Feature design: include cooling

[JannesHN]

A good feature to add to the design is a coolingsystem that can cool the samples down to 4 degrees (Celsius, obviously). Our current idea is just keep the chamber in which the rotor rotates as small as possible and cool the whole thing with a peltier ellement and use a thermocouple as sensor. We have a peltier ellement, heat sinks and fans. Like this right now:

The peltier module being used is an off the shelve kit used in minifridge projects.
Link to the product

[jurra]

First try:
we tried measured the temperature after 10 mins and the samples went 28 to 17 degrees celcius. This setup didn't work, becaus we have to go down to 4 celcius. We have to improve the insulation, or think of a different cooling system. But is good to build this before even trying to build a pcb to control the peltier.

[amchagas]

Hi,
It is not clear to me from the photos how you are removing heat from the chamber with the samples. Is the idea that the peltier cools down the heat dissipator and the fan is used to blow air through it, which then gets cooled air into the chamber?

Did you happen to test how cold the heat dissipator gets if the fan sitting on it is not on? This would be a nice way to know if this peltier plus heatsinks and fans is a good combo... In other words if running without the chamber, the system doesn't go below 4c then it might be you only need a beefier peltier+power supply? From the link have a 12V 5A peltier, there are other options available that can take more power in the same area (here is an rs search with more options), maybe this is the easiest way to test, without having to change much of the design? Just swapping the peltier?

A couple of things I can think of that might help, or might be absolute non-sense (I didn't make my mind about it yet :P ):

Ultimately, you want your samples to be at that temperature, correct? would it be possible to cool down to rotor by induction? As in attach the peltier to the motor casing (which doesn't move) and hope to cool down the rotor so that in turn the samples are cold too? Again, maybe this is a terrible idea, since the motor would be cooled down too, but maybe it doesn't affect it that much?

[amchagas]

also, on the peltier swapping idea, you might need a new circuit/power supply to drive it... but this is relatively straight forward to buy, or to design...