Design - Maximizing Passive Solar Gain - An Overview

A recent re-read of Mazria's book, The Passive Solar Energy Book, made me think that it might be time to pause and revisit the subject of passive solar energy for space conditioning.  Our house is designed around harvesting passive solar energy year-round that will eventually provide all of the heating and cooling we will need.  In winter, the energy will come through lots of south-facing windows, which is nothing more than typical for passive solar.  In summer, it will come from the Annualized GeoSolar system (AGS), which is not typical. (For info on AGS, click on "Featured Post" in the left column.) Harvesting heat from the summer sun instead of depending solely upon the whims of the winter sun, makes AGS a perfect adjunct to classical passive solar design -- or the other way around since we consider AGS as our primary energy source. However, I realize that only a handful of dwellings utilize AGS.  In fact, Wikipedia's description of AGS references only two sites in North America one of which is our project.

So, for all practical purposes, passive solar for conditioning means using the winter sun for heating.

But maximizing solar gain, whether from the winter sun or the summer sun, is more than solar collectors and windows. In order for passive solar to work, or at least work efficiently and economically, the building has to be designed for retaining and distributing solar energy.  (Hence the advantage of designing passive solar into new builds having sufficient thermal mass, air-sealing and insulation, as opposed to expecting a decent pay-off from passive solar for leaky, under-insulated existing structures with limited thermal mass.)  

A Little History
It is clear from the literature that passive solar has had its ups and downs.  It was boosted by the energy crisis in the mid-70's, especially it seems with respect to earth sheltering and greenhouses. (There are at least eight earth sheltered houses within a 30 mile radius of Collinsville, all but one built after the early 70's.)  In the '80s, energy prices came down long enough and low enough to dampen enthusiasm for passive solar. Then, after the turn of the century, the talk of "peak oil" and the reality of global warming combined with rising oil prices, emerging alternative energy and mainstreaming of energy certifications, such as Energy Star and Leed, heightened awareness for energy conservation.   But it does seem that, as building green is brought to scale, the emphasis is on reducing fossil fuel costs and preserving of finite resources.  Passive solar as the primary means for space conditioning seems to have been relegated to parts of chapters in the newer books on green building if mentioned at all.  However, judging from an occasional new book on, and the amount of online chatter about, natural green building, there are still quite a few purists building houses with straw bales, adobe, cob, rammed earth and rammed earth tires. And earth sheltering is still often combined with the natural modalities, e.g., Earthships that are earth homes built with rammed earth tires.

Maximizing Passive Solar
So maybe its time to revisit passive solar using Mazria's book and our project as bases for discussion.  Any building that is solely or largely dependent on solar for space heating and cooling must first maximize solar gain.  Then, because BTUs from solar are harder to come by than BTUs from carbon, it must be even more efficient than most green buildings at keeping BTUs where they can do the most good -- in during the winter and out during the summer.  

A way to look at passive solar is that it involves three phases -- harvesting, storing and distributing.  My blogs posts to date have co-mingled the three phases to the extent that maximizing solar gain per se has been lost in the discussion.  This series of four posts focuses just on maximizing passive solar gain. 


Suggested Reading
While I am using Mazria's book as a handy blueprint, a lot of information comes from a variety of other sources, principally the internet and the books I own or borrowed from the public library over the years. And I can recommend the reference list in Mazria as an excellent resource for hardcore passive solar literature. His book was published in 1979 and most of the references were published in the '60s and '70s with most in the '70s.

Earth sheltering and passive solar are complimentary modalities but one only has to peruse downloadable pictures of green buildings to realize that  passive solar without earth sheltering is the new norm. Nevertheless, while I am suggesting references, let me mention three go-to sources for information on earth sheltering: (1) the content and annotations in Don Stephen's paper on Annualized GeoSolar, (2) the well-annotated book produced by the University of Minnesota Underground Space Center titled Earth Sheltered Housing Design and (3) John Hiat's self-published book, Passive Annual Heat Storage, Improving the Design of Earth Shelters, that is invaluable for both its practical content and short bibliography. Unfortunately, both books are out of print but used ones are available online.  All three eclipse the methods that were popular in the '70s.  For a look back at those modalities, the classic is Rob Roy's Earth-Sheltered Houses, How to Build An Affordable Underground House.

Types of Solar Heating

There are three approaches to solar power for heating. The first and most popular is direct gain whereby the living space becomes solar collector, heat storage and distribution center all in one.  It is characterized by south-facing glass (northern hemisphere) and enough thermal mass properly positioned for absorbing, storing and radiating solar energy. 

The second approach to passive solar heating is indirect gain whereby thermal mass is positioned between the sunlight and the living space.  The mass absorbs solar energy and converts it into thermal energy for heating. The Trombe Wall, the

most famous iteration of indirect gain, is a masonry or water wall set just inside the south-facing windows. 

The third approach is known as isolated gain, so called because the collector and thermal storage system are isolated from the living space.  A classic example of this approach is a flat plate collector, whereby sunshine passes through glass (that is not window glass for the living area) and heats a thermal mass.  The heat is transferred via natural convection from the mass into the living space.  

Isolated gain

Of course, passive solar means that, regardless of approach, there are no mechanical assists like fans, blowers or pumps.

