Complaints from building occupants are one of the most difficult challenges that the facilities manager will experience. It’s a common problem, especially in open plan offices where those who feel the cold wear as many layers as they possibly can and huddle around a warm cup of coffee all day, while others sit in short sleeves all year round, trying to keep cool. It’s a challenge faced by most energy and facility managers and a massive challenge to keep complaints about thermal comfort to a minimum.
So when NOVICE project suggested that HVAC equipment be turned off for 2 hours in a leisure centre with a pool, some people thought it was a crazy idea. But we wanted to find out how much load shedding potential there is in a typical leisure centre. We decided to use the opportunity to find out what would be the impact of doing so on indoor conditions and, more importantly, would anyone even notice?
The first task was to identify which equipment could be safely turned off without impacting heavily on internal conditions like temperature and humidity and keeping CO2 levels within acceptable ranges. The team of energy experts working on the project identified all the non-essential HVAC equipment in the leisure centre, which included air handling units, pumps and fans.
To determine the impact on thermal comfort, we designed a survey to be undertaken with building users. The main aim of the survey was to determine whether or not the people who were in the building during the demand response event noticed any change in conditions during their visit. Crucially, to avoid any anchoring bias, we decided not to tell survey participants that we were running an experiment or had changed anything about normal building operation; we simply informed them that we were doing a thermal comfort survey to help us learn how to make the indoor environment more comfortable for them. We designed the survey to lead the participants into thinking we were only interested in their thermal comfort by asking a number of questions about which parts of the leisure centre they had visited, what their activity level had been during their visit and what was their overall satisfaction with temperature, humidity and air quality inside the building. The critical question that was asked at the very end was “Thinking about the time you have spent in the leisure centre today, did you notice a change in conditions during your time here?”
On the day of the test, a group of people was surveyed before any equipment was turned off – this group formed the control group as all of their answers related to the leisure centre under normal operating conditions. Then at 5pm on a Monday afternoon, the busiest time of the week for most leisure centres, all non-essential equipment was turned off for 2 hours. We continued to survey building users during and immediately after the simulated demand response event – this group formed the test group – all the time monitoring the actual temperature, humidity and CO2 levels to ensure that they did not drift outside acceptable ranges.
So what actually happened? After 2 hours with non-essential equipment turned off, conditions in the leisure centre had changed, but remained within acceptable ranges in all cases. In most areas, the temperature had drifted from its starting point by just 1.3°C, with reception getting slightly warmer while the pool and gym areas got slightly cooler. Humidity increased in all areas, and as expected, the largest increase was in the pool hall, where humidity rose from 70% to 99%. CO2 levels held roughly constant in the reception and pool hall, but increased to the edge of the acceptable range in the gym. But did anyone notice? The graph below summarises the results.
The majority of people did not notice any difference in conditions. Of those that said they did notice a difference, two of them were in the control group which means that if they did notice any change, it was nothing to do with our experiment. So 22% of people in the control group claimed to notice a change in conditions, when conditions were not changing. Only two people (25% of the test group) said they noticed a change in conditions that was matched by the data collected by the monitoring equipment. Effectively this means that there is almost no difference in perception of thermal comfort between the control group and the test group.
We carried out the same experiment in another two leisure centres to verify our results and check if they were repeatable. The results were mixed, with one leisure centre responding in exactly the same way as above, while the other had to end their experiment after 1.5 hours due to CO2 levels in the gym drifting outside of acceptable levels. We think that the smaller size and lack of natural ventilation in the gym were the cause of the quick rise in CO2 levels, but this was not proven, as our experiment was not designed to test which building parameters impact on indoor air quality. Still, in both cases, most building users did not have any complaints about their comfort and did not notice the change in conditions over time. One interesting anecdotal exception to this was a competitive swimmer who commented that their lap times were much slower than usual and (rightly) thought that the pool seemed more humid than usual!
This very small study shows that it is possible to participate in a load shedding demand response event without increasing the number of complaints received from occupants about their thermal comfort, even in relatively sensitive environments like leisure centres. The ability of a building to participate does depend on the particular conditions inside each building but it’s clear is that if building conditions change slowly enough, building users will not notice the changes as they occur and so there is no reason why more buildings can’t participate in demand response programmes. More case studies and experiments like this are needed in buildings of different types to persuade building energy managers to embrace the opportunities that are afforded to them by demand response, but our experience in the NOVICE project is a positive one.