The problem is that the isolators parts of coils heat must be removed.
Isolation materials have low power transmission coefficient (about 0.6-2.5 W/mK) with respect copper (>400W/mK).
Unfortunately attaching directly copper parts to vacuum chamber wall would shortcircuit the internal coils
This problem was detected during the thermal design of Rita and Patricia fusion reactors
Hi,
the less fancy way is to make your insulators as thin as possible and force cool them with water as the thermal conductivity goes with 1/mK.
The more fancy way is to use electrical insulators that have a good thermal conductivity such as Al2O3 (ca. 30 W/mK) or even diamond layers (ca. 1800 - 2000 W/mK, so much better than copper).
Thank you Johannes it can be a great solution (as using Beryllium oxide with similar transmission coefficient than copper), but then I have another problem:
Ceramics has higher melting point than copper so it is very difficult to cover it. Also great difference of temperatures and expansion coefficient can break it (I suffered a lot when trying that with Pulsotron-2 that broke a lot of targets)
No problem :)
As far as I know, BeO has not such a good thermal conductivity and it is poisonous, I think - so handle it with care.
What do you mean by it is very difficult to cover it due to the higher melting point? - Do you want to have Cu on top or the insulator?
My idea is to transfer copper coils heat (but not its electric current) to the vacuum chamber using the ceramic.
So it would be great to cover the copper with powder go to a ceramic owen and build a block but unfortunately the copper would mix with the ceramics because of its lower melting point
so why don't you just coat the copper with aluminum and oxidize the aluminum surface?
Standard way to remove heat from the coil is making the coil of a conductive (copper) pipe with coolant circulating inside or tracing coil wire inside dielectric pipe with circulating coolant. Other ways are less effective.
Attaching of a close-fitting metallic radiator to the hot parts may help.
> Attaching of a close-fitting metallic radiator to the hot parts may help.
In vacuum radiators would transmit power by radiation that is poor.
Perhaps using a thermal conductive but isolation compound covering the radiator between the metal parts and external vacuum chamber walls could help, I will make the calculations
> making the coil of a conductive (copper) pipe with coolant circulating inside or tracing coil wire inside dielectric pipe with circulating coolant
That is a good system used by Fusor to cool the inner electrode. In my case my coils are very thin, I should use less loops but a pulsed power (of some kiloamps) to one thicker with the coolant inside. What is better coolant? water does not withstands much more than 100ºC
Using water as a coolant one usually is able to prevent excessive heating of the coolant simply with adequate feeding. Limitations may arise from high dielectric constant of water and chemically induced electric conductivity of the water used. If those limitations are serious water can be replaced by transformer oil. But thermal properties of oil are nowhere near so good as that of water.
- Badges
- Science topic
- Similar topics
- Engineering
- Chemical Engineering