I had a recent request to post a photo showing the ageing of the corten and charred siding, this is it…
I haven’t looked at this in a while and was pleased to see how much my clunky rendering of the finished house that is part of the title page is fairly close to what the actual house looks like.
One of the challenges and concerns of this project has been the incorporation of storage into the space. This has been further complicated by our desire to accommodate the nine sleeping spaces, with the space required for the beds competing with space available for storage. I have this conversation with clients often, especially those who are seeking to live in a smaller space: what do you need to store? As I write this I think about the reactions of friends to something we as a family did a couple of times several years ago.
We called it a power fast, really the “p” should be capitalized, more like this: Power fast. Not an aggressive starvation, rather turning the electricity off to our house for a week and seeing how it went. It was a great experience, clarifying and thought-provoking that I won’t get into here, but almost as interesting was the reaction of people we knew. Some folks would be really confused about what we meant about turning the power “off”, unable to imagine such a thing, on the other side would be those that would tell of their experience living in a cabin without electricity for 8 years and getting their water from a dip well.
The storage of stuff generates the same sort of divide, some can’t imagine making do without the basement and attic to fill with boxes, others can pack most of their stuff into a backpack or at most a compact car and don’t know what the big deal is. We’ve had the basement and a large house of stuff and so no matter what this move required a paring, in fact so far it has been two parings to get down to what we are going to have in the new space, but we had to ask ourselves the same question “what are we going to store?”
Beyond the basic stuff that we will come back to, in making our decisions about what was coming with us we settled on love as the defining criterium. It would come with us if we loved it; handmade chairs from a chairmaker in North Carolina, furniture that I had made, a couple of other furniture pieces that we had a place for and held nostalgic value (this is a slippery slope), handmade pottery from potters that we know, other art, some made by Mary or I, most from others, again people we know, books, not a ton but several shelves worth, cookbooks, important in our house. The furniture obviously we have a place for on the floor, the art likewise on the walls, for the books and pottery though we would need to build a place. There’s camping stuff and sewing stuff, files and plans for the business, canning jars and I am sure other things that I can’t remember even if I did put everything in a pod when we moved out a couple of weeks ago (but not in yet, lease was up). There had to be a place made for those things and hopefully for the things I can’t remember right now too.
Understandably the storage in the office will mostly be the home of office stuff. Under and to the left of the bed are file drawers. The spaces above the drawers and on the top shelf above the bed will be open storage for paper and other supplies. On the left wall there are two openings, the nearer, a book shelf, likely to hold code books and other construction reference. the further is an opening to access the area under the entry landing. This is a pretty big space, 50 cubic feet or more, camping equipment?
If you are counting sleeping spots the murphy bed in the living room is number five and six. And in order to both create privacy for guests and expand the available storage we decided to completely cover the wall bed with a sliding cabinet. This cabinet would go from floor to ceiling and when stowed against the wall nest around the bed. The area in front of the murphy bed would be shallower bookshelves while the area around the perimeter would be two foot deep cabinets with both drawers and doors. When pulled fully away from the bed the cabinet would provide a third wall, a curtain extended between the house wall and the cabinet completes a very small and somewhat private guest bedroom.
The planning for this assembly started very early. When framing, we placed the beams in the ceiling that would eventually carry the tracks for the sliding mechanism. The cabinet is entirely hung from above with no support or guides at the floor. I have seen several examples of similarly operating cabinets online, but there was never any detailed information about the hardware that was used. After looking at a couple possibilities I settled on heavy-duty sliding door hardware, and I mean heavy-duty, think pocket doors that weigh a thousand pounds. I initially pursued using Hettich (sold under the brand name Grant) but after a back and forth with their engineers they felt their hardware wouldn’t work. My second choice was PC Henderson. With them I didn’t have to convince anyone that their product would work for my purposes, I just bought what I figured I needed, in this case the track rated for 1500 lbs. or actually two tracks, and four rollers. This isn’t even the strongest track they sell, there is one rated for 4500 lbs. I was working off the idea that the cabinet would weigh around 500 lbs and the stuff in the cabinet would weigh about the same, so somewhere in the vicinity of 1000 pounds total. Divided between the two tracks it puts 500 pounds of load on each track, 250 pounds on each set of rollers. This is well within the load rating of the hardware, but since the load is concentrated at two points 20 inches or so apart instead of distributed across a 4 foot wide door, we made sure to attach the track to the framing in the ceiling frequently, every 8 inches with a 3/8 x 5 inch quick drive lag screw. Each side of the cabinet then is directly supported by 3 screws (discounting the rigidity of the track itself) putting the load on each screw at about 170 lbs. The beams in the ceiling are 11 7/8″ LVL in a ceiling composed of 16″ I joists. The LVL’s are flush to the top of the ceiling joists leaving them 4″ above the bottom of the joists allowing space for the track to recess fully within the ceiling. We left a 1″ slot in the ceiling plywood to allow for the sliding of the connecting hardware, in this case the rollers that we used had 5/8″ threaded rod for the connection. You can probably see the connecting nut in the top of the sliding cabinet in the picture below.
I wanted the cabinet to fill the whole space between the floor and ceiling, no baseboard or crown detail, but since the threaded rods from the rollers hang down below the ceiling about 5″ that meant we had to assemble the components in place. We started by hanging the top bank (which you can see in the photo above) and attached the pieces below one by one. When finished this allowed for a gap at the ceiling of a little less than 1/2″ and at the floor about 3/4″. The disadvantage to this approach is that to remove the cabinet it will have to be cut apart.
