****
 A very useful rule of thumb is that a well designed and built air 
based Stirling engine using typical burner temperatures can deliver 
around 0.5 watt/ atmosphere/cc. ie a 1 litre engine running at 10 
atmospheres could be expected to produce 0.5 x 1000 x 10 = 5 kW.
5 kW/litre is "modest" by most IC engine standards but the 1 litre 
cylinder is far less challenging than a 1 litre IC engine.
****

> Apropos Stirling (small), how about this set up in the focal point 
> of a solar collector ? :
>
> http://www.bekkoame.ne.jp/~khirata/gallery/special1.gif
>
> from:
>
> http://www.bekkoame.ne.jp/~khirata/english/gallery.htm
>
> If it's not obvious, it is a one-cylinder displacer-less Stirling 
> (or Newcomen) engine. The 'displacer' is the shutter that can shield 
> the burner flame from the hot part.

All such designs are fun but of very low specific power output and low 
efficiency. To get good efficiency and power from a Stirling engine 
you have to work hard and smart. For high Carnot efficiency you need 
high temperature. For high delivered  % of Carnot efficiency but not 
necessarily high power density attention to all aspects of the 
regenerator and all the other obvious things is required. For power 
density you need suitable gas (essentially Hydrogen or Helium) and 
high pressures. Air Stirling at modest pressures is entirely doable 
BUT the engine will be very low lowered for its size compared to 
typical internal combustion engines.



> Then there are NASA closed oscillating Stirlings (I think that one 
> could try to build a low tech version using pistons and cylinders 
> from a car engine):
>
> http://www.grc.nasa.gov/WWW/tmsb/stirling.html

More

        http://www.grc.nasa.gov/WWW/tmsb/stirling/doc/lrgfp.html

12.5 kW. Helium. 1050K hot. 15 MPa operating (about 2000 psi).
Non contact seals (important).

20% overall efficiency.
(Carnot efficiency is 50% so tghis is 40% of Carnot which is good).


10 watt radioisotope powered Stirling with 14,000 hours running time.
James would be right at home with this one :-)

        http://www.grc.nasa.gov/WWW/tmsb/stirling/doc/smlrsg.html

Some of this material may be a little dated ;-)

        http://www.grc.nasa.gov/WWW/tmsb/stirling/doc/codes.html

    HFAST is written in Microsoft FORTRAN. For engine performance 
calculations it requires several seconds on an IBM 286 type desk-top 
computer that has 1 MB of RAM. For engine design optimization, which 
could require several hundred performance calculations, a 386 or 486 
type desk top computer would be more convenient.




        RM 

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist