There was a fuss in the media a few days ago about the dangers of overheating in homes due to the Green Deal. The article pointed out that highly insulated homes tend to keep heat in (surprise!) and that this can lead to overheating in the summer months, particularly in hotter times like those we are experiencing today (30 DegC). What the article failed to point out is that insulation also keeps heat out, just as well as it keep it in, insulation works in both directions. But let’s not let science get in the way of a headline.
I have worked on a number of projects where overheating was a concern and we either carried out analysis ourselves or we brought in other consultants to do it for us. In all cases we used Dynamic Simulation to do the work. For those of you who haven’t come across this, Dynamic Simulation involves calculating the heat losses and gains for a building, or part of a building, for a year. The gains and losses are based on an understanding of the use the building is going to be put to, the number of occupants, the orientation of the building, and all the thermal characteristics of the fabric and services. These calculations also use a weather file that contains recent actual temperatures and hours of sunlight for that area. The resultant calculation can determine where the overheating is likely to occur in a building and the data can show why it would occur, as the simulation is carried out in steps of 10 minute intervals or any other interval preferred by the person doing the work. The performance at every step is stored and can be viewed separately or as part of a daily, weekly or monthly chart to investigated trends and frequency of overheating.
The point of this is that overheating is something that we talk about regularly in housing as our climate gets progressively more unpredictable. The future may be uncertain, but higher standards for insulation and airtightness are guaranteed, and with them comes a higher likelihood of overheating. But fortunately we have the solution to hand. Dynamic Simulation. This is the tool that can warn us when and where in our buildings the overheating is most likely to occur, and having warned us of it, it can help us to simulate how to prevent it. We can simulate opening of windows by residents, mechanical ventilation controlled by temperature, the impact of external blinds or shutters, and anything else that we can apply to the buildings. If we can design it then we also have the ability to simulate it.
SAP contains an overheating check, but it is almost worthless, as it cannot tell you why the overheating is happening and when. This is because most of its calculations are annual and some are monthly. To be confident about overheating you need to use a proper tool.
There is no excuse for housing designers to design properties that overheat, despite residents efforts to reduce it. We need to provide residents with buildings that remain comfortable in normal conditions and allow for some protection using blinds or ventilation in unusual conditions. We need to simulate our designs and test their performance in hotter weather. If we don’t then our occupiers will go out and buy air conditioners and any CO2 savings that we thought that we had made by insulating the buildings and making them airtight will be lost.
Standards to pay attention to: EN 15251 Indoor Environmental Criteria
rory bergin's blog 
Pingback: To prevent overheating in homes the first rule is to keep heat out
Whilst i agree with your arguments, Dynamic Simulation can add unnecessarily to the design fees in most situations where a judgement can be made based on window directions, sizes, glazing type and shading.
As usual, the devil is in the details. For single family dwellings this may be true, for multi-family dwellings it is unlikely to be possible to optimise each living space and simulation will help the designer to quickly assess the impact. BIM integration between design and simulation tools is taking much of the time cost out of doing this.
Two weeks ago the Good Homes Alliance organized an event on “Climate Change and Overheating” and the program and presentations can be seen here: http://www.goodhomes.org.uk/events/138
Thanks Christian
An interesting article and timely considering the heat wave the UK is currently experiencing. I’m sure I’m not alone in having to unusually engage my brain at home trying to work out the best way to stay cool. Why is it we know exactly what we need to do at our place of work but when it comes to our homes we struggle? I suspect I’m not the only who finds closing the blinds at home to minimise the solar gain contradicting every urge I have.
Why do we resist using Dynamic simulation within residential homes? Regardless of scale, I would suggest it’s vital to any home. Ok, the budgets may be tighter but don’t we have to model SAP anyway?, so why not push a few more buttons to analyse Daylighting, air quality, energy, heating, glazing etc. As we shrink-wrap our homes and insulate to record levels we need to understand exactly what the impacts of our design decisions are going to be. Air quality concern is rising as these new passive-super-duper-haus’ bed in, issues with condensation are being realised. I heard recently of a house that used 1.75km (yes, I did mean to type kilometres..) of ducktape to achieve the required air permeability level, you wouldn’t be keen to raise young children in that house, with all the VOC’s from the glue…
With never seen before complexity and minimal margins of error within our residential buildings, rules of thumb simply can’t be relied on in today’s design teams. Nothing can substitute experience and knowledge, but Dynamic Simulation modelling certainly supports it.
Thanks Ian, I wonder what you think of the argument that tools like PHPP are more appropriate to residential uses because they are ‘simpler’ than Dynamic Simulation tools. I would be interested to hear how long it takes to model a home and carry out a simulations in IES versus modelling the same home in PHPP.
I have also been asked whether DSM vendors do any work to verify the results of their modelling. I know of a protocol called P-Star which involves calibrating a DSM model against actual results to carry out an annual simulation. Have you any research in this area?