Using reflux solar receivers to collect solar energy for dish-Stirling
electric power generation systems is presently being investigated by
several organizations, including Sandia National Laboratories, Albuque
rque, N. Mex. In support of this program, Sandia has developed two num
erical models describing the thermal performance of pool-boiler and he
at-pipe reflux receivers. Both models are applicable to axisymmetric g
eometries and they both consider the radiative and convective energy t
ransfer within the receiver cavity, the conductive and convective ener
gy transfer from the receiver housing, and the energy transfer to the
receiver working fluid. The primary difference between the models is t
he level of detail in modeling the heat conduction through the receive
r walls. The more detailed model uses a two-dimensional finite control
volume method, whereas the simpler model uses a one-dimensional therm
al resistance approach. The numerical modeling concepts presented are
applicable to conventional tube-type solar receivers, as well as to re
flux receivers. Good agreement between the two models is demonstrated
by comparing the predicted and measured performance of a pool-boiler r
eflux receiver being tested at Sandia. For design operating conditions
, the receiver thermal efficiencies agree within 1 percent and the ave
rage receiver cavity temperature within 1.3 percent. The thermal effic
iency and receiver temperatures predicted by the simpler thermal resis
tance model agree well with experimental data from on-sun tests of the
Sandia reflux pool-boiler receiver. An analysis of these comparisons
identifies several plausible explanations for the differences between
the predicted results and the experimental data.