 sec exposures; 10 seconds pause between exposures;
A Peltier 12 volt car fridge cooler cooling a metal plate;
A digital thermometer was used to measure air temperature (20cm distant) for each exposure.
A second digital thermometer was stuck to the cold metal plate on which the camera sat.

RAW DATA

** These are dark frames.  The average value of Red pixels is used here in the table as a measure of noise.  This was measured in MaxDSLR by using the "Info dialog" box set to Area.

The information about this experiment was posted here to the Cloudy Nights forum back on Xmas eve 2006.

The above experiment shows the effect of using a Peltier cooler.  Part 4 shows the effect of using a fan to blow air around the camera.

For more adventures in Noise go to Experiment Part 1  Part 2  Part 3  Part 4

Back to Astrophotography