Residential Renovation of a Schoolhouse

A Deep Energy Retrofit
1 Year Report


The material below is Gordon's report on how things have been going in their energy retrofit of a schoolhouse into a very energy efficient home after about 1 year.   Gordon gives a summary of the thermal performance to date, as well as some fixes to minor problems with the retrofit.


While Gordon is inclined to be modest about it, it seems to me that the house is performing very well indeed, and that Gordon's spreadsheet thermal analysis proved to be very close -- no small accomplishment.


Note that in addition to the 1 year report below, Gordon has provided an updated copy of the thermal analysis spreadsheet...


I'd like to thank Gordon and Sue once again for the great energy retrofit and for the excellent documentation!


For the full story on the original project...


From Gordon:


Schoolhouse – Deep Energy Retrofit -- 1 Year Report 

Several months have passed since the original article was published.  Perhaps it is time that we provided an update of the thermal and energy performance for you and your website-readers.  The following text summarizes both our performance and various ‘fixes’.  The spreadsheet has been modified to include summer 2009 data, and extended for a second year’s data (includes fall 2009 at this time).  I’ve tried to clarify a few labels, but otherwise it’s pretty much unchanged.   

While the spreadsheet displays data in a September to August format, we received occupancy late in November 2008.  Prior to that, varying numbers of construction workers lived on site – both in the dwelling and in on-site trailers.  As a result, the fall data (particularly electricity) was not representative.  In the text below the ‘first year’ is identified as 1 Dec 2008 to 30 Nov 2009.   


Before describing the details of the energy and thermal performance, perhaps we should reiterate the major goals of our renovation.  Being retirees, we wanted a comfortable and casual dwelling, with low and predictable maintenance.  We wanted to eliminate stairways - but we felt that the lowering of our overall operating costs was paramount.  Beyond that, we wanted to be less vulnerable to the seemingly out-of-control escalation of most operating costs.  This led to thoughts of simplicity, flexibility and making use of available solar.  The recent extreme variability in the cost of fuels, plus the implications of extended power outages such as the ’98 Ice Storm suggested to us that we should not be overly dependant on any specific fuel – even including solar.   

Thus we never had the goal of designing and constructing a solar home.  Nor did we intend to create an ultra low energy home.  Nor specifically a ‘Green’ home.  These other goals would probably have led to a higher solar fraction, near net-zero, or healthier environment.  We were striving for a practical retirement home, something appropriate to our needs.  We are indeed pleased with the result – and the solar fraction, the energy consumption, and thermal performance aren’t all that bad.   

- First Summer’s Performance (2009)

In the original article we had mentioned that a requirement for summer-time air-conditioning was not expected, and thus no provision was being made.  We probably should have mentioned that for most of southern Ontario, while it is indeed a ‘cold climate’ for building design purposes, some provision for summer-time air-conditioning of dwellings is very common.  Almost all commercial buildings have air-conditioning.  While winters are long, summers tend to be hot and humid.  In many areas, we tend to joke that there is barely three weeks between furnace shutdown and air-conditioning turn-on (and vice-versa).   

It has proved correct that air-conditioning will not be needed.  In May, we shut down the HRV and began opening windows.  Over the winter, the temperature of the outside masonry walls had dropped about ten degrees F.  The warming of these walls over the summer actually mitigated interior temperatures – peaking at about 75 degrees F for less than a week.  I was afraid that there might be some condensation on the mass – but none was evident (condensation had definitely been an issue with the un-insulated slab floor of the original dwelling).  The first half of the summer was cooler and wetter than normal, and the exterior mass walls were still quite cool.  By August, I was worried that the walls wouldn’t warm up enough, and contemplated engaging the solar air heater in the solarium to heat the main house -but August and September were warm and quite sunny and the walls warmed right up – false alarm!   

I should mention that throughout the summer, we made no effort to manage ventilation and temperature of the main house with the windows.  We tended to have three small (cross-ventilating) windows open from 1 May to 1 Oct and perhaps one or two more for a few weeks mid-summer.  Half the operable windows in the main house have never been open.   

Regarding the solarium, we did have the sliding patio doors partially open much of the summer – but it never got un-comfortably warm.  The trick really seems to be having two or more open doors.  In our experience, there is far more ventilation with two doors each partially open perhaps three inches, than one wide open.  We never did get out the ladder to open any of the upper clerestory windows.   

- First Twelve Months’ Energy and Thermal Performance (1 Dec. 2008 to 30 Nov. 2009)  

With the construction winding down in Nov.2008, and our occupancy commencing, the hydro (electrical utility) meter finally had a chance to come back to earth.  A full year of occupancy has elapsed, and we can now report a little more accurately on both our total energy usage and our space heating performance.   

Solar Recoveries:

Developing an accurate prediction of the expected solar recovery was one of the most difficult aspects of the design.  There are several sources of aggregate data for this area, but little correlation, neither in aggregate nor monthly distribution.  The equivalent of U.S. TMY seems to be lacking.  So, while we predicted 36.1 M Btu potential solar recovery and 29.5 M Btu usable, we really didn’t know whether it was reasonable or not – and certainly had little idea as to the year to year variance (and still don’tJ).  By our admittedly crude measurements, we figure we recovered a potential of about 37.5 M Btu (30.6 M Btu usable) over the first year (about 4 % more than we predicted) – but . in all honesty, this is probably a pure coincidence or luck.   

Our spreadsheet predicted the solar fraction to be about 49.3% of the total net space heating requirement.  We estimate the first year solar fraction to have been about 50%.  The actual calculation is somewhat dependant upon (i) the actual total heat load – which we expect to be 1 –2% greater than predicted due to the cold winter, and (ii) the internal and HRV recoveries which were likely lower than predicted due to slightly reduced occupancy.   

Propane Usage:

On 30 Oct. 2009 our propane supplier refilled our tank.  Thus our total propane consumption for both the kitchen stove and living-room fireplace was 62.1 litres ( $ 40.30Can ) since the previous fill of 12 Dec. 2008 (10 1/2 months).  As a result, it has now been possible to refine our previous spreadsheet estimates for propane consumption.  For that period, a better estimate would appear to be 24.8 litres for space heating and 37.3 litres for cooking.  For cooking, propane use was only about 0.95 litres per week – but we spent a significant amount of time on the road slowly moving our possessions.  As a result, for the coming winter, we will estimate cooking consumption to be about 1.20 litres propane per week.   

Purchased Space Heating:

Over this one year period, total space heating consisted of 2784 lbs of wood (or 0.8 bush cord), 25.1 litres of propane, and 36.3 kWh of electricity for a total gross site consumption of 18.17 M Btu.  The resulting effective net space heating requirement was 13.13 M Btu (or 3848 kWh).  With the dwelling being 2506 sq ft (interior net) and the winter 7152 DD F., the first year net space heating requirement was 0.73 Btu per sq ft per DD64F (or 4.28 W per m2a per DD17.8C).   

In recent years, (and for good reasons) the work of the PassivHaus Institute is gaining acceptance.  In all honesty though, our designs were undertaken, committed and contractors engaged even before we had seen any PassivHaus literature.  We just wanted a comfy house – and the renovation was darn near complete before we realized their performance goals weren’t all that much different from our own.   

The above first year space heating usage translates to 16.99 kWh / m2aTFA – quite good, but certainly not meeting the PassiveHaus Inst. criteria of 15 kWh / m2a.  While the actual space heating consumption was 13.3% higher than the PassivHaus criteria, the slightly colder than normal weather last winter could be expected to account for about 5%.   

It is noted that the average first year space heating performance of the PH CEPHEUS Hanover/Kronsberg project was 14.9 kWh/m2a.  Nearly half of the 32 townhouse units didn’t meet the criteria either – some were double the consumption!  However, over the first three years, the average consumption dropping from 14.9 to 12.8 kWh/m2a  - presumably due to resolving construction difficulties and possibly weather warming trends of that time and area.  If over time, our space heating performance were to show a similar trend, our usage may well meet the PassivHaus criteria.   

The PassivHaus criteria was originally developed for Germany, and is gaining advocates, if not acceptance, in many parts of the world.  Note that this criteria does not take into account the severity of climate (ie: heating/cooling degree days).  As it is unreasonable to consider that all people of earth have an equal opportunity to reside in the optimally temperate portions of the world, there is some support for a performance criteria that gives consideration to the degree-day principle.  If the relative severity of our winter was recognized, our retrofit would have significantly better performance than their standard. 


Total Site Energy Consumption:

Total site energy consumption (excluding passive solar and recoveries) was 16,100 kWh for the year.   

Total Source ( or Primary) Energy Consumption:

Total source energy consumption for the first year was 18,800 kWh.   The low level of source energy is primarily due to the efficient use of low footprint, sustainably harvested, wood heat.  This translates to a total source energy use of 83.2 kWh per m2a TFA, or about 70% of the PassivHaus limit of 120 kWh / m2a TFA criteria. 


Further Modifications:   

Repair of Insulation Deficiencies: 

As discussed in the original article, there had been two small areas in the attic that had no insulation for last winter.  The area of steel decking exposed to the cold was only about 15 sq ft, but the thermal loss was rather significant - as the cold was measurable 8 to 10 feet beyond the exposed steel.  This has now been rectified.  One hole has been filled with XPS foam and sealed with “Great Stuff”. 

 The other hole was around the chimney of the masonry heater and required a little more thought due to (i) Fire Code clearance, (ii) material fire rating requirements, (iii) the need for expansion/contraction provisions between chimney and ceiling, and (iv) minor moisture issues arising from wind driven rain and condensation.  In the end, we filled the area above the masonry chimney and around the SS manufactured chimney to a depth of 6” with a fireproof mixture of vermiculite, clay and water (about 10 to 2 to 1 by volume).  Clay was chosen because it acts as a binder for the vermiculite and can withstand some moisture repetitively.  When thoroughly dry, several mineral fibre batts (“Roxul”) were laid over the area.   

Replacement of Exterior Shrouds (HRV):

In this area, high winds are rather common during severe weather.  The original HRV exhaust shroud had no damper, and in strong winds, the flexible duct was measurably cold even through the duct insulation.  The plastic shroud has been replaced by a more robust aluminum shroud that has a relatively tight fitting exhaust only flapper.   

We were not impressed with the HRV’s inlet filter.  It caught insects, but little of the dust from nearby farming activities.  We wanted an inlet filter with four levels of filtration – course, medium fine and electrostatic– but we also wanted them washable for extended life.  The original filter in the HRV is only about 9” by 9” square.  The air resistance of the new filter stack would clearly degrade the blower performance in the HRV.  We decided to remove the HRV filter, and replace the original exterior air inlet shroud with a fabricated wooden box using spare siding materials.  In this manner, we could build a multi-stage filter about seven times larger than the original, and clean-up the air outside - before it enters the ducting.  At some point, we should recheck the balance of the HRV duct flows, but it seems to work as well, if not better than before.   

Additional Dampers for the Kitchen Stove Exhaust:

In advance of the start of last winter, we selected a spot on one wall central to the house as a representative location to consistently measure interior house temperature for data logging purposes.  Yep, Murphy was an optimist!  As winter commenced, we began to notice that the temperature readings were somewhat cool – but we were comfortable!  As winter progressed, this became considerably more so.  Sure enough, a relatively large area around the kitchen stove was noticeably cool – right in the middle of the house!  This area included the bath tub/shower stall in the main bathroomL.  The range hood had no integral damper, but at the roof level, the outer shroud contained a gravity type lid damper.  We soon discovered that with any type of breeze, the damper would oscillate.  During more severe weather, it just remained open for extended periods!   

Come spring, we examined the situation in more detail.  Under certain wind situations, rain could penetrate into the shroud.  Moisture would then drain down the vent stack - collecting on top of the range hood L.  Under certain quick cooling situations (early evening after sun-down), condensation could arise on the outside of the duct (inside the attic), and sometimes on the underside of the panel steel roof – which could then run down the outside of the stack.  The result was slight puddling on the floor of the attic, which is sprayed in-place foam.  Some of this water then drained down the seam in the duct dripping onto the top of an interior concrete block wall.  Over time, moisture had spread through a considerable area of this wall complexL.   

In a nutshell, we needed a good damper for the air, but we had moisture running down both the outside and inside of the stack- moisture that would be prone to freezing that would disrupt a damper!   

The solution took a bit of tinkering:   

- Note that all joints were silicon sealed for water proofing and pop-riveted for strength. 

- The original vent stack through the attic was light-weight 6” diameter aluminum. 

- A scrap of aluminum vent  pipe was cut to a 10” by 14” rectangle, and then rolled into a tube about 4” in diameter, 10” long with a loose, but lapped side seam. 

- A common 4” dryer damper was affixed onto the top end of this 10” piece of four inch vent pipe

(ie: oriented for air to exit tube at damper). 

- A 6” to 4” reducer was installed over the tube from the lower (other) end, such that the damper was located inside the taper, near the 6” end.  Several small holes (perhaps 3/16”) were drilled through the reducer in the taper near the 4” diameter (lower) end.  Held with the damper upwards, one can see that upward flowing air would open the damper, and additionally, downward running interior water would be caught by the tube, yet the small holes would allow the water to drain to the outside.

