Design Philosophy

The plan is to design and construct a test unit to measure the R-value of shade or curtain that would be well documented and could be reproduced easily.  It will necessarily have shortcomings, and further studies could be done to show correction factors that need to be added to the calculated R-value using the apparatus. In order to make this more practical, all of the test temperatures, etc should be published along with the R-values.This included the air gaps on both sides of the test material (window side and room side). There are many variables that are not covered by the test, including window temperature and size, test piece location (closeness to the window), temperature of the interior wall around the window, additional window coverings (like curtains in front of shades), and drafts from forced air heating. Computational Fluid Dynamics (CFD) can be used to create R-Value corrections after the models are tuned to match the test results.

DETAILS OF THE PROPOSED TEST UNIT

Size: The 2’ x 2’ “window” size will be used.  Although a slightly larger size of perhaps 3 ft x 3 ft would be closer to a commercial size and would have a slightly lower fraction of frame heat input, it would be bulky (nearly 4’x4′).

Window temperature:  The use of ice to maintain the cold face at (or very near) 32F is close to value for the inside of a real window at -20F outside temperature, and is easy to achieve with ice. It gives a reasonably high delta-T (72-32F) and heat flow for the test. Most of the inaccuracies for the test will come from heat-inflow through the “window frame” and these become relatively smaller as the heat flow through the window-covering increases. Hopefully, CFD can be used to determine the effect of window temperature on Rw.

Location of cold and warm surfaces relative to the test piece: The test piece will be at a constant distance of 2.5” from the cold face. The air gap on the room side of the test piece is more complicated. The convection effect on this side can vary due to window size, and room conditions. A standard distance is needed so that corrections for such variables can be made to the standard R-Values. The warm plate surface is thus set at a constant distance of 7.5” from the cold face, giving roughly 5” between the shade and the warm face.

Zeroing out effect of no curtain in place: The air gap between the window (cold face) and the room (warm face) can be determined by runs without any test material in place. The test results with a covering in place are reduced by this amount.  The Rw results include the resistance introduced by the stagnant air on both sides of the test piece test piece, as well as the resistance of the test piece.

Minimizing Unwanted Heat Flow: There are several heat flows that can distort the results from this design. They are: 1) heat flow in through the walls (casement in a real window) on the room side of the test piece, 2) heat flow in through the walls on the window side of the test piece, and heat bypassing along the wall (casement) from the warm side to the cold side.

  • Determination of heat flow with Heat balances: To analyze these effects, calculation of heat flows was made by setting different intrinsic R-values for hypothetical shades and predicting the effect of convective heat transfer on each face. The temperature of the air (and thus the frame) was calculated for both sides of the shade. The heat balance over the insulation covering the frame could then be calculated.
  • The heat flow through the test piece can be predicted adequately for a variety of R-values by using the commonly available convection heat flow equation (Q = Const * Area * (T_room – T_surface) 1.25).  If the wall insulation R-value is known this can be compared by Q = 1/R-Value * Area * (T-surface – T_avg_inside).
  • Determination of heat flow with Heat balances: To analyze these effects, calculation of heat flows was made by setting different intrinsic R-values for hypothetical shades and predicting the effect of convective heat transfer on each face. The temperature of the air (and thus the frame) was calculated for both sides of the shade. The heat balance over the insulation covering the frame could then be calculated.
  • Predicting Convection Effects: The heat flow through the test piece can be predicted adequately for a variety of R-values by using the commonly available convection heat flow equation (Q = Const * Area * (T_room – T_surface) 1.25).  If the wall insulation R-value is known this can be compared by Q = 1/R-Value * Area * (T-surface – T_avg_inside).
  • Heat Input through the Warm-side Wall (1):  The heat balances showed that the delta-T between the warm room and the warm-side interior of the test box was low enough that thick insulation (4 to 5″) could reduce outside heat in-flow to acceptable levels.
  • Heat Input through the Cold-side Wall (2):  The heat balances showed that the delta-T between the warm room and the cold-side interior of the test box was too high to reduce heat-inflow with insulation alone. This requires the use of a guarded-cold-box design.
  • “Guarded Cold Box” design: The “guarded cold box” is a 1″ thick passageway in the frame around the window that can be cooled so that the temperature on the cold side of the frame can be matched by the temperature of the inner-wall of the “raceway”. This assures that little if any heat flows in from the outside. It was calculated that 2 inches of insulation on the cool side of the frame allows a 1 deg F delta-T across the insulation with only 2% of the heat flow across the test piece.
  • Cooling the Air in the Raceway: The cold face of the “window” (an aluminum sheet) can be extended outward by 3″ on all sides to also cool the raceway. A small (2” x 2”) computer fan circulates the air in the raceway (part or all) over one or more finned heat sinks attached to the cold face.
  • Heat Flow Along Wall (3): Heat conducting along the walls (casement) from the warm side to the cold side and getting around the test piece would bypass the of the test surface and distort results.  This can be minimized by building the casement with highly insulating pink foam, and minimizing the cross-section of highly conductive wood in the frame around the test piece. The test piece would be mounted in a thin frame of 1/4″ thick plywood, and be surrounded by an additional 4 to 5″ of pink foam insulation which should be sufficient to prevent significant heat flow along the walls of the test unit.
  • Transporting Test Pieces: Transporting the flimsy test piece and frame will require more substantial carrier boxes. The test piece and frame must be able to slide out of the carrier box and into the test unit. These carrier boxes should allow test pieces to be prepared by interested parties anywhere and be shipped for testing.