Solar Energy, Vol. lg, pp. 253-257. Pergamon Press 1976. Printed in Great Britain END-CLEARANCE EFFECTS ON RECTANGULAR-HONEYCOMB SOLAR COLLECTORSt D. K. EDWARDS,J. N. ARNOLD and I. CATrON Energy and Kinetics Department, School of Engineering and Applied Science UCLA, Los Angeles, CA 90024, U.S.A. (Received 28 July 1975; in revised form 18 February 1976) Abstract--Results are reported of an experimental program to measure the effects of gaps between a honeycomb core and its coverglass and absorber plate. Nusselt numbers up to values of 2 vs Rayleigh number are reported for gaps of 0, 1.5, 2.3, 3.0 and 4.6 mm above and 0 and 1.5 mm below a 19 mm thick honeycomb core with 4.69 × 40.3 mm rectangular cells. The nontranspired honeycomb system was heated from below and oriented at 0 °, 15 o and 30 ° from the horizontal with the long dimension of the cells running horizontally. The results indicate that a well-designed honeycomb core will give good performance in a solar collector even with clearance gaps of 1.5 mm above and/or below the core. INTRODUCTION When a properly-sized honeycomb of a thin, poorly conducting, IR opaque, and solar transmitting and/or reflecting material is located between the absorber plate and cover glass of a solar heat collector, both natural convection and reradiation heat losses from the top of the absorber can be effectively reduced. A rectangular shape, with the long dimension running horizontally east-west, is desirable to increase solar transmittance for a fixed flat plate collector. The question arises in the design of such collectors whether or not it is important to minimize the bottom end clearance between the honeycomb core and hot absorber plate and the top end clearance between the core and cold cover glass. With such gaps, and particu- larly when tilted, will the honeycomb continue to inhibit natural convection heat losses to the cover glass, or will significant air flow occur up one cell through the gap and down another? Francia[1] employed circular glass cylinders 15 mm in diameter with a length-to-diameter ratio L/d of 17 over a tracking collector having a 6-to-1 conical concentrator. Buchberg, Edwards and Lalude [2] indicated that a rect- angular cellular structure with the long side of the rectangle running east-west (see Fig. 1) was preferable to the circular or hexagonal geometry for fixed flat plate collectors. The directional or wavefront[3] selectivity of such a geometry better accommodates the morning and afternoon sun when the collector plate is fixed. The solar transmittance of the cellular structure is appreciably higher for the elongated rectangle during morning and afternoon hours. The first study/2] indicated a broad optimum in cell configuration for L/d between 3 and 8 for a specified W/d = 3.4, and the authors recommended d = 5.3 mm for collecting solar energy at absorber temp- eratures in the neighborhood of 100°C. It was also pointed out that as an alternative to using selectively transmitting glass, one could use a selectively reflecting material such tPresented at the 1975 I.S.E.S. International Solar Energy Congress and Exposition, Los Angles, California, 28 July-1 Aug. 1975. as aluminized plastic-undercoated paper or paper-board with the aluminum coating overcoated with solar-clear, IR opaque resin. Later Buchberg, Lalude and Edwards/4] reported experimentally-measured performance data for d = 5.3mm, L/d =7.11 and W/d =3.4 and two other cell configurations. The measurements showed total heat los- ses by combined radiation, conduction, and convection out the top of 467 w/m 2 for an absorber temperature of 106.7°C and cover glass temperature 42.8°C under solar irradiation of 962w/m 2 with the solar incidence 11° off normal at noon and a wind speed of 4.7 m/s. Under less favorable conditions with 704 w/m 2 solar irradiation and the sun 46o off normal at noon and a wind speed of 2.3 m/s, the top losses were 391 w/m 2 with the collector at 71.7°C and the glass at 46.7°C. Lalude and Buchberg[5] carried out a more detailed optimization study than that of Ref. [2] and concluded that W/d should be 6 with L/d between 4 and 5 for 8&C collection, 6-8 for 80-95°C collection, and L/d = 10 for collection temperatures above 95°C. As explained in the optimization papers/2, 5] a high solar transmittance is desirable, hence the use of glass or aluminum to permit solar energy to be transmitted or reflected down to the absorber plate. A low effective IR emittance to limit radiation heat loss is required, hence the use of IR opaque glass or plastic overcoating and the use of high values of L/d. Bare metal honeycomb will not do. The effective emittance decreases with increasing L/d [6-- 10]. Low wall conduction to limit conductive heat loss is a necessity, hence the use of very thin-walled glass or aluminized paper or plastic; metal foil or metal hon- eycomb cannot be fruitfully employed. Finally, the Nus- selt number, which gives the ratio of effective fluid conductivity to true conductivity, must be small as must the fluid conductivity itself in order to limit conduction or convection heat loss through the fluid-filled honeycomb cells. The effective conductivity becomes large rapidly when a critical temperature difference is exceeded. The critical temperature difference is a strong inverse function 253