Peltier Cooling of Modified Canon Digital Rebel XSi (450D)

New Page 4

Peltier Cooling of Modified Canon
Digital Rebel XSi (450D)

Version III

by Gary Honis

UPDATE :
I am now providing a low cost modification service to convert your 450D or
1000D. CLICK HERE for
details.

The

first version
of the cooling system I made for the Canon modified 450D did not cool the camera
as well as I liked. It used a coffee canister as the cooling chamber and its
metal was not a good conductor. The
second version
of the cooling system used part of the original cooler for the cooling chamber,
instead of the coffee canister. It cooled the camera very well, dropping its
temperature 38 degrees Fahrenheit. The third version of the cooling system is
discussed here. It is smaller and lighter than the previous versions and cools
even better than the second version providing an even greater temperature drop
of 43.7 degrees Fahrenheit.

The original Cooler is branded "RubberMaid",
but the cooler is made by Vector, Inc. and branded under other names as well. It
operates on 12 volts DC and Vector also makes a 6-Amp AC/DC converter to operate
the cooler using 120 volts AC.

Below are some quick images of
how I used a peltier device from the 9 can beverage cooler ($25), a sheet of
aluminum ($10), and a styrofoam cooler ($3) to build a peltier cooling system
for my modified Canon XSi (450D).

Detailed assembly instructions for anyone
wanting to build a similar cooler can be seen
HERE.

Since the 450D (XSi) and the 1000D (XS) have the
same height and depth, this cooler fits the 1000D (XS) as well.

In the image below the peltier cooled Canon XSi
(on right) is in the focuser of an Astro-Tech 127mm triplet refractor. In the
background is a Meade DSI Pro imager in the focuser of an Orion ED80 used for
autoguiding:

The following photos were taken during the
assembly process:

Peltier, fan and switch parts removed from the
cooler:

Fan remounted to heat sink:

The cooler's on/off and cold/warm switch with
LED lights was taken off cooler and mounted in a project box.
Switch was not mounted on cooler but instead placed
on cables four feet from cooler to help keep the cooler weight low.

A very thin .025 sheet of aluminum (1/40 inch)
is formed into a rectangular shape just large enough to fit the Canon 450D body:

Holes are made for the 2" Barrel to focuser and
for the camera's flash nosepiece and the peltier/fan assembly is attached:

Placed a small 12 volt heatsink/fan inside to
circulate cold air:

Aluminum cooling chamber is covered with light
pieces of styrofoam for insulation:

Images of cooler on scope:

Version III - Dark Frame and
Cooling Tests:

I used two temperature probes to test the
cooling effect while taking dark frame images. One probe was placed on the
aluminum chamber wall and the other on the camera body.
I took dark frames every five minutes at ISO 1600
and measured the air temperature inside the cooler at the two probe positions.
The ambient temperature during testing was 77.4 degrees. A plot of the 43.7
degree Fahrenheit temperature drop of the cooler over time is shown in the graph
below:

I was very pleased with the above temperature
drop over time as measured by the two probes. It was an improvement in both the
rate of temperature drop and maximum temperature drop achieved over Version II
of the cooler, probably because of the smaller size of the cooling chamber of
Version III.

Version III - Dark Frame
Results:

For the graph below, I converted the ISO 1600
Canon five-minute dark frame raws to TIFs using Canon's Digital Photo
Professional and then recorded the standard deviation of the whole color image
(4272 X 2848 pixels) for each dark frame. Note the comparison to the dark frames
of the Version II cooler. It shows that the smaller size of the Version III
cooler and one internal fan as opposed to two internal fans, provides the same
reduction in dark frame noise.

Advertisements -->
Plastic Eyes |
Free
Business Cards

Notice in the above graph that noise increases
over the first ten minutes (the first two 5-minute exposures). It shows the
delay in cooling of the camera's imaging chip and electronics. I'll have to do a
test with no cooling (camera outside of the cooler) and add the results to the
above graph to see how much of a benefit the cooler provides over non-cooling.

Version III - Dark Frames:

Below is a comparison of the 342 X 342 pixel
center crops of the five-minute dark frames at ISO 1600 taken at ambient
temperature (77.4 degrees Fahrenheit) and after one hour of cooling ( 37.6
degrees) and two hours of cooling ( 34.0 degrees) cooler temperature:

Detailed assembly instructions
for anyone wanting to build a similar cooler can be seen
HERE.