Abstract
To investigate the effects of moisture movement on thermal efficiency, an experiment was performed on a low-slope roof system containing permeable insulation. Five replicate test panels of the roof system containing a variety of vapor retarders were evaluated in a climate simulator. The test panels were outfitted with temperature sensors, heat flux transducers (HFTs), and two types of moisture probes. These instruments were installed on both sides of the insulation material; additional temperature sensors were located at the interfaces between the various layers of the roof system. After initial steady-state testing to verify that the replicate panels exhibited similar thermal performance characteristics and to check the instrumentation, the panels were subjected to an external diurnal cycle representative of a cloudy southern continental summer climate. During this initial `dry run,' only hygroscopic water was present in the panels. Water was then added to four of the panels, and the experiment was repeated with diurnal cycles representing cloudy and sunny conditions being imposed. Significant differences in the outputs of the HFTs located on opposite sides of the insulation layer were noted in the `dry run' on panels containing impermeable vapor retarders. In this run, peak heat flows measured by the bottom HFTs were larger than those at the top of the insulation in spite of the temperature variations being imposed on the top surface. It is theorized that the difference is due to latent heat effects and that these effects were due to the presence of hygroscopic moisture. After the addition of water, all the bottom HFTs were surrounded with condensate and temporarily responded to the latent effects, whereas the top HFTs all responded as they did in the `dry runs.' A combined heat and mass transfer model that takes into account latent heat effects was used to estimate the latent and sensible heat flows for the test panels. The total (combined latent and sensible) heat flow predicted by the model approximated the output of the bottom HFT, while the top HFT output approximated the calculated sensible heat flow.
| Original language | English |
|---|---|
| Pages (from-to) | 1004-1012 |
| Number of pages | 9 |
| Journal | ASHRAE Transactions |
| Volume | 99 |
| Issue number | pt 2 |
| State | Published - 1993 |
| Event | Proceedings of the 1993 Annual Meeting of the American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. - Denver, CO, USA Duration: Jun 27 1993 → Jun 30 1993 |