Reality Show: Monitored Passive House results from Salem, Oregon

All — thanks for all of your contributions and comments about fine-tuning the standard. It’s going to be an exciting process. Continuing that discussion, let’s look at a few really good examples of certified Passive Houses that were modeled for various North American climate zones, and for which we have good monitoring data. The graphic below makes clear that generally, the climates of North America and Central Europe are not directly comparable. One small region–running from the northwest U.S. Coast into Canada–matches the Central European conditions.


Therefore, we’ll look first at a certified project in Salem, Ore., and evaluate how accurately the PHPP modeled the actual monitored experience.

The Salem, Ore., home that's been monitored for a year.

Again — as stated in the inaugural blog post, the core principles behind the Passive House concept, some of which date back to the early 1970s — are not in question. Minimizing the peak loads to a point when balancing the ins and outs (losses and gains) produces  a building that nearly reaches equilibrium. Such a building needs very little active energy input — and this only a few months of the year — to maintain comfort.

If the space conditioning meets our 1 W/sqft peak heating load and 0.8 W/sq peak cooling load requirements, then we get the icing on the cake: For mechanicals, we can either use point sources throughout the space or integrate the space conditioning in the ventilation air. In the Northwest, with next to no cooling requirement and lots of passive cooling potential, integrating conditioning and ventilation could prove to be the most cost effective solution.

Let’s look at how one Passive House project played out in Salem, Ore. The 16th & Nebraska project (also known as the Rue-Evans House, named for its owners) was built by Blake Bilyeu and his father. Blake Bilyeu is a pioneer — he took one of the first PHIUS Certified Passive House Consultant training courses offered. And by 2010, he had completed the 16th & Nebraska project. It became one of the first projects certified by PHIUS.

By U.S. Department of Energy climate zone definitions, Salem is considered a marine climate, characterized by:

  • mean temperature of coldest month between 27-65 F
  • warmest month mean of less than 72 F
  • at least four months with mean temperatures over 50 F
  • dry season in summer (month with heaviest precipitation at least 3x of driest month)

Here’s the Salem climate data at a glance:

And data for Bonn, Germany.

The climates are very similar: Average temperatures in Salem are a little higher by 4.6 F in winter in Salem, summer highs are the same, and precipitation is generally higher in the Pacific Northwest. Both locations have limited solar availability.

The project’s PHPP data at a glance confirms that both Passive House criteria are met: the annual heating demand criterion with 4.02 kBTU/sqft yr as well as the peak heat load with 2.9 BTU/hr.sqft. There is no need for cooling.

The general specifications of the exterior envelope components are:

  • R-45 in the wall, the roof has R-96, the floor over crawl space has R-51.
  • The average window installed U-value is 0.226 BTU/hr.sqft.F with orientation specific SHGC of 0.23 for E-N-W orientations and 0.46 SHGC for the South

Note: The window figures are unadjusted NFRC values. Passive House window calculations will result in slightly adjusted values. The PHIUS Technical Committee is developing a method for converting values and/or a protocol to more accurately calculate the window values needed for PHPP. The Tech Committee will make it available for comment in an upcoming PHIUS e-Newsletter.  Many thanks to Graham Irwin, John Semmelhack and Graham Wright, who already have devoted a lot of hours to this project!

Energy consumption at the project was monitored for a full year. Over that period, the home was occupied by its new owners (a young couple and a dog, and eventually the couple’s newborn baby. The owners blogged about their early experience — check it out.)

A detailed monitoring report on the first year was prepared by the company Ecotope. (Many thanks to the Ecotope team that graciously gave PHIUS permission to make the information available to the Passive House community. PHIUS plans to share more monitored data from other projects soon—stay tuned.)

Download the full report here. From the executive summary:

The 16th and Nebraska Passive House project located in Salem, Oregon is an impressive example of an energy efficient home. The home is built to the stringent requirements of the Passive House (PH) program. The home’s energy use for the first year post-construction place it in the top tier of the most energy efficient single family homes in the Pacific Northwest. The first-year stats for the project are listed below:

  • First Year Total Annual Energy Use: 5,413 kWh/yr
  • Electric Utility Cost per Year: $700
  • Energy Savings estimate over Oregon Code: 9,064 kWh/yr
  • EUI (using gross sq ft of 1,885 sf): 9.8 kBtu/sf/yr2

The Salem Passive House home blows away today’s code homes, and nearly meets — right now — Ed Mazria’s Architecture 2030  Challenge for 80% reduction:

Back to the original questions: how do actual results compare to what was modeled in the PHPP, and what conclusions can be drawn from potential deviations of the measured results to the modeled results?

To start, some of the results are rather unexpected:

The actual first year’s energy use came in between what was predicted and what is allowed in the Passive House program, 5,413 kWh/yr. This results in a 63% energy use reduction over an electrically heated home built to the Oregon 2008 Code. This represents a savings of $750/yr in utility costs.

The DOE Building America Program aims for 70% in overall energy reduction. With 63%, this is below what one might have expected. If the modeled results had actually been met the building would have been saving 73% over the code home. Reading on:

This home represents the upper limits of conservation that can be controlled by the designers and builders; or what can be achieved by applying most of the energy efficiency measures currently available. Space heating and DHW represent 13% of the home’s total energy use. The remaining loads are plugs, appliances, lights, cooling, and energy recovery ventilator (ERV) fans. Since PH is a modeled certification program, there is no guarantee that a home modeled to meet the PH standard will actually perform to the PH standard once built. It is clear that maintaining the PH energy use levels is a function of occupant behavior and lifestyle choices. More research and development of tools for modeling plugs and appliances in the PHPP program should be made available to the PH community.

The report points out higher plug loads and attributes this to the American lifestyle. As we have compared electrical loads modeled in PHPP and actual consumption, we find a large discrepancy between what’s modeled and actual results throughout many North American projects which seems to confirm the author’s explanation.  

Conclusion: We need more accurate protocols for the American household. We need to identify realistic stringent savings recommendations and adjust the initial assumptions in the household electricity sheet accordingly. This measured result also points to the potentially higher importance of the source energy criterion rather than the focus on annual heating demand.

The measured space heating demand constitutes only 1/4 of what was actually predicted. Once the additional household consumption is taken into account, though, in the internal heat gains the results once again are pretty close to what PHPP predicted. The modeled kWh amount for space heat in the diagram is reflecting the assumed 3.2 COP of the heat pump; if provided through direct resistance  that would amount to 3.2 times 694 kWh,  equaling a little over 1800 kWh/yr. The discrepancy is roughly the additional kWh use for the DVR and server, which is direct electric internal heat gain (no bonus through COP). That explains the total higher energy consumption.

In short, the predicted space heat demand result is indeed very close to the modeled prediction. On the flip side though, there is talk about energy used for cooling in the report and the PHPP modeled 0 energy use for cooling. Additional household consumption can be used to replace heating needs in winter, in summer it adds to the energy used for cooling needs.

The report concludes:

This 16th & Nebraska Passive House home represents the leading edge of current energy conservation. Insulation has been maxed out, the envelope is extremely air tight, and the glazing percentage has been reduced to 15%. The solar water heating system is providing 69% of the hot water energy, the home uses highly efficient appliances, low lighting levels, and a very efficient ERV. The remaining loads are a product of the American lifestyle and are the hardest loads to control without major impacts to lifestyle.

For the Northwest, the Salem example indicates that the PHPP is, as expected, reasonably accurate in predicting the annual space conditioning. Assumptions for household and plug loads need to be revisited and entered as correctly as possible, but this is not a climate but rather a market issue.

Just as in Germany, the Salem project proves Passive House is a good basis for net zero. From the report:

This project is a great example of what is possible with the PH program and represents one path to achieving net zero energy use. A 6-7 kW PV array on the roof with the current 3 people living in the home would take this project to a net zero energy status.

That’s a sizable PV installation, but, just like in Germany, forgoing active energy in favor of conservation is more cost-effective.

So we’ve seen what happens when we apply Passive House in a climate very close to the European climate. To predict space conditioning it works very well. American life style and market, we have some work to do.

Applicability in regards to humidity, hygrothermal concerns and the impacts on airtightness recommendations for this climate will be addressed separately at a later time.

Next, we’ll venture into more extreme North American climates to evaluate that experience. Stay tuned for more!


10 thoughts on “Reality Show: Monitored Passive House results from Salem, Oregon

  1. Nice report and coverage

    This is what we need to see more often if we want to get better at predicting homes energy usages. There are a couple of questions/ comments from me and would appreciate any feedback:

    1) Was the server (500 kWh/SF YR) accounted for in the PHPP? If not this would bring prediction and usage very close and a server is definitely not included in the standard values.

    2) Unfortunately I could not figure out with energy usages have been estimated (and if how) and which were measured. The way I understood the report is that no individual data was available which would leave the results questionable (even though there where not bad). However I agree on revising the standard numbers for plug loads, lights, hot water, etc. for cultural reasons but this should`t be so hard and maybe done a long time ago.

