NIST Time|NIST Home|About NIST|Contact NIST

HomeAll Years:AuthorKeywordTitle2005-2010:AuthorKeywordTitle

Experimental Verification of a Moisture and Heat Transfer Model in the Hygroscopic Regime.

pdf icon Experimental Verification of a Moisture and Heat Transfer Model in the Hygroscopic Regime. (1136 K)
Burch, D. M.; Zarr, R. R.; Fanney, A. H.

U.S. Department of Energy (DOE); Oak Ridge National Laboratory (ORNL); American Society of Heating, Refrigerating and Air Conditioning Engineers (ASHRAE); Building Environment and Thermal Envelope Council (BETEC). Thermal Performance of the Exterior Envelopes of Building VI Conference. Proceedings. Thermal VI. December 4-8, 1995, Clearwater Beach, FL, 273-282 pp, 1995.


moisture; heat transfer; buildings; building envelopes; MOIST; moisture transfer; calibrated hot box; manufactured housing; moisture analysis


The National Institute of Standards and Technology (NIST) has developed a personal computer model, called MOIST, for predicting the transient moisture and heat transfer within building envelopes. This paper summarizes selected results from a comprehensive laboratory experiment conducted to verify the accuracy of the computer model in the hygroscopic regime. This paper discusses three different multilayer wall specimens installed in a calibrated hot box. The exterior surface of the wall specimens were first exposed to both steady and time-dependent winter conditions, while their interior surfaces were maintained at 21DG (70DGF and 50% relative humidity. These boundary conditions caused moisture from the interior environment to permeate into the wall specimens and accumulate in their exterior construction materials. Subsequently, the exterior air temperature was elevated to 32DGC (90DGF), and the exterior construction materials lost moisture to the interior environment. The moisture content within the exterior construction materials and the heat transfer rate at the inside surface of the wall specimens were measured and compared to computer predictions. The moisture and heat transfer properties for the construction materials comprising the wall specimens were independently measured and used as input to the computer model. The agreement between predicted and measured moisture contents was within 1.1% moisture content. Predicted and measured heat transfer rates also were in close agreement. Accumulated moisture was observed to have little effect on heat transfer because moisture did not accumulate above the hygroscopic limit (i.e., the so-called fiber saturation point) and capillary water did not exist within the pore space of the materials. The insulation remained relatively dry, and the boundary conditions did not give rise to a latent heat effect (i.e., water was not induced to evaporate from one part of the construction and condense in another part).