Annualized GeoSolar does not fit nicely into any of the three types of passive solar heating.  It has attributes of both direct gain and isolated gain depending on the season. For summer, there is flat plate collector but it is located 20' in front of the house (isolated gain) and the distribution system and thermal mass are located under or adjacent to the living space (direct gain). For winter, it is all direct gain -- south windows and interior thermal mass.


The Merits of Passive Solar Conditioning
The advantages of passive solar are........
     -  Simplicity:  design, operation
        and maintenance 
     -  Feeling of comfort at lower room
        temperatures 
     -  Warmer floor
     -  Endless supply of free energy
The biggest disadvantage:  lack of control

Design Considerations for Passive Solar
Following is a list of the design considerations that receive the most attention by the references that I consult.  This is the first of four posts on passive solar.  Three forthcoming posts will enlarge on selected items from the list and touch on ancillary topics.

     Building
          -  Located for maximum exposure to winter sun
          -  Shape
          -  Orientation
          -  North side considerations
      Room arrangement
          -  Active living spaces
          -  Spaces requiring less heat and light
      Protected entry
      Windows
          -  Location 
          -  Ratio of glass to floor area
          -  Transparent vs. translucent
          -  Exterior shading from summer sun
          -  Interior thermal shades for nighttime and gray days
          -  Operable windows for warm weather cooling
       Heat storage
          -  Masonry
          -  Adobe
          -  Water walls
          -  Size and location relative to windows
      Surface colors
          -  Thermal mass
          -  Wood frame walls and ceilings
          -  Exterior surfaces
        Supplemental heat source

The next post on passive solar will focus on the first three major items in the list -- building, room arrangement and protected entry.

Construction - Short Truss Wall (Cont'd)

The first post on the short truss wall zeroed in on the unusual handling of the mudsill. This post is about building the rest of the wall then raising it onto the mudsill for alignment and nailing.  (Note: click on any photo to enlarge it for more detail.)

Pre-made Trusses

The jig as modified for the short trusses; notice the
pre-cut components in the background and the long
radial arm saw table for gang-cutting them

The trusses that will support the roof will extend from the south wall of the second story to the short north wall under discussion here.  The fact that I elected to erect a internal bearing wall in between before the back wall was in place meant that a line from the front wall to the bearing wall had to be continued on to the future back wall in order to know the height of the back wall. Since the tape measure would sag too much over such a span, it was necessary to scab together a couple of  2-bys on which to lay the tape measure.  I took the measurement in three places -- in the middle and at both ends.  Fortunately, the results did not vary more than half of an inch -- 43" plus or minus 1/4".

A truss prior to removal from the jig; gussets have
been nailed on

An earlier post detailed the use of a jig for pre-making trusses for the exterior walls.  I used the same jig for the short wall trusses by installing a divider that restricted the working part of of the jig to 43". I have an excellent job-site sliding compound miter saw but the handiest tool for gang-cutting truss components is the radial arm saw in the shop with its long and wide table. In short order, the side rails and short pieces that form the ends for 31 trusses were cut from salvaged 2 x 4s followed by the OSB gussets from new material.

Then I used the jig and two Pasload nailers to assemble the trusses.  Without thinking, all four gussets were nailed to place only to find out later that one on each truss had to be removed in order to have access for nailing the trusses to the mudsill after the wall was raised.

Building the Wall

Trusses made from recycled lumber except for new OSB gussets

The 2 x 6 tandem top plates were laid side-by-side for laying out the wall.  After the lay-out, one set was set aside and the trusses arranged at a 90 degree angle opposite the marks on the other set. As described in an earlier post, I stood the top plate on edge and nailed the trusses to it flush with its bottom edge. Then I nailed a parallel set of top plates to the trusses using spacers and shims on top of the first plates to position them for nailing. Overall, the wall was built in three +/- 20' sections that a friend and I raised in sequence starting from the east end.

Since the mudsills had already been installed, they were not available for nailing to the

The short wall upon completion

bottoms of the trusses before raising the wall.  So, in order to stabilize the bottoms for the raising, I attached a 1-by temporary brace.  The scaffold railings interfered with raising the two end sections directly into place.  Instead these sections had to be raised off-position and slid along the mudsill into the correct position.  As a precautionary measure, I nailed a second horizontal brace to the other side of the trusses before attempting the slide. Once the wall

The last section ready to raise; notice
the 1-bys bracing the bottoms of the
trusses for raising

was in position, the temporary braces were removed and the individual trusses were aligned flush with the edge of the mudsill, plumbed in an east-west direction and nailed to the mudsill with only one nail close to the exterior side.  The interior side would be nailed after the wall was plumbed in a north-south direction and braced. The decision to nail the outside first instead of the inside was dictated the tendency for the wall to lean slightly inwardly.

The plastic sheeting that was stapled to the mudsill on the day it was installed (first post on the short truss wall) was left in place under the trusses.  Eventually, after the pressure treated wood has dried sufficiently and the roof shades the sill from the sun, the plastic can be cut away without worry about the sill warping due to uneven drying in the heat of the sun.

The concrete contractor placed the anchor bolts in the middle of the 10" wall so they ended up only a couple of inches from the inside edge of the 2 x 12 mudsill when it was cantilevered 4" outward in order to be flush with the stucco.  I added an equal number of anchor bolts an inch or so from the outside edge of the concrete.  The extra anchors at least fell in the

The completed wall from the inside except for replacing
the gussets at the bottom that were removed in order
to have access for nailing the trusses to the mudsill; to
have built the wall without the scaffolding would have
difficult

middle of the 2 x 12 and moored the outside half of the sill before the wall was plumbed in an north-south direction.  In conjunction with plumbing, both sets of bolts were loosened or tightened as needed.

