GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted
Guest Blogs

Energy Predictions vs. Energy Performance

An analysis of energy use in the first year of living at Blue Heron EcoHaus

Energy modeling is one thing, but living in the house and measuring energy performance is another. The author of a construction blog looks back on the first year of occupancy in this Canadian superinsulated house.
Image Credit: Kent Earle

Editor’s note: Kent Earle and his wife, Darcie, documented construction of their superinsulated house on the Canadian prairies in a blog called Blue Heron EcoHaus. GBA published a number of those posts in a series that wrapped up last year. Recently, Kent wrote to say he has been monitoring energy use at the house and offered this followup.

We had spent a lot of time planning and designing a house that would be energy-efficient, aesthetically pleasing, and cost-effective. This is a fine balance to try to find. However, it really is a big guessing game until you actually live in the space and track its performance. You can run all of the computer programs you want, but you really don’t know how things will be until you’ve moved in and you’re living your normal life.

We had installed PV panels on the house to compensate for some of our energy use, with the hope that someday we could work towards a net-zero energy home. But this, too, seemed to be a big guess as to how well it would perform.

In the planning and designing stages of the house we ran a couple different energy models on the house. The first is called the HOT2000 program. This is an energy simulation and design tool for low-rise residential buildings. The software is widely used across Canada to support program, policy, and regulatory development and implementation. HOT2000 is developed and managed by the Office of Energy Efficiency at Natural Resources Canada (NRCAN). It was originally designed for use with the R2000 energy efficiency program, which was an early promoter of green home building in Canada.

We later used the Passive House Planning Package (PHPP), which is now a widely used software program for building highly efficient homes worldwide.

My intent with this article is to report the varying predictions of the these two programs, as well as our predicted solar generation, and also to show our actual energy use for the year of 2016 – our first full year in the new house. I’ll also report some considerations and possible options for the future.

Pre-build energy modeling

I had been very curious about this when we were in the early stages of planning the house. Most of what I read concerned predicted energy use in various homes, but I’d only come across one house that actually tracked and reported its energy use — the Mill Creek Net Zero House in Edmonton, Alberta. Although it used exceptionally little energy, it did not meet its net-zero target. That being said, it was very close.

I fully did not expect our home to be anywhere close to net zero, but we hoped that over the next number of years we could gradually build our solar panel array (as costs come down) to eventually reach our goal.

OK, let’s get to the numbers:

HOT2000 Predictions:

Annual space heating energy consumption: 7,159 kilowatt-hours (kWh).

Annual domestic hot water (DHW) energy consumption: 3,409 kWh.

Annual appliance energy consumption: 8,760 kWh.

TOTAL = 19,328 kWh/year.

PHPP Predictions:

Annual space heating energy consumption: 7,584 kWh.

Annual DHW energy consumption: 3,974 kWh.

Annual appliance energy consumption: 11,310 kWh.

TOTAL = 22,868 kWh/year.

PV Array Predictions (6.2 kW)

PHPP estimation: 7,321 kWh/year.

Solar installer’s estimation: 9,300 kWh/year.

So obviously there are discrepancies between the HOT2000 and the PHPP. Although their predictions of heating and DHW are quite close, surprisingly the appliance use was significantly different. Also, surprising, was the discrepancy in the solar predictions – I was a bit disconcerted by the drastic difference of 2,000 kWh/year!

For comparison’s sake, according to Stats Canada website’s most recent 2011 home energy use data, a Saskatchewan home consumes an average of 30,555 kWh/year (110 GJ), of which electricity for appliance use is 8,889 kWh/year (32 GJ).

Drum roll, please

Actually, first some clarifications. All I have is our actual overall energy use. I cannot separate out heating vs. DHW vs. appliances, unfortunately, although this would be interesting. The following information is taken from the solar panel’s generation and the electrical meter. I tracked each month and have recorded it below.

OK, now the drum roll.

Actual energy consumption and PV generation

January: Solar generated = 315 kWh vs. energy use = 3,323 kWh.

