Cooling aQuickCam

Modification by P. Ryczek
Modification by S. Weiller
Modification by Orion83

The thermal noise is the fundamental raison for limiting QuickCam exposure time at 4-5 seconds. So for imaging deep sky, one must cool the camera's CCD sensor. One can use several methods for cooling the sensor : ice, nitrogen, CO2, Peltier module... The cheapest and more practical on the field solution is Peltier module cooling. It's a thermo-electrical module where cooling is created by electrons motion in semiconductor. The endothermic side freezes whereas the exothermic side produces heat. So it's essential to remove heat from the hot side with an oversized air heater, which can eventually be helped by a fan.

As the peltier module's price is quite high, it's possible to collect Peltier module, air heater and fan from a icebox which can be connected on a 12V car battery. In France, such a icebox costs approximately 300 FF.
One can collect a 40x40 mm Peltier module (3.5A consumption under 12V), a 140x120 mm aluminum air heater, another one which size is 100x100 mm, a 40x40x40 mm aluminum block, some thermal compound, a 12V 80x80 mm fan, a safety thermistor, some supply cables,  a lighter plug + a 3m long cable... and a pitiful icebox ;-)

Thefollowing assembling has been made with a QuickCamColor 2 , ofwhich the CCD sensor (a color TC236P Texas ) has been worthwhily replaced by aTC237 CCD sensor (the same sensor as used in the Bernard Dalstein's Kitty 237   CCD camera). It's a Black&Whitemodel (in reality the Texas data sheet is wrong, the 5 and 6 pins(OUT1 and OUT2) are inverted, but it doesn't really matter,images taken are the best proof), which is more efficient in weaklight, and has a better resolution than the color sensor (becauseto make a color image, 3 pixels are needed). Be careful, it's notpossible to replace the sensor of a QuickCam VC and QuickCam pro,because there is no equivalent Black&White sensor to thissort of color sensor. It's still possible to find Color 2 inamerican surplus (see address page).

The aim is to use all which can be collected, the aluminum block will be reworked in the following way :

this in order to insert the CCd sensor on the cold finger, the sensor being mounted on " shrink model tulip holder " (1.78mm thread, Réf. Radiospares 187-2272, price 43FF)

Being given the size of the second air heater is 100x100mm, we could use a sealed housing, as Bernard Dalstein has suggested it for his Kitty CCD camera. This housing comes from a IP65 shunt box (moisture proof). The size of this box is 100x100x60 mm, and one can find it in all do-it-yourself shop at the electricity shelf. When all the mounting is housed in the box, one will seal it with silicone seal, as used for bathrooms.
every orifices, holes and mounting screws should be clogged with silicon seal. The opening in the housing for the CCD will be obturated by a glass window, which can be reclaimed on an ultraviolet filter or a photo Skylight filter (the price of a 36 mm diameter filter is about 40 FF). One has to seal the window with silicon joint.
the CCD sensor is fastened on the cold finger with two screws mounted on two small epoxy sheets, which can be collected from an old printed circuit. Above all, don't forget to spread thermal paste between the CCD sensor and the aluminum block. It's important to guarantee a good heat transfer. One can note the 1.778mm shrink holder. Not only does this holder allow to make a space between the sensor and the printed circuit (to insert the cold finger), but it protects the CCD sensor's pins from a dead short with the aluminum block.
The air heater is mounted on the Peltier module (don't forget the thermal paste between both parts) outside of the housing. So the cold produced inside the box  won't be reduced. Isolate the aluminum block fan with nylon screws, or with insulating spacers. The fan from the icebox will be very efficient to drain heat from air heater. One can put a small resistance bulb (about 60 FF) on the aluminum block to check the CCD sensor temperature.
One can notice silicone joint residues. That's not very pretty, but it's very efficient to seal the housing. To perfect the moisture insulating and to avoid hoar-frost making on the CCD sensor, one can put some small desiccant bags to absorb every moisture inside the housing.
The assembling is over, one can shut the box top in blowing dry air inside the housing until it is closed (one can use dry air can, or hair dryer).

