Originally posted by gs2c Hello S.Migol,I know you remodel your cooling k10d And I want to copy it in my K5.Does the tec Close to the cmos sensor?How did you do?
What I did with my K10D was a two-step process:
First I had it converted to full spectrum by Spencer's. At the time, they were doing a brisk business in conversions and they had an option to add a passive thermal control to it. I believe that it was some kind of heat transfer material that they put on the board or frame to help it shed heat. Comparing the local warm spots on an unconverted K10D and the modified one, I could feel that the bottom of the camera body and the LCD screen were warmer.
Second, I decided to find a way to help the camera shed heat from these warm spots. I picked up a complete peltier assembly from Tellurex that had a heatsink and peltier. Then I got a power control module with thermal probe from Oven Industries. Also, I got some thermal gel pad to help transfer heat on the uneven surface of the camera to the heatsink. I also used a bit of aluminum to wrap around the bottom of the camera to help transfer heat from that hot spot.
Here you can see the peltier device on the back of the camera:
And when it's off the camera, you can see the metal I've wrapped around the base:
Note in this second shot there are shims crammed into the edge of the mount. This is because with the weight of the heatsink device, there is sagging as the telescope moves.
When I first put the kit together, I didn't do anything special about the edge where the heatsink and the camera meet. During the summer as I used it, I noticed a lot of rime ice forming at this location. This is wasted energy, so I used an old wool sock to wrap around this gap. The sock helps insulate the gap and keeps the rime ice from forming. I do have to make sure that the sock doesn't create a gap between the heat sink and the camera. If there is a gap, then the peltier will not work and may fail.
In looking at the service manual for the K10D, I can see the reason for the hot spots I've noted. The LCD is pressed against several components on a circuit board. The frame that holds the CCD is connected to the frame that is also connected to the tripod socket. Thus, with heat transfer on these two spots, the setup does a pretty good job of removing heat from the camera. I am not sure if these are the same in a K5. For someone that has a K5, you'd need to run it for a while and feel the outside of the body to determine if a similar setup would work.
Note that I do not use a battery inside the camera. I always use an external power supply. Batteries generate heat and they don't work very well at low temperatures.
Eventually, I'll retire this K10D and just use it for IR pictures or widefield piggyback shots. I will likely replace it with the QHY10 astronomy camera because it uses the same Sony CCD. The advantage there will be no amp glow and more efficient peltier cooling.
If I were to recreate this cooling, I'd want to work closely with the people doing the IR conversion. I'd want to see if there would be a way to get better thermal transfer pad added to the frame of the camera. I'm not convinced that a cold finger to the CCD would make a difference on the K10D because of the bad amp glow. I've put the camera in the freezer and let it get down to -10C and there is still amp glow. Also, adding a cold finger to the sensor would break the weather sealing. During a session, the K10D will become covered in dew and thus I'd prefer to not have a risk of water intrusion. Another thing that I would do would be to assemble my own heat sink device (I may do this someday). I'd want to have the option to use a lightweight video-card style cooler. I'd also like to investigate how small a device I could use to better manage the power budget. The existing setup will pull 3 amps all the time. This requires a lot of battery power over the course of a night - my power budget for all the gear is about 50 amp hours.
Lastly, there are complete cooling boxes that can be purchased or if you are crafty you can make your own. I decided to go with a direct-contact approach rather than trying to cool the air around the camera in the hopes it would cool the components. I'd already tried this approach by using the camera without a cooler in the cold of winter and could see that temperatures were not stable and the camera would still become rather warm at least 10C over ambient. With the device in place, the EXIF temperatures are stable over the course of a night - varying just a few degrees - and are usually at or very near ambient.