Yikes! I was a pretty worried when I saw this. That being said, January was very cold and has very short, dark days (-20° to -30°Celsius most days, or -4° to -22° F. We kept the house around 71°F).

February: Solar generated = 553 kWh vs. energy use = 2,706 kWh.

February is always a cold month. Although you can see the solar was getting a bit more sunlight already as the days lengthened.

March: Solar generated = 603 kWh vs. energy use = 1,716 kWh.

This was getting a bit better still. I lowered the house temperature to 69°F. It was getting warmer outside and we were getting more solar gain.

April: Solar generated = 979 kWh vs. energy use = 1,385 kWh.

April was warm and sunny. Nice spring weather. Started to not need the in-floor heat on at all during the day, but still ran it during the night.

May: Solar generated = 960 kWh vs. energy use = 1,029 kWh.

Almost net zero for the month. It was a very nice month. We were running our river pump frequently to water new grass, which I think increased energy use quite a lot.

June: Solar generated = 1,434 kWh vs. energy use = 989 kWh.

Net positive in a big way. Beautiful month. Obviously the longest days of the year.

July: Solar generated = 956 kWh vs. energy use = 511 kWh.

The first two weeks of July were cloudy and rainy which is unusual for July.

August: Solar generated = 950 kWh vs. energy use = 645 kWh.

This month was very rainy as well, which again, is not normal. August is usually very hot.

September: Solar generated = 778 kWh vs. energy use = 611 kWh.

Cool and cloudy. I replanted grass seed and was running the river pump a lot again.

October: Solar generated = 315 kWh vs. energy use = 1,478 kWh.

October sucked! 315 kWh is the same as January! It snowed on October 4. We had to turn the heat back on. There were only two or three sunny days all month.

November: Solar generated = 390 kWh vs. energy use = 1,750 kWh

Cloudy month, but we had some mild days mid-month with above-freezing temperatures. Still, we generated more solar in November than we did in October, which should not happen.

December: Solar generated = 229 kWh vs. energy use = 2,857 kWh

Shortest days of the year and extremely cold (-40°F). What do you expect?

Actual totals:

Actual solar PV generated = 8,189 kWh.

Actual household energy consumed = 19,000 kWh.

Actual total net energy used (consumption – PV) = 10,811 kWh.

Energy use was less than expected

I’m extremely pleased with these numbers! I’ve been waiting for two and a half years to know what our actual energy use would be.

The annual output of the 6.2 kW solar array at the Canadian superinsulated home of Kent and Darcie Earle was close to predictions and, overall, supplied about 45% of total energy use.

We actually used less overall energy then was predicted by either the HOT2000 (19,328 kWh/year, although it was close) and a lot less than was predicted by PHPP (22,868 kWh/year), which is surprising. It makes me wonder how close we would be to meeting the Passive House standard given that our actual energy use is 3,868 kWh less than it predicted… Hmm. Maybe we should have tried to hit that airtightness target of 0.6 ach50 after all. Oh well.

Nonetheless, the overall energy use of 19,000 kWh is very good (and such a nice round number, too!). We did not do anything different in terms of our behavior, except to just be smart and not be wasteful. I still baked bread every weekend and we used our larger appliances just like we normally would. We have two refrigerators and two deep freezers in the house. All the lights are LEDs. We try to hang our clothes to dry. We used our wood stove occasionally, maybe two or three times per week, but mostly just for ambiance and occasionally on the extremely cold days.

That being said, based on the predicted numbers, the heating energy likely accounts for about 50% of our overall energy use. That makes me wonder, too, how much better we could do if we burned wood a bit more often?

PV estimates are close

As for the actual solar PV generation (8,189 kWh), it pretty well split the difference between the installer’s predicted 9,300 kWh/year and the PHPP prediction of 7,321 kWh/year. I think this past year was on the cloudier side for sure. We had a lot of rain in the spring and even more in the fall, which is very unusual. Followed by an extremely early snowfall, which seriously cut into our PV generation (see October – brutal). It probably would be closer to the installer’s prediction on a typical year. (We’ll have to see what 2017 brings.)

