Created September 2010

See below for update
October 2011



DSLR Peltier Coolbox

I decided that I needed a cooling solution for my DSLR after a frustrating summer of difficult to process 'noisy' images, especially in the red channel of my DSLR photographs, after doing a bit of research on the web, I found a few DIY variations to cool a DSLR chip, some quite drastic which involve rehousing only the necesary components of the DSLR, and others involving a 'cold-finger' inserted into the camera, neither appealed, as my 40D is a recent purchase, and I also want to use it for general photography, so I settled on a peltier cooled box instead which the camera is placed inside, with a hole at the front to allow connection to the telescope.

Peltier Coolbox

The unit started life as a mini 'England' beer-cooler which I bought from a car-boot sale for a couple of pounds after yet another failed World-Cup attempt, It had a 40w peltier cooler unit with a 12vDC and 230vAC PSU, so was ideal for use in both the observatory or on location with 12v power if required


The components were stripped out, which composed of a 40w peltier, heatsink, fan, thermostat, dual voltage PSU, and supply cables, the PSU was rehoused in a ABS project box from Maplins (ML47), and I used a multicore data cable to extend the wiring to 3m length, using 2 conductors together for the actual peltier supply lines to avoid them overheating when running (approx 3A @ 12V) and the remaining cores for the power for the fans and thermostat

The main box is an aluminium 2 part 1mm walled case, again from Maplins (XB71) which measures 210x160x80mm, and was just big enough to house the camera with the USB attached and room to be able to open the CF card slot with the camera in situ. A template was made from cardboard and then transfered to the box for cutting, which was done by chain-drilling the apetures and cutting away the unwanted parts with a chicel, and finally cleaned and shaped using a file, the finger-grip of the camera protudes slightly so a indent had to be made into the case to allow the camera to sit flush with the front of the box, this was done by hammering a ballbearing against a peice of wood which I had routered out a depression

Next was to build a 'cold-feeder' plate, which would run through the box and allow connection to the camera via the tripod screw, and also to provide a cold-sink for a fan to circulate the cold air inside the box, as this attaches directly to the camera, and therefore the camera's metal chasis, it should also help speed up the cool-down times of the camera. A length of 28mm copper plumbing pipe was flattened in a vice and shaped to suit the box, and an extension piece was bent to suit the small internal circulation fan. Holes were drilled where the fan is attached to allow air flow through the cold plate, and the surface which connects to the alluminium box was made as flat as possible using an oil stone, this was then connected to the aluminium box with 3mm screws and heatsink paste applied to allow maximum transfer of the cold provided from the peltier which was to be mounted on the outside of the box.

A hole was drilled in the copper to allow the insertion of a tripod connecting bolt, and other cutouts made to allow the cable from the camera power supply, and the temperature probe to be fitted.

The Peltier was fitted to the outside of the box and the heatsink insulated from the housing using a sheet of packaging foam, again heatsink paste was applied to allow maximum heat/cold transfer, the wires were connected inside a small ABS box (Maplin N80BQ) which was screwed to the case, small holes were drilled to allow the wires fro the peltier and fan to exit the box and also into the box to provide power to the internal cooler fan. The main fan was simply screwed to the heatsink using 5mm screws which fitted into the slots of the heatsink.

Insulating foam sheet has been glued to the outside and bench tests have shown that it will achive 13°c below ambient temperature, hopefully with a thicker insulation and better sealing of the case this will reduce further still, I will add further information and actual results when I have had the opportunity to test the unit out in the observatory

September 2010


UPDATE TO PROJECT - October 2011

I had hoped to give this cooler a firm testing throughout the warmer months, but due to my 40D camera developing a fault I was not able to try it out as planned, so after recently buying a 500D as a replacement, I realised that the cooler would require some modifications to get the new camera to fit

The holes in the box were enlarged to allow for the physical differences between the cameras, and the new camera fitted better than the previous as it protruded further out of the box, which makes it easier to attach to the telescope.


I also decided to disconect the thermistor from the box to allow the unit to run continuously, in the thought that further cooling will be achieved, as I figured that this may have been operating at about 3 ºC which would be the ideal fridge tempearture, but not ideal for achieving maximum cooling, I simply left it connected loose in the project box and added a bit of insulation to the hole where it was sited originally

20mm plastic foam packaging was then used to fully insulate the box, this was glued onto the alluminium using silicone glue, I had tried previously using double sided carpet adhesive tape, which worked well until it got cold, and the insulation would easily peel away from the tape, the silicone had a strongest hold to the insulation in cold tests, and would be adequate for the job, unsure of its thermal properties though! The meeting edges of the insulation were also glued the same

The hinged side of the box was difficult to fit insulation to permanently because of it being an obstacle to the hinged back cover, so I fitted the permanent insulation to form a recess to allow a final piece of insulation to be inserted after the camera has been set up and the box closed for imaging, this piece is not glued, but simply stays tightly in place because of the interlocking tabs (though I keep forgetting to remove it when opening it up)

A depression was cut into the front face of the insulation to allow easier scope connection, and also the areas inside the cool box where the camera protrudes through the holes was removed using a sharp blade

I need to add a small bolt as a means of opening the box, and fit the thermometer to the housing permanently for monitoring use, but otherwise 95% complete


Bench Test

As the box had been tested previously with thinner insulation and approximately 13ºC below ambient temperatures were recorded, I decide I wanted to see if the removal of the thermistor and the added insulation would make any differences

The testing thermostat probe was again positioned centrally to the box (without camera) and test started with a temperature of 21.2ºC

After 30 minutes a drop of almost 12ºC was achieved and at 1 hour the temperature had dropped 13ºC to 8.2ºC - no different to the previous test! It seems the thicker insulation, neither the removal of the thermistor seemed to make any difference


The last test however re-assured me that this project would be worthwhile, I decided to place the temperature probe between the cold sink and the case of the box, so that it was reading the temperature of the actual metal areas within the box, surprisingly the temperature reached 0.3ºC after just 20 minutes, which was quicker than I was expecting following the tests.

Conclusion is that providing I can begin to cool the camera for a minimum of 1 hour beforehand, the box will perform at approx 13ºC below ambient, but continued use will gradually decrease in temperature, further field testing required, or possibly a bigger wattage peltier to be fitted


October 2011