Water Cooling Experiments

With nighttime summer temperatures reaching close to 26.5C (~80F) here on the West coast of Florida, my STL11000M is hard pressed to achieve temperatures beyond -12C during imaging. I decided to build an inexpensive test rig so that I could test different methods of water cooling and associated problems, especially as related to condensation and non-water cooled temperature delta comparisons. I decided from the start that I wanted to use a closed loop system because of the advantages provided by the used of distilled water and additives (Water Wetter) that improve cooling efficiency and more importantly, protect the system against corrosion. The test rig was inexpensive to set up and provided a way to test four different water cooling strategies. These strategies are:

1) Ambient temperature water cooling
2) Water immersed radiator
3) Ice immersed radiator
4) Cooled ambient air water cooling.

Of these four strategies, only #3 and #4 are worth mentioning. Strategy #4 proved satisfactory but as I will outline shortly, only marginally so.

Below is a picture of the the test rig. The pump was the most expensive part at around $50 and is rated to 320 gph. The total cost for the system was about $80. The reservoir is made of 1-1/2" PVC parts. There is a 10 foot 3/8" copper coil inside the cooler. This is used as the heat exchanger. The cooler is a small 9qt Coleman. The whole rig is very portable and with the barb connectors included with the STL11000M there is no coolant loss when the hoses are disconnected from the camera.

The image below shows the inside of the cooler during the ice test. It shows the 10 foot - 3/8" copper coil.

The next image shows the cooling parameters during the ice test while taking darks, as well as the ambient temperature, which was recorded at 32C (~90F). The cooler had no problems achieving a 46.4C delta and could easily go lower. However, as night fell and the ambient temperature dropped to around 26.5C (~80F), condensation started to show up on all the fittings and hoses, including the hoses inside the camera. At that ambient temperature the camera could easily maintain -20C at 60% power, but condensation became unacceptable.

The image below shows the setup for test #4. This strategy used a "cooled ambient" by using a couple of Igloo ice substitutes in the closed cooler. There is no ice bath or cold water in contact with the copper tubing, just cooled air. With the camera running, the coil temperature was maintained at around 21C (~70F) for six hours while the camera ran at -14C in 26.5C (~80F) ambient. The ambient Delta between the ambient air and that inside the cooler was about 8C (~10F) at 80F ambient and 70F for the copper tubing. Maximum Delta with this strategy was ~42C with no condensation.

Conclusions:

The camera is limited to about -15C in warm humid environments, but only because of condensation. This all dependent on dew point of course.

There seems to be no substitute for a cool air environment around the camera when it comes to efficient cooling. This is counter-intuitive as water cooling is so much more efficient. However, camera design plays an important role.

Very slight drops in ambient air temperature, and/or slight environmental air movement, made dramatic differences in cooling efficiency. This was tested indoors by manipulating ambient temperatures and air flows using the A/C and fans. A 3F degree drop in ambient temperature (from 79F to 76F), coupled with adding a subtle air flow around the camera resulted in a cooling gain of -4C.

The camera is able to achieve -10C comfortably without water cooling at an ambient temperature of 24.5C (~76F). At this same ambient temperature, the ice bath strategy produces too much condensation to be a viable alternative. This could be different at other sites where the dew points are lower than those here on the West coast of Florida.

Strategy #4 produces no condensation. However the difference between water cooling and air cooling is a paltry -6C. Is this worthwhile? I think so, but just barely worth the trouble.

The camera design sets a limit as to what can be accomplished with water cooling because of the risk of damage from condensation.

The design of the camera is such that hot air coming out through the vents on the case back is being drawn back in by the fan. This results in a loss of cooling efficiency. This might be one reason why cool air circulating around the outside of the camera improves cooling efficiency to the extent that it does.

A camera design immune to condensation issues seems quite feasible and should not be a difficult design to implement.

I believe that there are no substitutes for a more robust camera design in terms of air cooling efficiency and immunity to condensation. That said though, I believe the STL series cameras are the very best available in the amateur market today.