One should obtain thefollowing results at standard temperature :

The first trial is to take two images in complete dark with an 30s exposure time. The first one is taken at 34,4°C, so without any cooling. The second one is taken after 10 minutes of cooling at 10,3°C (3.5 A consumption under a 12V supply). The results are the two following images. They have been voluntary brighten to underline the thermal noise :

on the left side, 34°C - on the right side, 10°C

Comments would be frivolous....

One can now make realtrials in deep sky imaging.

The following images have been taken at the center of a town, at the focus of a F/D=8 102mm fluo refractor.
CCD sensor temperature : 1.5°C . Some clouds, medium turbulence. Neither dark frame nor flat field frame have been made.

Click on image to enlarge it :

	ALBIREO 9 exposures of 5 sec.  Some stars dimmer than 12th magnitude appear.
	MIZAR - ALCOR A single 5 sec exposure. The small in the up and right corner is 7.59 th magnitude.
	M13 2 (only two !) exposures of 5 sec. What a pity that more images haven't been taken ! It 's very promising...

You canfind many other examples of cooling images in the gallery page.

We wanted to put thecooling QuickCam to the wall, so we tried to image nebulae andgalaxies.

These images have been taken at "le col de Chavade", in the french department  "Ardèche", at an height of 1300m , at the focus of a F/D=8 102mm fluo refractor. CCD sensor : -6°C, dark sky, very pure atmosphere  and very low turbulence. Neither dark frame nor flat field frame have been made.  

	M27 - Dumbell Nebula 40 exposures of 5 sec. One can notice the very low handles of the nebula.
	M51 - Retrievers Galaxy (???) 50 exposures of 5 sec. Only central bulbs of M51 and its satellite can be seen. To make the comparison, here is an image taken by B.Daversin near Paris. The refractor is exactly the same, but the camera is a HX516 Xpress Starlight. 8 exposures of 2 min. It's impossible to fight !!!

Conclusion

The trials of this cooling QuickCam are very interesting on every stellar objects such open clusters, double and variable stars and asteroids. Under a good sky, the 13.5  th magnitude is reached with a 102mm fluo refractor. Some serious studies can be considered : variable and double stars watching, asteroids astrometry, etc...

Nevertheless, we have to face the facts : it's almost impossible to image dim and nebular objects with a 5 seconds exposure. One must modify the software in order to compel the camera not to compose images of which the exposure time is higher tan 5 sec.

But we mustn't forget that the price of the whole cooling camera is lower than 1500FF, the black&white sensor and cooling system included. For this modest price, numerous applications are possible, and it's a great pleasure to see in real time (dim stars on the computer display.

Modifications by P. Ryczek

As the previous system,this one uses a cooling by Peltier module, but the difference isthat the CCD sensor isn't on the printed circuit itself, but isput between two plates, of which one is made of aluminum. Moreinformation on the web site of P. Ryczek .

images by P. Ryczek

Modifications of S. Weiller

Take a soda can. Pierce a1"1/4 diameter hole in the bottom of the can. Cut the tightpart on the top. Make small spits all around the hole with a 1 cmgap between each one. Turn down each split with pincers becausethey have a very cutting edge. Recover a bottle top or a bottombottle, and drill a hole to help the USB cable going through it.
Put the QuickCam with somefoam to stuff. Put some adhesive plate on the plastic cover plate.The QuickCam can remain in this way indefinitely. Put the coolingsystem when you use the system.

Trials ofthe system :
a)  Dark frames with andwithout cooling system

b) Imagesof mars with and without cooling system. The Control image isnicer (less turbulence at this moment), but we notice that somehot spots are missing on the cooled image with the same exposuretime.

Modifications by Orion83

An originalcooling system : recovering a CO2 refill (as used for paint ballguns), and spraying some CO2 on the rear of the CCD sensor (aQuickCam VC in this case). The advantage of this system is itneedn't to unsolder the CCD sensor. More information on the website of Orion .

Images from Orion83

Home Page