Still, based on the actual numbers, our solar panels did cover nearly 45% of our overall energy use for 2016. We would, however, need to double the number of our solar panels (add another 6.2-kW array) to meet net zero consistently. Who knows, maybe in the coming years the costs will drop more and perhaps government incentives will increase. One can hope.

Comparing our house to the average Saskatchewan home consumption of 30,555 kWh, we did very well. We used 37% less energy then the average home. And when you take into account the solar energy generated, that drops us further, to 65% less energy then the average house! Sweetness.

Considering that we are completely on electric energy, it makes sense to make the house as energy-efficient as possible. The cost of electricity for us is $0.12224/kWh (while the cost of natural gas power is about $0.04/kWh equivalent), which works out to an electricity bill of $1,321.54/year (10,811 kWh x 0.12224).

We do, however, have to pay a basic service fee of $32.61/month (even when we are net positive in a month), which sucks, and then 5% tax. That brings our absolute costs for the year to $1,798.50/year or $149.88/month, which is about half the cost of our previous homes power and electricity bill.

I’m OK with that.


  1. Dana1 | | #1

    PV output is higher in the early years.
    A silicon PV array usually puts out 5-10% more power than the panel ratings would suggest in the first few years, and degrades slowly over time. The rate and degree of degradation varies with vendor & technology. So all else being equal (which it isn't- weather & insolation are highly variable), it's not surprising that the output would exceed simple-model predictions.

  2. user-1119494 | | #2

    Thanks for numbers!
    Always nice to see how reality compares to predictions.

    I suppose that, if one were trying to live off grid or better match output to need, one would tilt the panels far higher (60deg? 70deg?) and triple or quadruple the panels... Might be better to keep paying the monthly connection fee for now!

  3. Expert Member

    Thanks for the update. It's great to be able to see how the house measures up. So many of the projects featured here disappear into a void once the final instalment of the construction blog is posted.

    I think both the Hotcan and PHPP predictions were close enough to what you have ended up using to be reassuring about the decisions you made based on them. They aren't spot-on, but any program that can come that close is very useful.

    Nice house you seem to enjoy living in. Not too many compromises to achieve the energy savings. Looks like you achieved your goals. Well done.

  4. AlanB4 | | #4

    Thanks for the numbers, they
    Thanks for the numbers, they are very interesting.

  5. StollerB | | #5

    Great work Kent! Thanks for

    Great work Kent! Thanks for sharing such valuable information. It was especially helpful for you to provide the comparison between modelled and actual values, and very interesting to see the differences between Hot2000 and PHPP. Fascinating that, even with its incredibly detailed inputs, PHPP so grossly overestimated your annual energy consumption, and on what seemed to be a quite unusually cloudy and cool fall as well. Had it been a more typical fall, your heat gains through windows and reduced building assembly losses would have lowered your heating energy consumption further. Since you straddled the standard PHPP input for interior temp (21 degrees C) your heating energy should have been more accurate. Perhaps the semi-regular wood stove use did throw that number off more than one would suspect? As Marc Rosenbaum (or some other unacknowledged person before him!) said: "There's no such thing as Net Zero houses- only Net Zero families!"

    Since air leakage can have a big impact on heating energy, I'm curious if you performed a blower door test on the house, and how that result compared to the infiltration rate you had used in the PHPP and HOT2000 modelling.

    Anyway, great article, and thanks again for sharing!

  6. David Murakami Wood | | #6

    Thank-you for this, Kent. We've been following your project as we started our own on Wolfe Island in Ontario. Out of financial necessity we've been unable to install PV initially and have decided to put this to our long term advantage by using the first panel-less year and a bit as a baseline to arrive at a more accurate assessment than the models. We of course have the advantage of being a little further south and the temperatures never dropping as low as where you are. I wanted to ask whether you had considered supplementing your PV array with a small wind turbine? We've certainly been considering that here. We really appreciate all your posts.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |