consider noiseless Peltier thermoelectric cooling
Open, NormalPublic


consider noiseless Peltier thermoelectric cooling

aombk created this task.Oct 26 2014, 1:01 PM
aombk updated the task description. (Show Details)
aombk raised the priority of this task from to Needs Triage.
aombk added a project: Brainstorming.
aombk added a subscriber: aombk.
sebastian assigned this task to Bertl.Oct 28 2014, 2:43 PM
sebastian added a subscriber: sebastian.

Already planned, Herbert please be so kind and add a short development progress comment.

sebastian triaged this task as Normal priority.Oct 28 2014, 2:43 PM
allan added a subscriber: allan.EditedOct 31 2014, 8:37 PM

Sounds like a valuable addition to offer as an add-on to the external chassi of the camera. For normal use-cases the less complex passive solution favours something that doesnt add cost/weight/power use.

In any event going via the metal of the chassi is enough to conduct heat, i cant imagine there are any applications where you need peltier cooling on-die.

Adding it onto, rather than into, the camera, means you can also figure out a way to cool the hot side of the peltier element. For which there is ample room behind the camera where the fpga is. I understand that to be the main producer of heat.

Bertl added a comment.Nov 8 2014, 7:58 AM

The basic idea so far is to have a metal plate covering the entire sensor PCB (front side), pressing against the sensor base (thermal connection) similar to the heat sinks used for CPUs.

If we manage to get the peel-a-way sockets, this plate could cover almost all of the sensor area and extend through the PCB as well as on all four sides.

Such a heat transfer plate can also be used to attach peltier elements or heat pipes which would allow to efficiently cool the sensor.

But we will have to do some test in this regard first before we decide that we want to go this way.

black magic's ursa mini cooler: