Construction - French Drains - Design, Fabrication and Installation (Cont'd some more)

Grading of the Building Site and Setting the Laser
My excavation of the building site left uneven contours in which water pooled after a rain. The first thing that the contractor did with a track loader just before a frog-strangler rain was to smooth and slope the grade.  As a result, the site was dry enough in a couple of days to resume work.  

The laser beam was then set up at 5' above final floor height to serve as the bench mark for all subsequent excavation and grading.  The proper fall for French drains is the same as for soil pipe -- quarter inch per foot which is a 2% fall.  The laser was then set for 2% to avoid having to calculate the fall as the trenching progressed.  For instance, the easternmost trench began at the north end at 14'  (5' from laser to floor level plus 9' below floor level)  and ended at the south end near the future rain garden at 17' automatically.

Rock for Backfill
Several truckloads of two kinds of rock for backfill were delivered by the contractor -- natural pea gravel (the brown pile in photo) and 1" clean quarry gravel (the light gray piles). Since the French Drains were made from

In goes 1" clean

single wall culverts (French drain construction) and might be damaged by backfilling with 1' clean from a height of over 10', we decided to bed the drains on a few inches of pea gravel then cover them for a few inches with more pea gravel before dropping in the 1" clean.   In addition to protecting the culverts from the coarse rock, the gravel would also facilitate water flow into the perforated culverts as well as carry water parallel to, but outside, the culverts -- at least for a while. Our soil engineer had warned that any gravel not protected by the proper geo-textile fabric would eventually silt up and behave more like the adjacent soil. As far as the the geo-textile wrap for the culverts was concerned, it was designed to filter out silty soil whether it contains gravel or not, thereby rendering future siltation of the gravel moot.

Last-Minute Changes to the French Drain System
The drawing shows the planned French drain system having seven north-south

N-S French drains in green and yellow; manifold in red (click on photo to enlarge)

drains with some of the eastern-most connecting with a diagonal "manifold" that empties at a future rain garden. While the internal diameter of the seven drains is 8", the connecting manifold is 12" and of double-wall construction in order to bear the weight of heavy vehicles on the driveway. 

What the drawing does not show are two last-minute changes to the system.  One is a long east-west French drain connecting the north ends of the seven north-south drains. The culvert-less trench was lined to a height of 6' with geo-textile fabric before backfilling with rocks and closing at the top with a "burrito wrap" of the fabric.  

The original design for all seven of the N-S drains was a pea-gravel-protected, fabric-wrapped, perforated, culvert backfilled with 1" clean and topped off with soil.  The trench was not to be lined with fabric. The second last-minute change was to use the same fabric, rock backfill and burrito wrap that was used for the E-W connector for the eastern-most (#7) drain even though it already had a wrapped

Digging E-W connector to a depth of 9 feet

culvert at the bottom. The rationale for these changes was to intercept ground water as it flowed from N-E to S-W anytime the water table rose decidedly above the height of the wrapped culverts (9' below floor level) even though the probability was minimal. 

Trenching, Installing and Backfilling
The trenching was done by a backhoe with a 24" bucket beginning with the N-S connector then the westernmost N-S drain and progressing eastward with the rest of the drains.  As soon as a trench was dug and the bottom lined with a couple of inches of pea gravel, the pre-made wrapped culvert was snaked in and lowered into place with ropes by enough volunteers to ensure its safe handling and proper orientation in the trench.  As soon as the drain was in

"Snaking" in a pre-made French drain

place, it was embedded in more pea gravel then backfilled with 1" clean to within a couple of feet of the top.  Then enough soil was swung in from the next trench being dug to bring the backfill to grade.

Construction - French Drains - Rationale, Fabrication and Installation (cont'd)

This is the second of three posts on the French drains.  The first post delt with the rationale for and the prefabrication of the drains from culverts.   This post bridges from the prefabrication phase to the installation phase in the third post.

Overall Design
The individual homemade drains were detailed in an earlier post and can be seen in the distant background in second photo.  The

12" double wall conduit for the manifold

perforated portion of each 8' drain is 60' long to which enough additional un-perforated pipe was added to reach daylight downhill near the future rain garden or to empty into a common 12" manifold that went to daylight near the garden. The perforated sections and one 20' section of unperforated pipe were assembled ahead of time. The remainder of the system had to be assembled on site as the trenches became available.

 

Entire system ready for installation

Professional Help
The need for an elaborate system of French drains and its cost was not anticipated when budgeting originally. Consequently, I tried to imagine ways of trenching and backfilling without professional help (original plan). However, it became clear that the additional cost of help could somewhat, but not entirely, be justified by savings of time, materials and equipment rental. 

In terms of time, it took only two days to lay the French drains in mid-April which allowed us to get a leg up on what is usually our wet season in May and early June. This was perhaps the primary benefit of seeking help.

In terms of materials, rock dropped from a height of 5' is sufficiently self-compacting to support foundation footings and slabs. Therefore, rock is preferable to soil for backfilling because soil has to have the correct moisture content then must be compacted in shallow layers (lifts) with compacting equipment. In terms of material costs, the pay-off from using professionals is that they have the right combination of equipment to minimize the amount of dirt that has to be removed.and replaced with expensive rock.

In terms of equipment rental, no mini-excavator or compactor was necessary -- no small savings.

Brian Hayes Construction
Consequently, we did the right thing in hiring Brian Hayes, a local contractor, at a time when he was not overbooked and able to do the work himself.   As luck would have it, his Dad (with a lot of help from a youthful Brian) had DIYed their home-place in which, as a science teacher, Mr Hayes had incorporated sustainability concepts that were way ahead of the curve.  

Brian was not only willing to work with a DIYer but seemed to take genuine ownership of our energy-neutral project (perhaps partly in his Dad's memory?).  He provided the rock and did the trenching and backfilling for the French drains. He facilitated the rapid installation of the AGS system, he graded the house footprint to final depth in preparation for the rock sub-base for the concrete slab and he dug the shallow trenches for the foundation footings -- all of this in essentially five working days despite having to deal with an amateur track-loader operator (me) and a volunteer crew.  Conservatively, he saved us as much as two months time over DIYing the French drains and the AGS system.

Brian's invoice equaled +/- 20% of our total home-building budget.  The ratio for the French drains to the AGS system was 8:2, i.e., 80 % of the 20% went for the drains, a cost that is somewhat easier to reconcile by knowing that any conscientious construction let into our wet hillside would have required them irrespective of our need to use them to protect the AGS system.  The 20% for installing the AGS system was more in line with what was budgeted.  

Our hope is that unanticipated synergies during the remainder of construction will help to offset the French drains costs.

Prime Examples of Unanticipated Synergies Immediately
Our project already has enough life to attract unanticipated synergies, most of which so far have taken the form of volunteer labor and opportunities for salvage.  Brian brought another dimension.  His 35-year experience in the field, and I, with my research-based design, were able to collaborate amiably on the fly as the French drains and AGS conduits were laid, producing much better outcomes than I would have had sticking to the original design. Since part of Brian's business is razing old buildings (60 or more per year), he now plans to watch in our behalf for salvaged lumber opportunities - another unanticipated synergy.  Still another:  I thought our narrow dead-end street precluded semi-truck deliveries, which meant unloading onto a main street and using the track loader to schlep individual skids several blocks (as I did with the pallet of geo-textile material). Through Brian's connections, a vacant contractor's property at the end of the street will be available as a turn-around for semi's.

First the Driveway
But back to the actual construction of the French drains, the first thing Brian prescribed was to site and rock a driveway to give access for the delivery of rock for the French drains and a place to dump it.   It also would be mandatory for the

Rough-in for drive and turn-around area near future garage

Ready-Mix trucks and others delivering materials later. Accordingly, I used garden hose, then marking paint, to outline the driveway to the street and the turn-around area near the future garage.   With the track loader, I removed about a foot of soil. 

As a naive DIYer laying out his first driveway with garden-hose-and-paint precision, I envisioned a nice driveway from the time the first gravel arrived until we moved in. Wrong!   By the time gravel truck after gravel truck arrived, it was deeply rutted and had to be repaired over and over.  And, when dumping space closer to the house footprint was usurped by installation of French drains, the driveway to the street became the default site for several more truckloads of gravel which spilled over the sides of the original driveway footprint to the extent that it became impossible to tell exactly where the driveway was supposed to be.  Oh, well!

To continue the story on French drains go on to the third post.

Construction - French Drains - Rationale, Fabrication and Installation

This post is the first of three on the French drain system installed below the house to prevent ground water from comprising the Annualized GeoSolar system.  

Why Are French Drains So Important for Our Project?
Just to review, we are using a passive solar system called Annualized GeoSolar in lieu of conventional heating and air conditioning.  It captures the heat from the summer sun and

stores it in the soil under and adjacent to the house. 

The keys to the AGS system are conduits (red in top drawing) angled slightly upward from the solar collector in front of the house (blue). A total of ten conduits fan out through the soil a few feet below floor of the house then terminate in a common solar chimney behind the house (green).   Heated air traveling slowly and passively from the collector to the chimney gradually raises the ground temperature to a floating year-around preferred temperature of 74 degrees (+/- 4 degrees). When heat is lost through the envelope of the house in winter, it is replenished from the enormous thermal

mass under and adjacent to the house. Since heat seeks cold, the heat that enters the envelop during summer goes immediately into the mass to augment the heat already being produced by the solar collector for the upcoming heating season.  Of course, during cold months, the chimney has to be closed to keep counterproductive cold air from dropping into the conduits.

Ground Water Problem  
For the system to work, the soil under and adjacent to the house must remain perfectly dry because water carries the heat from the conduits to the water table before it can be used to condition the house.  In dry climates with a low water tables, all that is needed to keep the soil dry is an "insulation-watershed umbrella" (orange in the second drawing) which not only insulates the soil for a distance outward from the house to increase the size of the thermal mass, it also keeps rain and snow-melt from saturating the soil in the thermal mass.  In wetter climates with high water tables, the umbrella is not enough.  The water table under the house (curvy blue line in second drawing) must be lowered sufficiently that it is unable to steal heat from the conduits.  In our case, French drains are necessary because the water table is known, through the monitoring of four piezometers over several seasons, to rise in the late spring to the level of some of the conduits and much too close to the others.


Typical French Drains

Perforating with circular saw

Typically, a French drain lies close to the surface of the soil such that it is possible to fabricate the drain safely by either entering a shallow narrow trench or reaching in from above. Also typically, the drain is about a foot square in cross-section and contains clean stones of consistent size, say, 1" in diameter, with a 4" perforated pipe wrapped with geotextile fabric embedded in the stones a couple of inches from the bottom.  Then the entire drain is wrapped in geotextile fabric. Most of the water trickles through the rocks. The pipe provides rapid egress for water when the rock bed is full. The outer geotextile fabric keeps soil particles from entering and clogging the rock bed while that around the pipe provides additional insurance against a clogged pipe.

Hog ringer with hog ring  

Homemade French Drains
Our situation is unique in that, protection for the AGS conduits required the drains to be eight to ten feet below floor level (black line at the bottom of the first drawing) and situated in narrow trenches too dangerous to work in.  Consequently, I pre-made the drains and had them ready to drop into the trenches with ropes from above.  

The drains were fabricated from eight inch single-walled culverts that were perforated on both sides in the lower hemisphere with a circular saw whose blade guide was rigged to control the depth of cut. The perforations were about 4" apart. Then three culverts were joined together with one unperforated section using split unions secured with wire to make 80 ft lengths. Finally, the perforated sections of the culverts were wrapped with perhaps the only geotextile fabric extant that will not be clogged by our wind blown loess (silt) (research on fabrics).   The 12' width of the fabric was cut by the supplier into 36" strips which was the perfect dimension for overlapping  the culvert on top, rolling the edges

Hog-ringing the seam

under twice and fastening with hog rings. In order to eliminate the possibility of the culverts rotating relative to each other during handling and thereby causing some of the perforations to be in the top, less effective, hemisphere, hex headed sheet metal screws were used to secure the culverts to each of the wire retained split unions. 

Since at the time of lowering into the trenches, the textile fabric roll on top of the culverts

was to be used to position the assemblies in the trenches, it was important to prevent rotation of the fabric on the culverts during handling.  So a few sheet metal screws were driven through the roll into the culvert. 

Finally, the up-slope end of the pipe assemblies were closed by doubling the fabric back

Hog ring in place

over the culverts a few feet, wrapping and folding then securing with hog rings.and sheet metal screws similar to the way the fabric was handled along the length of the pipe.

Here we have focused on the individual drain. A succeeding post discusses the entire French drain system comprising seven legs -- some joined to a common "manifold" before reaching daylight and some going to daylight directly. Excavating and backfilling the trenches.are covered as well.

For a complete story, continue on to the  second post on French drains. 

Odds 'N Ends - Whole Wall R-value

The Envelope and Green Building

The skin of a building comprises the walls, roof or ceiling, floor, windows and doors collectively known as the "envelope".  Green building is largely a matter of keeping heat from entering or exiting the building through its envelope and doing so with minimal impact on finite resources.

Heat Transfer

Heat is transferred in three ways: 

• Conduction - through solid objects, called "thermal bridging" when it is applied to green building
• Convection - through fluid motion (air is a fluid), called "air infiltration" in green building parlance
• Radiation - heat transfer in a straight line through space such as sunlight passing through a window and warming a floor

R-factor

Insulation is just another solid object through which heat passes. The better the insulation,

the slower heat passes through it.  In fact, the "R" in "R-factor" means "resistance" to the

passage of conductive heat and is usually expressed either in terms of R-factor per inch or R-factor for the entire thickness.  For example, 2" thick solid foam is labeled R-10 which makes it is R-5 per inch while batt fiberglass insulation for 2 x 6 walls is R-19 or R-3.5 per inch.

Comparison of fiberglass batts vs. blown-in cellulose

Air Infiltration
Even though it is not its primary function, insulation can also be a barrier for air infiltration. Blown-in foam insulation totally stops air leakage. Cellulose can be packed so densely that little or no air passes through it. Loose fiberglass (not batts) can be similarly densely-packed. The low-tech rice hulls that we will be using for insulation in a 15" wall also makes air leakage relatively moot, although we still plan to do all we can with caulk and spray foam to stop leakage. Fiberglass batts, on the other hand, are difficult to fit into the space between studs precisely enough to stop air.  And it is even more difficult to do so around such obstacles as electrical and plumbing components.  For dense-packed fiberglass and cellulose, for spray foam and for rice hulls, it is inappropriate to base the R-factor on resistance to conductive heat loss alone. All of these insulations, ramp up whole wall R-value by eliminating air infiltration either nearly or totally.

Whole Wall R-values

The effectiveness of an insulated wall is compromised when structural members, like studs and headers, penetrate completely through the wall.  They serve as bridges for heat transfer (hence "thermal bridging") that degrades the R-factor for the whole wall. (Have you ever noticed how snow melts faster over cathedral ceiling rafters than over the insulated spaces between them?)  Therefore, the whole wall R-value that takes into consideration thermal bridging is an important concept.  A 2 x 6 wall on 24" centers, for example, goes from the R-19 stamped on fiberglass batts to R-13.69 due to heat loss through the studs, a 27% drop.  A 2 x 4 wall on 16" centers experiences the same 21% drop, going from R-13 to R-10.*  To brag a little, the super-insulated home we are building not only widens the space for insulation to 15", but it also uses truss walls instead of stud walls to minimize thermal bridging.  Altogether we expect a whole wall R-value of over 50 which will be necessary for sole dependence on passive solar heating and air conditioning.

Air Sealing is the Secret

The recommended minimum for wall insulation in our climate zone is R-18.  Not even 2 x 6 walls insulated with fiberglass batts with a whole house value of R-13.69 meets the minimum, much less the ubiquitous  2 x 4 walls.  However, if the envelope is thoroughly sealed against air infiltration, the house is significantly more comfortable than a drafty house with the same amount of insulation.  Conversely, a house with serious air infiltration might be impossible to make comfortable with any amount of insulation. 

Air sealing is finally getting its due.  Building codes for new construction now require that all potential air leaks be caulked or spray-foamed before closing up the walls and ceilings. And blower-door testing can be added as a way of measuring the thoroughness of air sealing. The reward for doing the right things for sustainability -- sealing and testing -- nets a speedy return on investment through lower energy costs.

Replacement Windows
The energy benefit, from replacement windows derives not from upgrading the glass itself from double pane to single pane, considering that single pane is R-0.85 and double pane is  R-1.5 - 2.0 -- nothing to write home about.  The benefit from window replacement is the opportunity it provides for sealing air leakage through and around the window openings, especially the leaky compartments on either side of a window that previously housed the window weights. Replacement of metal framed windows has the added  benefit of significantly reducing conductive heat loss.
_____________

* These whole wall figures were used on page 12  by Michael Morley in his book, "Building with Structural Insulated Panel (SIPS)"  while comparing SIPs, with their minimal thermal bridging, with conventional stud wall construction.

Timeline - Design Evolution - Insulation

2025 Important Cautionary Update

Immediately after moving into the house 2 1/2 years ago, we began seeing a tiny insect, particularly around the windows, only to realize that they were rice weevils.  The detailed discussion of the weevil problem renders this post academic. 

*        *        *        *        *        *        *        *        *        *        *        *        *        *

The Choices

As our design evolved, I waffled as much on the insulation as on French drains, the extent and composition of the earth contact walls or the design of the foundation walls. Hands down, spray foam would be the best choice for several reasons and worst choice for several other reasons.  Its positives are a high R value (at 6.5 per inch, it is twice that of most other insulations), it blocks air infiltration and it adds strength and rigidity to the wall or ceiling. Its negatives are that it mostly comes from petroleum (as I understand it, the addition of soy is more greenwashing than creditable), has high embodied energy, out-gases VOC's for a while after application and it is incredibly expensive especially since it is not DIY-friendly and our design for a super-insulated house calls for exterior walls and cathedral ceilings that are extraordinarily thick.

So what else?  My second choice was cellulose for both the walls and the ceilings. The advantages of cellulose are that it is made from recycled paper, which is about as green as it gets, it does a pretty good job of sealing off air infiltration, particularly in walls as thick as ours will be, it can be sprayed into walls before drywalling much like spray foam and it is relatively inexpensive.  There is one important disadvantage of cellulose for our cathedral ceilings -- for a decent R value, it has to be dense-packed which requires the space between the ceiling and sheathing to be entirely enclosed so as to be able to pack the insulation densely. With our mini-attic (cathedral ceilings) design, the space that exists under the sheathing precludes dense-packing.  The same goes for loose fiberglass which would also have to be dense-packed to function well in a cathedral ceiling.  Fiberglass batts, in my opinion, are not an option for super-insulation. However, we may use them for the 6" exterior walls of the garage (but not the wall between the garage and the house that will match the other exterior walls).

Rice Hull Insulation

Since none of the conventional insulations suited our needs perfectly and still needing to reduce costs, I went searching for alternatives.  I had reread Don Stephens' article on Annualized GeoSolar dozens of times but blew off his argument for rice hull insulation as impractical for us. About a year ago, when searching for alternative forms of insulation, I finally Googled rice hulls.  Up pops The Rice Hull House, a slide show on truss walls and rice hull insulation emanating from Washington, LA.  Also I found an article written by Paul Olivier on the  attributes and physical properties of rice hulls.  (I suspect that Paul was the one posting the slide show as well.)  

Price Quotes
After getting a price quote on hulls from a southeast Missouri rice mill that was

considerably higher than the price Paul mentioned in his article, I exchanged emails with him seeking input.  He advised looking further, that they should be available at $15 per ton. A freight quote for a grain trailer-load (18 tons) from Arkansas or Louisiana was $500-900. Eighteen tons at $15 per ton plus freight would run $770 to 1,170 for as much as my rough calculation suggested we would need to do our entire house.   If not, depending on the amount we were short, we would either finish up with cellulose or order another trailer-load.  

A quote for doing the whole house with cellulose was $4,700.  God only knows what it would be for spray foam.  And the quotes would be higher now because our design has recently morphed into walls and ceilings that are several inches thicker.

Properties of Rice Hulls
As explained in the article by Olivier, rice hulls possess a high silicon content which makes them essentially inert when it comes to combustibility, mold growth, vermin and insect support -- even more so than for cellulose.  The thermal resistance (R) factor for rice hulls when poured or blown into a wall or ceiling cavity is R-3 per inch, which is similar to loose fiberglass and cellulose.  Interestingly, Paul also said diatomaceous earth can be added to the hulls as a further measure against termites. The earth particles find their way under the insects' shells and abrade them to death.   

Timeline - Design Evolution - Energy Efficient Roof

Past Three Years

Cathedral Ceilings

From the beginning, we envisioned cathedral ceilings in lieu of an attic but was unsure as to how they should be constructed to maximize energy conservation. My first choice for the roof-ceiling complex was structural insulated panels but as discussed in the previous post, SIP discussion, they were not budget-able.  So, that's OK -- we'll just throw up some 2 x 12s, sheath and waterproof the tops, drywall the bottoms and insulate in between, right?  Whoa, I was soon to find out that it's not that simple.

Moisture Condensation

Moisture condensation occurs when the ceiling is not air tight, which includes most ceilings.  As air moves from the

Download pic:  2 x 12 rafters for a cathedral ceiling

interior, it takes moisture with it that condenses as it approaches the cold side of the roof in winter, rotting the wood, encouraging mold and lowering the insulation R-value.  And this phenomenon is much more critical for cathedral ceilings than for ceilings under attic space.  The moisture reaching attic space is dried by attic ventilation before it can do damage.  But not so with enclosed cathedral ceilings.

The internet is rife with the pros and cons of using the typical sheathing-rafter-drywall approach for cathedral ceilings.  My interpretation of the chatter is that 2 x 10s or 2 x 12s with sheathing and drywall will work under some circumstances and not others. When urethane-type spray foam fills the space between sheathing and drywall, air infiltration is nil, even without a sheet plastic moisture barrier, so moisture condensation is moot.  But spray foam does not fit our budget.  When it comes to dense pack cellulose or dense pack fiberglass insulation, arguments fly back and forth -- some say they inhibit air infiltration enough that condensation is not a problem and others warn against using them in a closed cathedral ceiling.  Batt insulation is leaky enough that, even with a plastic moisture barrier, is probably not worth the risk.

Mini-Attic with 2 x 12s

Download pic:  Ventilation for cathedral ceiling

The nay-sayers advocate using what essentially is a mini-attic at the top of the rafters to allow any air passing through the ceiling to exit into ventilated space.  Then, even if some condensation takes place, it readily drys because it is in contact with moving outside air. This is the cautious approach I have decided to adopt so we can use any type of insulation and not worry about condensation issues. And, to stop air from entering the ceiling in the first place, I plan to use at least one layer of carefully-detailed 6 mil plastic sheeting as a moisture barrier below the rafters before installing the tongue and groove natural wood ceiling which, unfortunately, has more potential for air infiltration than drywall unless it is backed up by some sort of solid sheet material like 3/8 drywall or Masonite.  (Recent update:  As voiced in subsequent posts, I learned that plastic sheeting above the drywall is ill-advised for our climate; it will not be included in our design.)

To create a mini-attic, I plan to use structural (construction) screws to fasten 2 x 4s on edge and on 24" centers at a 90 degree angle to the

Download pic:  Natural wood ceiling

rafters and fasten the sheathing to them. Then, by keeping the insulation flush with the tops of the rafters, the "attic" will comprise the 3 1/2 inch void between the insulation and the sheathing that will be ventilated through the soffets. I had already planned to use foil-faced OSB board for sheathing in order to reflect radiant heat in summer. It is typically installed with the foil side down, which should also allow it to handle better than the "raw" OSB side of the sheathing any moisture condensation that reaches the mini-attic and takes a while to dry.

R Factor

Since I plan to insulate both the walls and the ceiling with rice hulls that are rated at around R-3 per inch, a 2 x 12 roof gives at least an R-36.  But, consistent with super-insulating, I would like to shoot for R-50.  Consequently, I plan to secure with structural screws a 2 x 4 edgewise to the bottom of each rafter in order to provide for an extra three and a half inches of insulation.

However, the above modality does not address thermal bridging through the rafters.  I plan to handle this problem by ripping 1" thick foam board into strips and sandwiching them between the bottom of the rafters and the edgewise 2 x 4s.  They will also increase the height for the loose insulation by another inch.  The roof will then be at least R-50 with thermal bridging controlled.

Mini-Attic with I-Joists

An alternative choice for a mini-attic system would be 15" I-joists instead of  the stick-built 2 x 12 system described above. The advantages would be simple installation, fabrication from renewable sources and minimal thermal bridging.  Their disadvantages would be greater difficulty mating them to the 15" exterior walls than with the 2 x 12s and higher cost. Not only are the I-joists more expensive than new 2 x 12s per linear foot but the stick-built approach will allow the use of materials recycled from tear-downs for an additional savings.

Recent Update 

Much of the above design became moot late in 2016 when I decided in favor of 16" and 18" tall roof trusses in lieu of joists -- either 2 x 12s or I-joists -- and a double layer of roof sheathing with a 3 1/2" space between layers to serve as a dedicated mini-attic.  Makes me grateful that the building inspector was happy with architectural drawings that were sufficiently nonspecific that I could improvise on the fly.

Timeline - Design Evolution - Stick Built Exterior Walls

Past Five Years

Ruling Out Alternatives for Exterior Walls

One of the first decisions we needed to make was whether to use an alternative to conventional stick building for the non-earth-contact walls.  The options are plentiful -- rammed earth, straw bale, Earthships, adobe, cob, cordwood, earth bags -- to name those that Daniel Chiras covers in his "The Natural House -- A Complete Guide to Healthy Energy-Efficient, Environmental Homes".  

My first choice however was structural insulated panels even to the extent of buying Michael Morley's book on SIPS and getting preliminary quotes from a couple of vendors.

The panels are structural enough to eliminate conventional framing, have built-in insulation, allow no air infiltration, are constructed off-site to cad-cam precision and go up so fast that the house is under roof in a matter of days.  SIPs work especially well with timber framing (most timber-framed houses today are enclosed with SIPs)  but the two together would have exceeded our total budget!  Time to look for an affordable alternative.

Stash of Salvaged Lumber

We are sitting on quite the stash of lumber that I salvaged from old buildings -- three houses, two garages and several outbuildings, plus freebies picked up through Craigslist. Most of the dimension lumber is 2 x 4s which, if installed in the usual way, would not provide the R-45+ wall we need for a super-insulated house.

An important but poorly understood reality is that, irrespective of the R-13 the manufacturer prints on the insulation, the whole wall R- factor for a 2 x 4 wall is no more than R-10 when thermal bridging through the studs and plates is factored in. And that's even before air infiltration enters the equation.  Similarly, due to thermal bridging, the actual R-factor for a 2 x 6 wall is R-14 instead of the  R-19 printed on the batts.

So, how do we build R-45+ walls using the salvaged 2 x 4s?

Double Wall Construction

Early on, I had considered using a double-wall construction which is basically two separate walls tied together at the top and bottom plates with the amount of space between the walls dictated by the R factor goal.  And for the double wall, I was thinking about using 2 x 4s rotated 90 degrees from normal studs in order to reduce thermal bridging and to better manage inconsistencies in the salvaged lumber. However, before I had worked out the details, I made a lucky strike.

Truss Walls

While searching for a cheaper alternative to conventional insulation, I remembered Don Stephens mentioning rice hull insulation in his article on Annualized GeoSolar.  When I Goggled rice hulls, I found a blog that detailed the construction of truss walls in conjunction with rice hull insulation. The blog was posted by a Louisiana company building low-cost, Habitat for Humanity-like, housing using wall trusses filled with rice hulls for insulation (Rice hull house with wall truss design).  The important difference between their trusses and my double-wall idea was that the trusses can be made individually in a jig and assembled in the wall as if they were studs.  This approach is much simpler than building and raising two separate walls then joining them at the plates, particularly for someone working alone.

At the time of this posting, the wall truss jig was nearly assembled in the workshop in anticipation of building the 65 or so wall trusses ahead of time on days too inclimate to work outside. The jig will yield trusses that are highly standardized despite the inevitable inconsistencies in salvaged lumber.

Odds 'N Ends - A Broader View of Sustainability

The Food Chain Starts with Plants
Only plants are capable of converting the sun's energy into food (photosynthesis). Plant-eating insects and animals -- herbivores (with minor input from omnivores) -- convert plant tissue into animal tissue.  Without plants and plant-eaters, there would be no higher forms of life, including humans.  Unfortunately, many plants, insects and animals at the lower end of the food chain are on the road to extinction due to habitat loss and degradation.

In my opinion, anyone interested in a broader view of sustainability should read the eye-opening "Bringing Nature Home - How You Can Sustain Wildlife with Native Plants" by Douglas W. Tallamy, professor and chair of the Department of Entomology and Wildlife Ecology at the University of Delaware in Newark, Delaware.

Three Problems
The gist of the book is that plants capable of supporting the essential herbivores are under assault on three fronts.  First, habitat for natives has been replaced by agriculture, lumbering and urban sprawl (think herbicidal control of milkweed, the sole food for monarch butterflies, and loss of their winter habitat in Mexico due to illegal lumbering). Second, pests that hitch a ride on imported ornamentals cause extinction of natives (think chestnut blight, Dutch elm disease and Emerald Ash borer).  And finally, alien plants, with no natural enemies, out-compete natives (think Russian olive, Japanese honeysuckle and kudzu).

To make matters even worse, alien plants hog resources (nutrients, water and sunshine) but, since they are rarely eaten by native herbivores, contribute nothing to the food chain. The ubiquitous foundation plants, ornamental trees and shrubs, as well as cool weather grasses, are all problematic in this regard.


Habitat Fragmentation - Habitat Islands
Instead of the original thousands of acres of contiguous native habitat, the habitat remaining today is in the form of isolated islands that are usually degraded by alien invasives, foul air and chemical run-off.  "Bringing Nature Home" means gardening and landscaping with enough native plants to support wildlife.  Native trees, shrubs and prairie plants bridge between isolated islands and provide sustenance at the bottom of the food chain that may actually be better than that provided by the islands themselves, particularly in heavily populated areas.  

Co-evolution
Native plants support native herbivores because the plants and insects/animals evolved together over millenia.  The native herbivores do not recognize alien plants as food and, if they were to be tempted, likely do not have the ability, bestowed by co-evolution, to overcome the plants' defense mechanisms.  A white oak tree, for example, supports 534 species of lepidoptera (moths and butterflies) while a Bradford pear tree or Japanese honeysuckle bush support practically none (Tallamy).

Resources 
About half of Tallamy's book is advocative and half is helpful hints for getting started with natives and guidance on region-specific plant selection.  

Wild Ones (www.wildones.org) is an organization that promotes landscaping with native plants and, of course, the eradication of non-natives.  As members for several years, Dorothy and I have visited many beautiful private and institutional native landscapes. And we have mingled with the choir -- a unique subset of interesting environmentalists who are only too happy to share, not only their knowledge, but plant seeds as well.

Our Progress
Although we are at least two years away from completion of our home, we have already eliminated most of the alien plants that overran the property initially. Our new natives are beginning to flourish in areas that will not be disturbed by the construction and their progeny will dominate after construction.

Fortunately, here in the hilly bluffs of the Mississippi River, the habitat fragmentation is less intense than in the surrounding countryside--more like peninsulas than islands. Our hope is that someday our 4+ acres of mostly natives will be a bridge between two adjacent (struggling) peninsulas.

Oh, by the way, did I mention that native gardens and grasses slow global warming? Yep, in two ways--by storing far more carbon than foreigners like fescue, zoysia or Japanese yews and by avoiding the carbon inputs of watering, fertilizing and mowing (Native plants and global warming).

Timeline - Design Evolution - Foundation Walls

Past two years

Uninsulated Foundations
Nothing riles my sustainability sensibilities more than the sheer ubiquity of above-grade block and concrete foundations.  "Walk-out basements", with even more exposure, are prized even if they face northwest.  

Walk-outs facing northwest.

Many newer houses have retrofitted interior insulation, especially for finished spaces,  While better than no insulation, the thermal mass of the concrete ends up on the wrong side of the insulation for maximum performance.

A serendipitous example of heat transfer (conductive heat loss) through an uninsulated

foundation is seen in the photo to the right. Dorothy planted a tropical plant (Christmas cactus) next to the house never expecting it to make it through one winter, much less five. The heat conducted from the basement warms the soil enough for he plant to survive the cold season. In the other picture, notice how the snow melts first near the wall. The tan remnants of the cactus on the ground at the top of the photo indicate dormancy, not death.

Our Problem
Over the past two years, I have spent an inordinate amount of time worrying about the design of the foundation under the stick-built walls.  As our original concept of earth-sheltering morphed from earth contact on the north and west sides of the house into earth sheltering on on the north side only, the need for stick-built walls increased. And more stick-built walls meant more energy-problematic foundation walls.

Protecting the Thermal Mass

A lot of the heat that the AGS System generates and stores under the floor would bleed out through the foundation walls and be lost unless the entire foundation is insulated.  But doing this on a low budget is a challenge.  I looked at several ways.

Dry-stacked cider blocks would work structurally and affordably (per Rob Roy in his book, Earth-Sheltered Houses- How to Build an Affordable Underground Home), but we would be back to the waterproofing and insulating conundrum he wrestles with. The Complete Block System would work well and, like the cinder blocks, would be DIY-friendly but too expensive.  A DIY-poured concrete wall in rented forms would not be a bargain, would be lots of work, would be too tall for the frost-protected shallow foundation (see below) and has the same bad choices for insulating and waterproofing as the cinder blocks.

Insulated Concrete Forms

As it turns out, the best value for the foundation walls for us will be concrete poured in insulated concrete forms (ICFs). Setting up the forms, which stay in place after the pour, is surprisingly DIY-friendly.

The ICF walls with 8" of concrete and 5" of Styrofoam will be about 13" thick and will fit nicely with the width of the stick-built wall trusses above them.  And ICFs automatically provide slab edge insulation.  This is a big deal because insulation of the slab edge is important for energy conservation and is one of the things that green building certifications such as LEED and HERS highly value but something that normally is difficult to achieve in the field.

Conventional Foundation

A conventional foundation wall in conjunction with a slab floor in our climate must rest on a footing, the bottom of which must be below the frost line -- 30" below grade. Without proper insulation, the foundation is an energy nightmare, losing heat through the wall and, in the process, sucking energy from the adjoining slab.

Frost Protected Shallow Foundation

In our situation, there will be a footer and no insulation under the slab

However, if the wall is insulated on both sides and insulation is laid horizontally over the footing at the base of the outside of the wall and extended outward for a couple of feet, the footing becomes frost proof and can be raised above the frost line and the wall on it shortened as well. The result is a "frost protected shallow foundation".

For us, ICFs are a reasonable solution.  They provide the insulation for the wall above the footing while the insulation in the insulation-watershed umbrella for the AGS system serves also as the horizontal insulation over the footing that characterizes the frost-protected shallow foundation.