So much to know...

Shawna HendersonNovember 18, 2014

It's hard to wrap your head around the amount of number crunching that goes into energy analysis, and how to prioritize your time and effort when working on a project. There are energy modelling tools that can be used that make it easier, but in a lot of cases, the easier the modelling tool is to use, and the less actual, real-world performance testing data you input, the more likely the result will be a ball-park figure that doesn't relate to actual energy usage, or actual heating or cooling loads.

Here's a good example from Jeremy Begby's November 17, 2014 blog post at IE3. He ran a comparison in WrightSoft (one of the industry's standard software  for heat loss/heat gain calculations) using a real-world example. In the first run, they used the default 'loose' air leakage rate, and in the second run, they used the actual blower door test results (single point method). The results were dramatically different:

  • The heating load using the blower door results was 1.4 times higher than when the software's generic 'loose' envelope rating was applied
  • The cooling load was another 3/4 ton higher using the blower door test results
  • While infiltration accounted for 36% of the total heating and cooling load using the 'loose' envelope rating native to the software, the blower door test results indicated that infiltration made up a whopping 64% of the total load

The estimate not only undersizes the equipment significantly, but also understates the importance of improving the building envelope when upgrading the HVAC systems in a house. Using real-world data points like blower door tests becomes much more important as you start to approach low/zero energy space conditioning and water heating targets in existing housing.

For example:


Last month, I wrote an article for Oil and Energy Magazine that looked at passive solar design, and I sketched out some of the ways that adding indirect gain in the form of an attached sunspace or closed-in porch can improve the building envelope. I don't model a sunspace addition in WrightSoft, instead, I build out the house-as-is model, ideally using the blower door test results. I typically cross-reference that with a monthly energy use model built in Hot2000 (because I'm Canadian, eh? Plus, I've been using it for over 2 decades, and can consistently get within ±10% of actual energy use).

Taking the as-is model, I run several scenarios for envelope improvements (first air sealing reductions, then additional insulation, then window replacements).

Then I model the proposed sunspace design using an excel sheet that I built out a billion years ago based on manual heat loss/heat gain protocols. I have it set up so it gives me monthly gain/loss totals for the sunspace, so I can see what I need to do to reduce monthly losses and when I need to incorporate shading or increase overhang sizes. I also calculate the heat loss 'credit' the house achieves with the buffering effect associated with the reduced area of wall/foundation area that is exposed to the elements and seasonal temperature extremes. Here's part of what that analysis looks like, this one with a wormy bit of ductwork in the top of an attached greenhouse for heat (but not air) transfer between the food- (and moisture-) producing greenhouse and the living space:

In addition, I do seasonal shading modelling when there is SouthWest and/or West-facing glass to make sure that there will be no sneaky overheating in the summer months I especially like this suite of tools for solar design modelling from Sustainable By Design, out of Washington state.

The only thing missing in my analysis is real-world monitoring. I follow up to find out what energy bills are when I can, and there is a good correlation between the model and the actual use, where the work was completed as specified (that's another topic for discussion...). In an ideal world, clients would jump at paying for the privilege to be monitored. But, mainly, they just pay me for consulting.


Leave a comment

Please note, comments must be approved before they are published