So much anchorage may seem like overkill but our location carries three types of risk: tornadoes, earthquakes and subsidence. We have tornado alerts every year, sometime several times, and actual tornadoes nearly every year. Seismologists say that the odds are pretty high for another major earthquake at the New Madrid fault near the Mississippi River in southeastern Missouri.  If one should happen, the seismic waves will follow the gelatinous river floodplain to St Louis with the potential for major damage. Finally, subsidence from cave-ins of abandoned underground coal mines occur regularly in southern Illinois, including in Collinsville where a reasonably new school had to be razed a few years back due to subsidence damage.  The old mine under our site is a little over 200 feet down and, even at that depth, we have to worry about subsidence.

Just like the other 15" exterior walls, the short wall will be insulated with rice hulls to an R-48. The 5" space between the tandem top plates will provide access for blowing the hulls into the wall cavity after the sheathing and drywall are in place.

The first set of top plates were nailed to the wall before raising.  Consistent with common practice, a second set of top plates were necessary to bridge the joints in the first set and establish continuity and alignment. The roof trusses will rest only on the outside-most "top" top plate. Because of the pitch of the roof, there will be space between the inside "top" top plate and the trusses. Consequently, I used less-than-perfect salvaged 2 x 6s for the inside "top" top plate but bought new 20 foot long 2 x 6s for the all-important outside "top" top plate to which the trusses will be fastened.

The jig for pulling the measurement for the roof trusses

Measuring for the Roof Trusses 
Working alone, it would have been impossible to pull an exact measurement for the roof trusses without some sort of jig. One of the new twenty-foot 2 x 6s for a top plate was perfect for making a jig.  I scabbed an extension to it and used it to span the distance between the front and the back walls. Beforehand though, I put a shallow saw kerf in one edge and tacked opposing keepers to the sides of the board in several places.  After hoisting the 2 x 6 to the top of the walls and standing it on edge near the west end, I used a level to make the saw kerf even with the outside edge of the front wall framing and clamped the 2 x 6 to the middle wall to steady it on edge.

Then, to pull the measurement for the trusses, I hooked the end of the tape measure in the

The vertical lines on the 2 x 6 flush with the wall framing; notice that
the variance between locations is only 1/4"  (to see the lines better,
click on the photo for blow-up

kerf and stretched it along the top of the 2 x 6. The task was simplified by having the keepers tacked to the sides of the board that kept the tape measure from sliding off of the edge.  I used a level to mark a vertical line on the 2 x 6 that was flush with the outside framing of the short wall and recorded the length.  I repeated the measurement at the middle of the wall and near the east end.  Instead of using the tape measure each time, I could compare plumb lines with the original mark on the board.  To my surprise, the variance between the three locations was only 1/4".

I might say parenthetically that the technique for measuring just described is only one example of many techniques that my working-alone mindset comes up with after having read early on John Carroll's book, Working Alone.  I recommend it for any serious DIYer.   The triangular braces in the nearby photo that are clamped to the mudsill while I was using it as a straight edge to assess the levelness of the concrete wall (previous post) was suggested by Carroll.  The four that I made exactly to his prescription have been enormously helpful in many ways.  (Ever tried planing a door while holding it between your legs?  Try using Carroll's braces.)

I needed two other measurements for the trusses. One was the difference in height between the front and back walls, the value of which gives the roof pitch.  This task was made easy by the rotary laser. The other measurement was the horizontal distance between the walls which I obtained in one area only.  Since the situation is a right triangle, the difference in height and the distance along the slope would be sufficient to calculate the horizontal distance between the walls .  But as a DIYer, I was more comfortable pulling all three measurements.  And for added comfort, I asked my mathematician brother-in-law to make the calculation and found that his calculation coincided with the measured distance.

Construction - Short Truss Wall

Since the concrete earth contact north wall is only 12' high instead of a full two stories, the difference must be made up by a short stick-built truss wall.* This post, the first of two on the short wall, focuses on solving the problem of the concrete wall below it being unacceptably out of level.  It describes the installation of the mudsill first, building the rest of the wall horizontally then standing it on the mudsill for nailing instead of attaching the mudsill, as is typical, before raising the wall.

After cutting the mud-sills to length, they were used as straight-
edges in the middle of the wall for determining the height and
spacing of the shims that would be necessary to keep the sills
 level when laid flat

The Unlevel Conundrum 
The rotary laser and a taut mason's line told me that the concrete wall was too low in several areas along the 58' span -- as much as 3/4" in one area. If the truss wall were to be built with standardized trusses already fastened to a mudsill, shims of various thicknesses would have to be used under the sill to keep it level.

There would be two disadvantages to this approach.  First, there would be a space under the sill that would have to be closed in some fashion to seal out bugs and critters and to eliminate air infiltration/exfiltration.  When I helped my step-son, Keith, build his house a few years ago, I saw how difficult, frustrating and shotgun-ish it was to try to force mortar into the gap under the mudsill after the wall

The shims were screwed down; the sill will
not sit directly over the concrete -- its outside
edge will be flush with the stucco and barely
extend past the bolts on the inside; additional
anchors will be added closer to the outside
edge of the concrete

was in place.  On the exterior, the termite shield was in the way of injecting the mortar so it had to be done from the interior with no guarantee that the space was thoroughly filled. Incomplete filling would be even more likely with my 2 x 12 mudsills than with Keith's 2 x 8 sills.  

The second disadvantage of filling the gap after the wall was raised would be that part of the wall and the roof above it would be supported by shims with whatever additional support a hit-and-miss mortar bed could offer.  A decent mortar bed would offer more support and having an uninterrupted mortar bed is even more of an issue our case. The outside edge of the 2 x 12 aligns flush with the stucco, which means it is cantilevered +/-4" over the insulation, cement board and stucco.  
Ideally, the anchor bolts should be located near the center of the sill.  In our case, with the inside edge of the sill being only an inch or two beyond the anchor bolts that the contractor installed in the middle of the 10" wall, the outside edge of the sill would not be sufficiently anchored.  Consequently, I will be adding many more anchors near the outside edge of the concrete, such that they are closer to the middle of the sill, after the wall has been raised and nailed to the sill.

Builders nowadays typically use a sill gasket between the mudsill and the concrete wall (actually between the sill and the termite shield) as an air seal and a moisture break.  With our 3/4" gaps, the typical 1/4" thick gasket would not suffice as either one. The DIY method detailed here provides air sealing but not a moisture break. However, the two layers of 6 mil plastic that sandwiches the insulation will keep the concrete bone dry then the house wrap and metal cladding will overhang the top third or so of the termite shield.  The contact between them will not be conducive to capillary attraction that is necessary for moisture wicking.

Holes for the bolts were drilled and
the sills were dry-fitted

The Plan 
After my experience with Keith's house and with the installation of the first truss wall, it made sense to install the mud sill first, get it level and well supported by mortar then stand the rest of the truss wall on top of it. Accordingly, I would use shims to keep the sill level while bedding and bolting it to the concrete wall. Then I would build the wall on the floor with trusses attached only to the top sill, not the mud sill, and stabilize the bottom with a temporary brace.  After the wall was raised, the individual trusses could be nailed to the mud sill.

Installing the Mud Sill 
For the rest of the exterior truss walls, I will be using 2 x 6s for tandem mudsills over the 13" wide insulated concrete walls.  When pouring the concrete walls, I intentionally placed the anchor bolts off-center in the concrete and protruding an extra 1 1/2" from the concrete so as to be situated midway between the mudsills and tall enough to receive a 2-by bridging across the sills.  However, the contractor for the north wall placed the bolts in the middle of the wall, making it awkward to use tandem mudsills. Consequently, I settled for a single sill using pressure-treated 2 x 12s.  Before attempting to

The near sill is still in the dry-fitting
position; the far sill has been
inverted and the termite shield
nailed to it

level the sills, I bored holes in them for the anchor bolts in such a way that their outside edges were flush with the surface of the stucco below them.  Four inches of the sill protruded out over the insulation and stucco which meant that less than 8" lay on the concrete.  With the bolts in the middle of the 10" concrete wall, the holes were located within a couple of inches of the inside edge of a sill which was an advantage for installing the sills but not an advantage for anchoring the wall long-term.

It would have been difficult if not impossible to have bedded the sills in mortar without shims to level them and support them at the correct height as the excess mortar was squeezed out.  The 2 x 12s were dead straight and 20' long so I used one of them as a straight edge leaning against the outside edge of the bolts to place shims about 8' apart and just outside the line of bolts. When the sill was settled into the mortar bed and bolted to place, the shims would dictate the correct placement of the inside edge of the sill while the outside edge could be beat with rubber mallets until it was level with the inside edge.  With the straight edge still resting on the shims, I used Tapcon screws to stabilize the shims so they would not move when laying down the mortar bed, reinstalling the sills and settling the sills into the mortar.

The near sill has already been bedded
level in mortar; the mortar is in place
for the far sill

Keeping the mudsills separate from the rest of the truss wall afforded the opportunity to attach termite shields before the sills were installed.  Instead of using roll flashing and bending it with a braker, I used drip-edge that is used for roofing.  It came already bent but the disadvantage was that, instead of one continuous piece, it took six 10' pieces to span the wall.

The sills were too heavy for one person to handle alone so step-son Keith came to help.  First, we dry-fitted the sills over the anchor bolts.  Then we used roofing nails to fasten the drip-edge to the underside and flush with the outer edge of each sill, overlapping adjacent drip-edges a few inches.  The overlaps will be pop-riveted eventually.

Sill covered by
plastic sheeting

After the shields were attached, we installed the sills one at a time so as to be sure it could be completed before the mortar began to set.  The mortar was tooled to place so that it stood proud the shims by at least 1/2". We dropped the sill over the bolts, added washers and nuts and began tightening while beating on the inside edge of the sill with rubber mallets to squeeze out excess mortar.  As soon as the inside edge was against the shims, one person stood on the outside edge of the sill while the other beat it with a mallet.  Eventually, the excess mortar exuded out and the sill was level crosswise as well as longitudinally.

Wet pressure treated lumber warps under direct sun due to uneven drying.  In order to preclude warping, the sills were stored under cover until they were installed.  The sun came out before the installation was complete so I kept the sills wet with a sprayer then covered them with plastic as soon as the installation was complete.  The object was to make them dry evenly, loosing moisture from all sides simultaneously.  

After the rest of the wall is raised and nailed to the mudsill and I know where the trusses will fall, I will install a concrete anchor between each of the original anchor bolts,  Where the drip edges overlap, I will join them with rivets and caulking.

View showing the termite shield; parging with stucco is
incomplete due to a cold spell before it was finished

A subsequent post, will detail the raising of the stick-built wall on top of the mudsill.

______________
*  The reason for not extending the concrete to a two-story height was a matter of cost and ease of insulation.  The truss wall is cheaper and easier to insulate to an R-50.

Design - Indoor Air Quality

According to one of the contributors to Christina Fisanick's book, "Eco-Architecture (Opposing Views on the Merits of Green Building)", the EPA says that outdoor air is 2 - 5 times healthier than the average indoor air.  Any house like ours that will be sealed well enough to be energy neutral would have an indoor air problem without purposeful control of pollutants.

Indoor air quality can be manged by a combination of two strategies:  controlling the amount of internal pollution in the first place followed by continuous replacement of stale air with fresh air.

Control Measures

• Combustion ventilation:  Vented range hoods; vented water heaters; heating stoves with piped in make-up air and sealed combustion chambers; no open fireplaces
• Moisture control:  Range hoods; bathroom exhaust fans; dryer vents; dehumidifiers if necessary
• Low or no volatile organic compounds (VOCs):  Formaldehyde-free building materials; low or no VOC paints, finishes, carpet and upholstery; outside storage for and use of VOC cleaning supplies and shop chemicals such as acetone and paint thinner
• Radon mitigation during construction or as a retrofit
• Air barriers between attached garage and living quarters; no duct work penetrations

Ventilation

• Enough strategically placed operable windows for adequate ventilation when the     outside environment cooperates
• Heat recovery ventilator (HRV) or energy recovery ventilator (ERV), either free-standing or tied into the central HVAC to replace 30 - 50% indoor air with outdoor air every hour 

How Does Our Design Stack Up?

We will have a range hood matched to the BTU output of the burners but I remain ambivalent as to its configuration.  Venting to the outside is best in a normal situation but the disadvantage of venting is that, in winter, a lot of heat is lost.  Since we will have an ERV, maybe it makes more sense to use a hood with high-end filtering to comb out the worst pollutants then duct the ERV close enough to the hood to remove the remainder.  I need to do more research and consultation on range hoods before we make choices.

We will have no other sources of open combustion such as stoves and fireplaces but will have a gas tankless water heater and gas clothes dryer properly vented to the outside.  Since we are building new, we expect to have good control over VOCs.  Radon gas will be intercepted and led to daylight by the gravel backfill in the French drains and the AGS conduits and we have used plastic sheeting under the slab.  Air barriers between the garage and living quarters with no un-caulked penetrations are not only required by code but make perfect sense and we will be in compliance.

We are deliberately limiting operable windows to the least number necessary for proper ventilation because fixed panes are cheaper and are less likely to leak air. Since there are a lot of days here in the St Louis area when it is too cold or too hot or too humid to open the windows, we will depend mostly on an ERV to exchange +/-30% of the indoor air each hour year-round.  (Heat Recovery Ventilators (HRV) are better suited for colder, drier climates; the ERV functions like a HRV but also helps to control incoming humidity.)

Energy Recover Ventilator

Ours will be a free-standing ERV since we will have no conventional heating or air conditioning with which to integrate it.  The installation will be so unsophisticated that it becomes a reasonably easy DIY operation with perhaps some professional help in balancing the system.  The ERV will ventilate the bathrooms, eliminating the need for separate bathroom fans and, as discussed above, will likely replace a vented range hood.

As a value-added feature, the ERV will be situated so that it pulls air through vents in the partition between the living quarters and the earth contact wall so as to bring more air into contact with the wall than would otherwise be possible, thereby enhancing the performance of the AGS system.  

ERV in summer mode.  Hot, humid incoming air is cooled
and dried by the outgoing air by passing near each other
in the heat exchanger without actually physically mixing

Construction - Insulating the Earth Contact North Wall

Intentional Compromise
In a perfect world, the earth contact north wall would have been a full two stories high, i.e., at least 16 feet.  But to have done so would have increased its cost so significantly that I decided to stop at 12' with the concrete and carry the wall to second story height with a short truss wall.  The latter could be done with salvaged lumber and my free labor as opposed to paying for professional labor and buying several more yards of concrete to increase the height, not just of the 12" thick wall itself, but the three substantial deadmen behind the wall as well.

Moreover, it will be easier with the truss wall to achieve an R-48+ to match the other external truss walls than to insulate a concrete wall to this R-value.  But the advantages of a higher wall of concrete would have been an additional 200 sq ft of earth contact for the AGS system and more of the wall totally impenetrable by air infiltration/exfiltration. However, I am betting that (a) the large floor area of the house, (b) the +/-10' of uninsulated north earth contact wall, (c) a partial earth contact west wall and (d) the insulation/watershed umbrella will provide all of the insulated and dry thermal mass the house will ever need without the additional 4' of concrete for the north wall.  Therefore, I think the additional 200 sq ft of earth contact would have been expensive insurance for a problem that probably does not exist.

Tweaking the Platon Damp-Proofing 
As described in a prior post, the Platon damp-proofing system was used on the lower 8' of

Original Platon installation

the tall section of the wall before backfilling began.  The top 4' still needed damp-proofing before it could be insulated.  I decided to keep the Platon system below the bottom of the metal channel that holds the insulation. Otherwise, the extra bulk would cause unevenness of the insulation because the top of the channel would be directly against the concrete and the bottom would hang over the bulky Platon material. Consequently, I installed only enough additional Platon to bridge the gap between the original material and the bottom of the channel.  As will become apparent presently, the two layers of plastic sheeting that waterproofs the metal channels and insulation will also waterproof the concrete even better than would the Platon material.  The exception to this arrangement was the first four feet at the ends of the wall where the insulation extended down nearly 6' behind what will eventually be retaining walls running more or less perpendicular to the wall.  Here the channels did overlay the Platon material and did flare out somewhat at the bottom.

Materials Preparation
First is was necessary to calculate how far the insulation should extend below the mudsill of the truss wall.  Code specifies a distance of 8" between the grade and the mudsill. Figuring downward from the mudsill, the first 8" will be exposed stucco, the topsoil over the

insulation/watershed umbrella will be 8" deep and the umbrella will be 6-7" thick, so the wall must be insulated to a depth of at least 23" below the mudsill to get below the umbrella. (Below that, of course, the wall should remain uninsulated so that heat can pass back and forth through the wall between the earth and the living space as a function of the AGS system.  For more on AGS, click on the "Featured Post" in the column at the left.)  I settled on 32" of insulation 3 1/2" thick for an R-14 to a depth of 8-9" below the bottom of the umbrella. The nearby photo shows these dimensions drawn on the cement board before stuccoing (click on the photo to enlarge it for better detail).

Parenthetically, in order to achieve an overall R-48, the inside of the wall will have to be insulated to another R-34, which is the subject of a future post.

I used a plywood blade in a circular saw to cut two thicknesses of expanded polystyrene foam board -- 2" and 1 1/2" -- into 24" x 32" pieces .  I also cut 
3 5/8" wide 20 gauge galvanized steel channel (the kind that is used as the top and bottom plates for steel studs for wall framing) into 24" lengths using a metal blade in a radial arm saw.  I cut 1/2" cement board into 48" x 28" pieces using a corded fibercement nibbler.  Next, I cut lengths of 6 mil plastic sheeting into 8' wide pieces.  At this point, all I needed in order to begin insulating the wall were 1 1/4" x 3/8" Tapcon screws for fastening the channels to the wall and 1 1/4" cement board screws with high-low threads for fastening the cement board to the outside of the channels.  I then followed the same installation procedures that were covered in a previous post with a few deviations that were so minor as not to warrant special discussion here.

Installing the Insulation
As described in the previous post, the channels were readied for installation in the shop by cutting to length, drilling holes for the Tapcon screws in half of the channels and then screwing channels with holes back-to-back with channels without holes.  I started the installation at the east (left) end of the 12' wall so that the channel with the holes would be facing to the right each time so as to make it easier to use the hammer drill in my right hand; if I were left-handed, I would have started at the other end of the wall.

The first four feet of the wall needed to be insulated to a depth of +/-6'.  The additional depth was necessitated by the transition from an 8' wall to a 12' wall which will be accomplished by a retaining wall running more or less perpendicular to the wall and butted up against the first four feet of insulation.  (The first book on earth sheltering that I read some 8-9 years ago was Rob Roy's Earth Sheltered Houses in which he cautioned against butting a retaining wall against the concrete house wall without insulation between the two in order to reduce heat loss and to eliminate moisture condensation during warm months.) The same 4 x 6' configuration was necessary at the west end of the wall for another retaining wall.  In between, there was a span of approximately 52' where the insulation was only 32" in height.

Before the channel and the insulation could be started, the 6 mill plastic had to be positioned so that roughly half of it would be trapped under the insulation and half lapped

Insulation installed on about half of the wall with the inner
layer of plastic showing beyond and below the foam;
 bare concrete is in the distance; notice that the metal
 channels are shorter than the insulation

over the insulation, completely sealing off the channel-insulation complex. The rolls of plastic I buy at the farm supply store comes 24' x 100'.  I divided the 24' dimension into thirds.  The 8' wide pieces provided 4' of plastic between the insulation and the concrete wall and 4' between the insulation and the cement board. And the individual pieces of plastic were overlapped by a couple of feet.  In this way, the metal channels were totally sealed against rust-inducing moisture, rather than relying solely on galvanization, and the concrete wall was not just damp-proofed but was actually waterproofed in the process.

Without going into all of the details covered in the previous post, the installation was a

Completed installation of the foam; notice the outer layer
of sheet plastic, thrown back and weighted down on the
floor of the scaffold, that will be brought forward and
 draped over the foam before the cement board goes on 

matter of screwing a pair of channels to the wall over the inner layer of plastic, mating a piece of 2" foam board with a piece 1 1/2" thick, slipping the left side of the foam into the channel, slipping the next pair of channels over the right side of the foam and applying pressure leftward while drilling the holes and inserting the Tapcon screws. The 32" length for the foam and 24"length for the metal meant that the foam protruded below the foam 7-8" in order to keep the metal well up inside the sheet plastic to keep it dry.  I also kept the tops of the channels an inch or so below the tops ot the foam in order to keep any rough edges from perforating the sheet plastic. The inside diameter of the channels is 3 5/8" and the foam is 3 1/2" thick so there is enough tolerance that the insulation can be moved somewhat.  Accordingly, after all of the insulation was in place, I used a mason line to level the tops of the foam to match the bottom of the mudsill that will eventually hang over them.  Any remaining gaps between the sill and the foam board will eventually be sealed with mortar as described in a subsequent post.

Installing the Cement Board
Installation of the cement board required two cordless drills and a driver for phillips head cement board screws.   Before taking a piece of cement board to the wall, I used one drill

Completed cement board installation; notice ample outer
 layer plastic sheeting exposed below the board; the
backfill will press it against the edge of the foam and
the wall below to seal the channel/foam assembly against
moisture

with a 3/16" masonry bit to drill a hole in the upper left corner and the upper right corner.  I As soon as the left side of the board was butted against the right side of the previous piece, aligned flush with the top of it and proved level with a torpedo level, I used the second drill with a bit designed for metal to drill a pilot hole in the metal track through the hole previously drilled in the board in the upper left corner. Then, while holding the board steady, I used the driver to fasten the board to the track with a cement board screw.  The remainder of the screws could then be installed in a similar manner, i.e., masonry bit, metal bit and cement board screw.

Installation of the cement board is best done by two people because it is heavy and difficult to manage working alone.  I overcame the problem by leaning a 2 x 4 against the wall at the level of the bottom of the board as a support.  However, the rest was rarely at the right height necessitating a lot of juggling which probable quadrupled the installation time.  It didn't help either that the ground was uneven and snow-covered part of the time and beginning to thaw the rest of the time.  In either case, the slipperiness underfoot made things interesting.

Stuccoing the Cement Board
This last step in insulating the concrete wall followed exactly the procedures described in detail in the previous post.  The only difference was it had to be done in late December instead of warm weather, which complicated and slowed the process considerably.  Not knowing when the stuccoing might get done, I am publishing this post without pictures of the stucco on the assumption that the curious will be checking out the previous post anyway.

Honest Perspective
The DIY concrete wall insulation described here and in the previous post is a good example of a green building approach that would not be practical in a production setting because it is too customized, cobbled together from disparate materials and time-consuming.  In a way, though, it is a microcosm of our entire house project.  Jo Scheer in the book, Eco Architecture says, "Though extreme eco-architecture may not be a solution to a thoroughly sustainable building industry, it certainly provides ideas. It is a model of ideas and concepts that beg to be assimilated".  Who knows what impact on sustainability some of our "impractical" ideas might have somewhere down the line.  

Construction - First Exterior Truss Wall

The actual "first" exterior wall was described in the previous post  but its ten-foot-plus height from the second story floor and a preponderance of windows caused it to be a hybrid instead of a pure truss wall.  The wall described here provides an opportunity to detail the building of an 8' wall using trusses made ahead of time in a jig. Rather than going over the details of truss wall construction when blogging on the rest of the exterior walls, I will link back to this post as a reference.

The wall is the exterior wall between the garage and the living quarters.  By not having windows like most of the other stick-built exterior walls, it allows us to focus on the use of trusses instead of 2 x 4s or 2 x 6s for framing.  The fact that it has a door only means that a couple of trusses are positioned differently than would be the case for a plain wall.  While the garage will be insulated and passively conditioned to a higher comfort level than most garages, the design for the wall is no different than for the rest of the exterior walls, i.e., 15" thick and an R-value of just under 50.

Although I intend to refer back to this post in the future for the details of truss wall construction, it is atypical in one respect.  The wall was put together on the floor of the garage and tipped to place from the exterior side.  The remaining truss walls will be assembled on the house floor and tipped up from the interior.  I will try to nuance the difference during this narrative.

Building the Wall
I took eleven trusses for the wall from the stash of pre-made trusses and built the wall on the floor as is typical of most wall construction.  I failed to photograph the wall while it was laying on the floor so I've included the following photo from the most recent post as an example of horizontal assemblage.  

After cutting the 2 x 6 pressure-treated bottom plates (mud sills) and the salvaged lumber 2 x 6 top plates to length, I laid them side-by-side and used a tape measure to mark lay-out lines on 24" centers as if laying out stud positions for traditional walls. The future pedestrian door to the garage was laid out in the process, centered over the mini-ramp that had been formed into the garage floor to make the doorway ADA compliant.  I stood the trusses on edge in the approximate positions they would occupy in the lay-out. I stood straight pressure-treated 2 x 6s on edge against the bottom of the trusses with the lay-out lines facing the trusses.  The trusses were moved to match the lay-out lines more closely and "eye-balled" perpendicular to the 2 x 6 longitudinally. Then, starting at one end, I aligned each truss accurately with its lay-out line on the mud sill and nailed it with one nail.  I then used a rafter square to make sure the truss was perpendicular to the sill before adding more nails.  

Digressing for a moment.  The rest of the 8' exterior walls will be framed on the floor of the house as opposed to the garage floor.  For them, getting the framing as flush as possible on the interior profile for smoother drywall would be more important than for the sheathing and cladding side of the wall. So, the following procedure applies specifically to the other walls but I used it for this wall while knowing that any minor discrepancies would be reversed and face inward and the garage wall would be the smoothest. A smooth interior profile is another reason for using the straightest 2 x 6s next to the floor which puts them on the interior side when the wall is raised.  

