[Space] Heating

[Christoph Maier]

On Thu, 2010-02-11 at 23:48 -0800, Joachim Pedersen wrote:
> So, 555timers are really small ICs, which can't really push that much
> current, so you would need to have the 555 trigger some kind of power
> FET methinks. Also, resistive heating is notoriously inefficient...

It is, because you lose a lot of heat in the controller, in particular
if you don't use PWM. 

However ... we want to heat some stuff up to the point where transistors
are cozy and warm, not higher. 
The most efficient way to do this is to use the regulating transistors
_themselves_ as heating elements - mounted together with the devices we
want to heat on a good thermal conductor (i.e., a block of aluminium),
and wrap a good thermal shield (that also serves as padding to prevent a
reasonably solid piece of metal hitting something at landing) around it.

The electronics would contain power transistors as heating elements, an
adjustable temperature regulator, e.g., 
LM56 or LM57 [ http://www.national.com/cat/index.cgi?i=i//73 , 
http://www.national.com/cat/index.cgi?i=i//71 ]
and a few resistors and capacitors. 

Christoph