    3) The biggest challenge I have with the report is the lack of information on how the numbers of the code house came about! The heating demand amounts to 40 kWh/(SF YR) (TFA based) which is very low. I am not an expert on Oregon code but this doesn`t seem right. I am skeptical because 2 reasons:
    a) A house that uses 40 kWh/ (SF YR) for heating is considered a low energy house in Germany and gets government incentives
    b) I modeled a home for MN code and moving it to Portland OR it shows out 80 kWh/ (SF YR) annual heating demand


    • Phillip,

      Your comment is well taken. Neither the server nor the DVR were accounted for. That does point towards the need to interview the homeowner with much attention to detail in regards what they are intending to have in their home. That is not always clear in the beginning of a project, even close to the move-in date sometimes hard to predict.

      A couple of ways to deal with that: figuring some higher defaults that more closely represent the plug loads in American households and that are automatically worked into the per person usage assumption or to hope for the best to get the most accurate information from the home owners to be (difficult obviously if it is a spec house).

      This was by the way also the biggest difference we found in our investigation how REM/Rate and PHPP compare. The predictions in regards to heating and cooling demand were actually closer than we would have expected. The household electricity was off in some cases by a factor of 7!

      I agree with you, the report does not reveal any information on the code home comparison and how the numbers were derived and it indeed seems low. The PH discussed here uses 1.2 kWh/sqft yr for heating. If we use the factor 10 rule of thumb of PH reduction in heating energy compared to existing building stock, then the existing home should be using roughly 12 kWh/sqft yr (TFA – I think your numbers were meant to read in metric). This would put the existing home at 18000 kWh just for heating if indeed electrically heated. You are right, they are assuming a very good home in comparison here. The reduction factor through PH measures is more like 6 or 7.

      Still, 63% over code home! With optimization possible to get to 70%. I think this is a great result.

      • Kat,

        Thanks for the reply. In fact all my numbers were metric, sorry for the typo. I would be interested if your REM/Rate PHPP compare showed larger differences if the house is occupied by much fewer people than the PHPP verification mode accounts for. The PHPP figures a lot of the electricity consumption in correlation to people. In my opinion this might not be very accurate because many loads might not be linear to # of people. And as we all know the real number of occupants and # of occupants shown in the PHPP verification mode often quite differ in the US. Do you have any studies or publications on this?

        One more comment on the one year billing report: If all the data they had was the total energy consumption I am impressed by the results drawn from it. However it is a slippery slope in my opinion….

        • Agreed, zooming in in more detail is definitely needed. In this particular case it might have influenced the report/monitoring decisions that the company who did it was not directly involved with the Passive House design or PHPP calculations. This is good and not so good. Not so good because the details of different calculation conventions and necessity to reflect those in the monitoring set up might have gone overlooked. On the other hand, if an outside third party organization with no stake in the outcome evaluates (and I understand that the company who prepared the report had been somewhat skeptical at first), then to convince them and get them excited about the results is worth quite a bit. They did dial down the success somewhat by choosing, as you pointed out accurately, an already very efficient comparison baseline but even then, still 63% total energy consumption reduction! I think this is a full success for PH in the North West.

  2. Kat:
    Good to see that PHPP is reasonably projecting expected results. Comparing at Passive House projects in varied climates is necessary to increase confidence in PHPP results.

    1) Is it safe to say the American lifestyle includes more computers and other electronic media devices that produce both increased PE load and IHG? Increased IHG also indirectly increases need for cooling – thus further increasing PE?

    2) I, and believe other CPHCs, have a good grasp of PHPP for use in heating demand determination by balancing Transmission and ventilation losses with radiation and internal gains in a heating dominated climate.

    Where I would benefit is developing a better understanding of cooling/dehumidification balancing and other connected load calculations.
    Using PHPP for a hot dry or hot/humid climate for a residence may help in walking through the process while developing a better understanding of what concepts/strategies PHPP is attempting to calculate for cooling (loosing heat) and other connected loads.
    In short, training specific to tabs Summer, Shading-S, SummVent, Cooling, Cooling Units……..through IHG would help.
    I believe a better understanding would apply to mixed climate summertime and non-residential all the time (presuming non-residential has high IHG and therefore elevated need for cooling).

    3) I continue to look for what IECC 2009 and 2012 standard reference residential building (defined in chapter 4 of IECC) will result in on Passive House measures (Annual heat demand, annual cooling demand, annual primary energy demand, and air leakage at 50 Pa (1 lb./ft2). The only direct answer in code language is air leakage.
    If the measures were known/published, comparing Passive House to code would be greatly simplified. I will continue to ask those training in the IECC for this information, but would appreciate knowing if anyone at PHIUS or the US DOE knows this information.
    Until then, is it reasonable to predict that Passive House will result in 80 to 90% reduction in annual heating demand and 55 to 65% reduction in primary energy demand? I understand this spread will decrease if each new (every 3 years) IECC makes a 15% improvement.
    The IECC has a presumed IHG (gain) formula, presumed ventilation rate, and no mention of solar radiation gains. Thus it is in essence only asking the designer to address transmission loss (or as they term the UA).
    Having this information would better my ability to explain Passive House to those who may have interest.

    Appreciate any thoughts.

    • Joe,

      Your request for advanced training in climate specific design problem solving and PHPP use reflects what we have heard from many other fellow CPHCs.

      The PHIUS Certified trainer group (:// is responding to that request from the community and has started to plan for advanced training sessions. The advanced (and I mean advanced, this session is not for wussies :) ) THERM and WUFI combined workshop rolled out earlier this January in Chicago and is currently being scheduled at most of the 2012 CPHC training locations throughout the country. Final dates will be available very soon. A second 2-day part to the advanced THERM/WUFI training has only been conducted in pieces so far but is currently being put together for advanced integrated mechanicals and advanced PHPP climate specific design solutions. This makes for a fully loaded 4-day advanced course for already seasoned CPHCs. More info soon!

      Your advanced PHPP suggestions are extremely well taken and we will definitely work them into the design of that advanced PHPP design day.

      Yes on higher electrical consumption, yes on consequentially higher IHG, more cooling and higher PE. Plug loads are one of the issues and differences (in this case cultural) we are hoping to fine-tune during this comparative exercise of PHs in various North American climates.


  3. Is it really enough to claim that since Bonn and Salem have similar climates based on the Köppen–Geiger climate map – a map which is not without valid criticism?The Trewartha might be a better map for comparing climates… unless you think Philadelphia and Miami should be in the same climate zone… It also shows that much of the US is actually fairly comparable to EU.

    To validate if the climates really are comparable, we’d have to know how the house would be predicted to perform in both locations… I’d be interested to know where the submitted 16th & Nebraska’s specific space heat demand would be when the model is migrated to Koeln (closest PHPP climate to Bonn, ~20 miles/36 klicks). Any project we’ve taken from Portland/Salem to Koeln has resulted in a 2.5-4.5x increase in SSHD. This isn’t a unique occurrence, as when we’ve migrated a project we’ve done (or someone else has done) to comparable climate/HDD locale in EU, we’ve consistently seen worse specific space heat demands than in the US. Additionally, much of the northeast (Philly/NYC/Boston) is colder than either Central Europe or said ‘NW sliver’ – but it is still easier to achieve PH there than here in Seattle – and considerably easier there than in EU.

    Furthermore, if we’re going to run around educating people that PH and the 2000W Gesellschaft are complementary, than I still don’t think it is responsible to fudge U.S. electrical defaults to conform with our poor standards of consumption – rather, we should be educating and attempting to do the opposite. That’s not to say consultants shouldn’t be auditing planned appliance/equipment usage – as that would be highly prudent.

  4. Nice report! Glad to see some monitored results. I understand that the modeled results came pretty close to the actual consumption. Mike highlighted the different climate maps and data sets and some may be more accurate than others. What if some of the data set we use in PHPP is not as good as this one?

    So, I’ve always wondered why PHIUS does not require monitoring as a say, one year anniversary verification? I guess if the home performs to the predictions, the homeowner would only be interested in monitoring for a couple of months, then they just pay their minuscule utility bill and stop looking at the monitor…

    • Roger,

      Happy to hear you say this. Many of the recent conversations about the value and accuracy of modeling (it is theory in the end – lets face it, and only as good as the assumptions are to begin with) are echoing the same sentiment.

      We are thinking about offering highly discounted monitoring systems to every project that certifies under PHIUS+. There are more people interested in this feedback than one would think. So, lets provide it! And best of all, the PH community keeps collecting monitored data to assure quality of the model and improve the assumptions. How else would one verify one’s assumptions? “No feelings without the data”, as Thorsten Chlupp from Alaska said so famously during the 2010 5th national PH conference in Portland OR.

      It also greatly helps to fine-tune the mechanical system after a few months in operation, as you are pointing out. If performance turns out to be grossly off, one can go back in and look for a mistake that might have been made in more detail and hopefully correct for it.

  5. I’m confused about the overall energy use allowed by PHPP. The bar graph shows 6462. I thought the standard was 120 kwh/sq m or 11 kwh/sq ft/year. For a 1500 square foot house this would be 16,500 kwh. What am I missing?

Leave a Reply

Your email address will not be published. Required fields are marked *