Thermal comfort in environments with different vertical air temperature gradients

Abstract

The use of displacement ventilation for cooling environments is limited by the vertical temperature gradient. Current standards recommend a temperature difference of up to 3 K/m between the head and the feet. This paper reviews the scientific literature on the effect of vertical temperature gradients on thermal comfort and compares this to the results of our own experiments. Early experiments have demonstrated a high sensitivity of dissatisfied test subjects to changes in the temperature gradient between head and foot level. Recent studies have indicated that temperature gradients of 4‐5 K/m are likely to be acceptable, and the mean room temperature may have a greater sensitivity on the percentage of dissatisfied (PD). In new experiments, test subjects have evaluated the thermal comfort of different vertical air temperature gradients in a modular test chamber, the Aachen comfort cube (ACCu), where they have assessed vertical temperature gradients of ΔT/Δy = 1, 4.5, 6, 8, and 12 K/m at a constant mean room temperature of 23°C. The results of the different temperature gradients are in contrast to ANSI/ASHRAE Standard 55 (Thermal Environmental Conditions for Human Occupancy, Atlanta GA, American Society of Heating, Refrigerating and Air Conditioning Engineers, 2013) as the PD increases almost constantly with higher vertical air temperature gradients. The PD for the overall sensation increases by approximately 7% between gradients of 1 and 8 K/m. The evaluation of our own tests has revealed that vertical temperature gradients of up to 8 K/m or higher are likely to be acceptable for test subjects.

Relief Demand Calculation in Humanitarian Logistics Using Material Classification

“…N l represents a natural scenario combined with natural factors. For example, this study divided daily apparent temperature into two states, average night-time and average daytime temperature, and 20 • C was defined as a watershed [70][71][72]. Both temperatures have two states: H (>20 • C) and L (≤20 • C) values.…”

Predicted percentage dissatisfied with vertical temperature gradient

“…In the study upon which ASHRAE 55 and ISO 7730 specifications are based, Olesen et al [3] exposed 16 seated subjects (8 males, 8 females) with light clothing (0.6 clo) in a tiny chamber (length: 2.0 m; width: 1.4 m; height: 2.0 m) at four vertical temperature gradients (0.4, 2.5, 5.0, and 7.5 °C/m) for 3 hours. Cooling panels in lower walls/floors and heating panels in upper walls/ceilings created the vertical temperature gradient, while the influence of radiant asymmetry was not controlled or investigated [8,9]. The mean indoor temperature was adjusted according to the subjects' preference in an initial test without a vertical temperature gradient.…”

“…A study by Möhlenkamp et al [8] evaluated the percentage dissatisfied with five vertical temperature gradients (1, 4.5, 6, 8, and 12 °C/m) between head and feet for seated subjects (sample size between 42 and 96 for each condition). The experiments were conducted in the Aachen comfort cube (length: 2 m; width: 2 m; height: 2.5 m) to simulate vehicles or meeting rooms with high thermal loads.…”

“…These studies of the effect of vertical temperature gradient on thermal discomfort allow no consistent conclusions for the following possible reasons. Wyon and Sandberg [4], Zhang [15], Yu [10], and Möhlenkamp [8] all observed that the mean room air temperature or operative temperature affects dissatisfaction with vertical temperature gradient. The overall room thermal environment that affects whole-body thermal sensation is a dominant stimulus for overall and local thermal discomfort.…”

“…The limit of 3 °C/m between head and feet is stipulated in ASHRAE 55 and ISO 7730 (Category C buildings), based on the work of Olesen et al [3]. However, numerous laboratory and building field studies have reported that the limit could be higher without causing significant occupant discomfort: 4°C/m by Wyon and Sandberg [4], 5°C/m by Tanaka et al [5] and Yu et al [6], and 8 °C/m by Liu et al [7] and Möhlenkamp et al [8]. The considerable inconsistency indicates that there might be other influencing variables, beyond differences in experiment size and setup, that have not been considered in previous studies.…”