- A second 6” to 4” reducer was installed beneath the first one, oriented in the same direction, almost touching one another, and both fastened to the tube.  .  In this manner, water running down the outside of the stack (ie: the first reducer) would be caught by this second reducer.  Prior to installation, a 1” plumbing fitting was fastened into a hole in this second reducer near the base of the taper.  Thus all water, (both running inside and outside the stack) would drain out to this fitting. 

- A third 6” to 4” reducer was installed near the bottom of the 10” tube, oriented in the opposite direction.  

- A second 4” dryer damper was installed on the lower end of the 10” tube, inside the taper of this third reducer.  The damper was oriented the same as the first, such that air entering this lower damper would exit the first damper. 

- The length of this assembly was measured and a slightly shorter section of the original vent stack was removed. 

- The assembly was mounted into the vent stack. 

- Chunks of scrap “Sonotube” were cut as forms and the vent stack and assembly was slowly (~8” every other day) insulated with “Great Stuff” from the attic floor, up to the water drain outlet. 

- A tube for the water drain was then installed. 

- The circular gap that collects the exterior water was masked with tape, and the foam insulation was extended up to this gap. 

Thus we now have two dampers within the vent stack and one at the roof level.  The dampers now only very briefly open in a very major wind gust – but readily open with the range hood on low volume.  The moisture problem has been resolved, and the house is now noticeably tighter.  Hopefully warmer – we’ll see. 

It is cautioned that this type of procedure is not likely in conformance to most codes, as it reduces the kitchen vent stack diameter – but commercial products simply do not address the situation.   

Damper for the Main Plumbing Stack: 

Even though the ensuite bathroom has a small window and an exterior wall (no plumbing in that exterior wall), the thermal analysis indicated that due to the mass and level of insulation, no special provision for heating would be needed.  However, the plumber chose to locate the main plumbing stack directly behind the back wall of the bath/shower stall.  Additionally, the contractors chose to not properly insulate either the hot/cold pressure lines nor the vents and main stack.L   

In hindsight, this is absolutely unacceptable for a low energy home.  One does not want condensation from the cold lines within the walls – insulate them!  Once one starts to consider every BTU, insulating, to minimize the hot water line loss, becomes more practical.  However, an un-insulated stack and associated vents might as well be considered a 4” hole in the wall right into your shower!  In some jurisdictions, a minimal quality of insulation for about 10 feet is suggested but not enforced.  For a low energy home in our climate (7150DD F), we would suggest a quality level of insulation for double that distance!  The implications of not doing so are simply dam uncomfortable!   

Correcting the situation properly afterwards is often impossible.  In our case, it’s not just the messy matter of removing drywall from the adjacent room, but additionally tearing down a double blocked load bearing concrete masonry wall too!  A proper repair is just not an option.L  A damper on the main plumbing stack might lower the thermal loss by a third – but is unlikely to be acceptable to code.  But when faced with the bathroom from hell, a damper starts to look pretty good!   

However, an open rooftop stack is not just a matter of an oscillating cold air column.  It also involves the entry of a fair amount of snow and cold rain – making thermal considerations worse.  It also means, this moisture must be significantly reduced – as it will wholly compromise the action of most any damper in freezing conditions.  Even if this externally sourced moisture is eliminated, the exiting sewer gas is very high in humidity, and could compromise the operation of a damper.   

We have tested several different concepts - all seem far from ideal.  The approach used for the kitchen stack is not feasible in this situation.  While it captures internal moisture, it would allow sewer gas entry into the attic – just not an option!  In the end, we connected two 90 degree vent elbows together to create a “U”.  We mounted a dryer damper between the original stack top and this inverted “U” – turning the assembly downwind from the prevailing wind direction.  In this manner, driven water is unlikely to compromise the damper, but internal condensation, when frozen, may prove to be a problem.  To minimize this, the exterior of the stack from the floor of the attic to the roof is being insulated, in an attempt to keep it relatively warm.  But this is not likely to be fully effective.  As this is a trial, it will be necessary to climb to the roof on several occasions this winter to examine the possible frosting problem.  In the meantime, while the bathroom is not comfy, it is marginally warmer.   


The Bottom line, when building or renovating, a quality level of insulation over the main stack and nearby vent pipes is critically important for low energy housing in cold climates.  Hindsight is 20-20.   


Best wishes





Gary December 15, 2009