I slipped enough shim(s) under the 2 x 6 and the end of the truss for the truss and the 2 x 6 to be in simultaneous contact with the shim(s) before sending home the first nail  Since the trusses, having been assembled in a jig, are quite true and uniform, the shims merely compensate for any unevenness in the floor that might cause misalignment of the straight 2 x 6 with the trusses and result in a rougher interior profile.

The next step was to attach the interior-most (current situation; exterior-most for the rest of the walls) top plates in exactly the same manner.  (I use the plural form "plates" because the pressure treated 2 x 6s would have had to have measured 20' to have spanned the entire length of the wall.  It is hard to find straight pressure treated two-by-sixes this long so, I used two boards of varying lengths for each plate so that the junction between them on one side of the wall did not fall opposite the junction on the other side of the wall.  The top plates were salvage lumber so I used two of varying lengths such that the junction of one set of top plates were staggered not only with each other but with those of the bottom plates. 

Finally, it was a simple job to nail the second set of 2 x 6s to the trusses.  The trusses were already properly aligned as for perpendicularity and therefore already matched the layout lines on the top plates.  Although the 2 x 6s were salvaged lumber, the two used for the exterior-most top plate were pretty straight while the one of them used for the inner top plate was bowed slightly.  I made sure the bow was facing towards the center of the wall so that it would not hang over the top of the wall and interfere with proper mating to the floor joist as described below.  For the other walls, a plate that bowed outward would make for bumpy drywall or sheathing so I have made it a practice to fasten all crooked plates with the convex profile facing inward towards the middle of the wall.  Since the second set of plates had to be suspended while nailing, I cut spacers to fit over the plates nearest the floor that held them in place but slightly too low.  Then it was a matter of shimming them a little into a nailing position flush with the edge of the truss.

Only one other job remained before the wall was ready to raise.  I scabbed together the segmented plates with short boards and drywall screws.  This maneuver stiffened the wall for raising; the scabs were removed later.  In fact, one of the scabs interfered with proper placement of a truss so its installation was delayed until after the wall was raised and the scab had been removed.

The wall aligned, secured and ready for covering (click on the image
to enlarge it for more detail)

Raising and Aligning the Wall
I used a spirit level to check the north concrete wall and the south truss wall and found that they were both plumb.  Therefore, I could use the distance between them at the floor as the measurement for the length for the wall while allowing a 1/4" tolerance.  The three guys helping me raise the wall were skeptical about such a close tolerance only to have to witness me puffing up and strutting around when it went to place exactly as planned.  

After it was in place, I secured the top by clamping it to the adjacent floor joist to which it would eventually be nailed after the wall was aligned.  The wall had no choice but to be plumb in a north-south direction as it fit tightly against the concrete north wall and the truss south wall.  And fitting against the floor joist took care of its straightness at the top. All that remained for alignment was getting it plumb in an east-west direction after fastening it to the joist.  Plumbing also automatically aligned the bottom longitudinally. The top of the wall ended up being pretty level despite the concrete floor being anything but (see Major Concern below).  Minor differences will be handled as the rest of the rake wall is stick-built on top of it. 

Fastening the Wall To Place
I nailed the top of the wall to the second story floor joist and to the existing south wall. Fastening to the joist took the place of a second layer of 2 x 6s commonly found in double top plates to bridge over joints in the first layer and to tie together intersecting walls.  The need for a double top plate was further diminished by having two top plates side-by-side to begin with.

I fastened pressure treated two-bys to the north concrete wall with robust (1/4") Tapcon screws and nailed the north end of the new wall to them.  I supported any gaps between the bottom plates and the insulated concrete foundation or the concrete floor with composite shims under each truss.  I then used the 1/2" galvanized anchor bolts, that were installed into the top of the foundation when the concrete was poured, to tie down the bottom of the wall.  Because one of the bolts fell in the doorway and had to be cut off, there were only three bolts for the entire wall so I added three more using 1/2" concrete anchors.

Shielding the Wall from the Weather
Since the trusses have gussets made from interior plywood and OSB board and, since my

snail-paced construction schedule means they will be exposed to rain and snow unduly long, I immediately covered the wall with lumber wraps.  I expected to be using relatively expensive 6 mil plastic sheeting (at nearly $100 for a 100' x 24' roll) until realizing that the lumber wraps were free for the asking from my local lumber yard.  I used a lot of staples and a few batten boards to fasten the wrap. How well it resists the wind remains to be seen.  At least I get to test it on this short and easily-accessible wall before using it for the tall wall described in the previous post.

The inside to the left also needed covering

Looming Problem
The concrete floor on which the wall sets is not level, in one area being 1/2" out of level over a distance of only 6' or so.  Nor is the concrete foundation perfectly level within itself or level with the floor in many places.  Consequently, there were some serious spaces under the bottom plates that have to be air-sealed in some manner.  At the time of this writing, I am not entirely sure how to do it, whether to use mortar or spray foam or caulk or ???????.

This will be a common problem for all of the exterior walls so I will have much to say about it in future posts starting with the next post on building the short truss wall on top of the earth contact north concrete wall.  The top of the concrete wall is decidedly unlevel so the challenge of standing a level wall on it without gaps between will be addressed.