Published: Jan 1990
| ||Format||Pages||Price|| |
|PDF Version (284K)||19||$25||  ADD TO CART|
|Complete Source PDF (14M)||19||$104||  ADD TO CART|
There has been no general agreement on how to perform, analyze, and report hot box tests on the thermal performance of reflective insulation systems which consist of a series of airspaces separated by reflective foils. This paper presents the results of experimental and analytical studies on the horizontal heat transfer through single- and multi-layer reflective insulations installed vertically. The multi-layer insulations were commercially available reflective insulation products. The experimental studies were carried out in a guarded hot plate and a calibrated hot box. The results showed the effect of aspect (height to width) ratio on the convective heat transfer in the air space. The experimental results were compared with a detailed three-dimensional finite element model of the conductive, convective, and radiative heat transfer which occurs in the reflective insulation system. The results were also compared with the original references used to determine the thermal resistance of plane reflective air spaces that are presented in the ASHRAE Fundamentals Handbook. These two analytical methods were also used to compare a wood and extruded polystyrene support structure for the reflective insulation product.
The results showed that the guarded hot plate did not adequately model the convective heat transfer, due to the low aspect ratio, and that a full-scale facility is needed to properly test a reflective insulation product. It was also indicated that the thermal performance of reflective insulation products are not comparable with the results for plane reflective airspaces in the ASHRAE Fundamentals Handbook due to the thermal bridges which characterize most reflective insulation products.
reflective insulation, multilayer insulation, thermal insulation, thermal resistance, R-value, calibrated hot box, guarded hot plate, finite element method, thermal performance
Jim Walter Research Corp., St. Petersburg, FL
Sandia National Laboratories, Albuquerque, NM
Computer Simulations, Inc., Russellville, AR
University of Massachusetts, Amherst, MA
Paper ID: STP23319S