None of the photos show the cabinet at completion (we’re still making doors and drawers), but this outlines the process we used to build the thing.
Here is a quick video of the sliding cabinet in action:
Then there is the basic stuff. Clothes: we will need a closet. Broom and vacuum, other cleaning things, toilet paper, towels, bedding, pots and pans, ironing board, iron.
The kitchen is divided between three separate units, the cabinets on the wall, also home to the fridge, dishwasher and oven. The dishwasher is a 18″ unit and the oven is a 24″ wall unit placed below the counter with a drawer underneath. The island is the second component. It houses the sink and has generous prep space. And last the tall kitchen storage, a ten foot tall, twelve-inch deep cabinet divided between a lower section behind doors and an upper section of open shelves. The tall cabinet will also store/display the handmade plates. There are 8 plate shelves on each side of the cabinet that will allow the plates to be stored in the open. You can see them going ladder-like up the side of the cabinet in the photo above. I didn’t mention a dedicated cooktop surface because there isn’t one. The stove top cooking will take place on portable and stowable induction hot plates. There are good reasons for this, there is no gas in the house so we were going to be cooking on an electric surface one way or another, the limited counter space would have become even more limited with a permanent cooking surface incorporated into the kitchen, being able to use the same surface to prep and then cook makes the small space more versatile. Even so it was hard to embrace the idea when we had always, since we had a choice, cooked on gas.
So we bought a Volrath single eye induction hot plate about 9 months ago to try before we got too far along on the kitchen rough in. It took some getting used to, like any new tool, but it has been called a success by the person in the house that does the majority of the cooking. I won’t go into all the advantages of induction cookery, but I will share this one that is germane to the purpose of the project: it is something like 70% more efficient than other stove top methods.
So on to the clothes. Downstairs between the bedroom and the office sits “the closet that holds everything” This would be a 105″ long, 28″ deep space that is our clothes closet, washer and dryer, ironing board, vacuum and broom storage, and some high spaces that will have something in them that we will learn about later. The photo above shows the left side of this space, the tall narrow space on the far left, nearly hidden by Paul, is the broom and vacuum space. The large spaces to his right are for the stacking washer/dryer, between the washer space and dryer space is a narrow slot that will house a pull-out fold-out ironing board. And to the right of that is the clothes closet, all 70″ of it. All of this will be behind doors when finished. For the observant you may have noticed there isn’t a dryer vent in the dryer space. The dryer we are using is ventless, it condenses the moisture from the exhaust air and drains it down the washing machine waste line. Additionally the dryer is heated by a heat pump, a relatively new use for heat pumps in the US, that only requires about a third of the energy to dry a load of clothes. We actually don’t use a dryer often, preferring to hang the clothes outside to dry. The most efficient method of all.
Three more quick mentions of spaces where we took advantage of storage opportunities. The bed has drawers that pull out from the sides to store linens.
And the upstairs bays both have window seats that have cabinets underneath.
Enough about storage.
The panels are at an angle to the support because the house in oriented on a southeast/northwest axis. Even so the panels are not facing due south, more like 20 degrees to the southwest. If we had angled them more and gotten them facing due south we would have run out of room on the support for the solar water collectors. the panels are Solar World 280w with enphase microinverters. Permanent power is scheduled to be turned on next Monday and at that point the panels will start to generate.
One last photo…
After the roofing membrane was completed we got busy putting in place the multiple components that make up the green roof. This started with 2″ of expanded polystyrene. The insulation board of choice for this application is extruded polystyrene, think rigid pink foam insulation at a big box store. But the truth is I had the EPS as a leftover and the only down side is it absorbs more water than XPS which reduces its insulation capacity. Without the added layer of foam that roof is already approaching R-70 so I was willing to sacrifice a little R value to save the money in this case.
The next thing to go down is a drainage layer. This is to provide a free-flowing area for the water that drains through the soil to exit the roof without much obstruction. For this I chose a product called Enka Drain, made by a North Carolina company. The particular drain mat we used consisted of an open fiber mat with an egg carton kind of profile topped by a water retentive layer that was like a thick wooly moving blanket. The wooly blanket holds about 1/4″ of rain to slow the desiccation of the roof soil, a problem for roof soils in hot climates, for example Georgia.
The drain mat we used didn’t come with an integral soil barrier cloth so that is the next layer. Here we used a very heavy non-woven soil barrier that can move a lot of water, 150 gallons per minute per square foot. That is well beyond what the roof could shed or what would ever fall out of the sky. The soil barrier will prevent the roof soil from clogging the drain mat so the roof remains free-flowing. We brought the soil cloth up the edges of the parapet to serve as a protection layer between the roof membrane and the gravel edging that would end up around the perimeter of the roof.
It is common to use gravel as an edge material for green roofs, but it requires something to keep the gravel from mixing with the roof soil. Commercial edgings are often perforated aluminum sections adhered to the roof keeping the gravel in place, but allowing water to drain. Those edgings are pretty expensive so I was looking for some alternative that would be simple and cheap and readily available. After kicking it around for a long time, I came back to an idea I had early on and rejected. Beside my shed I had a pile of header block left from a recent job building a garage, these are 8″ concrete blocks with a 4″x4″ notch out of the top side commonly used as a form for a concrete slab poured into the top a block wall. I realized though that I could quickly cut the block in such a way that once the gravel and soil were in place only a thin line of the top of the remaining block would be visible creating a clean 1 3/8″ concrete separation. And if we placed the block on a bed of gravel it would allow for water to drain freely between the soil and the gravel edge. We also could use the gravel bed to place the block level allowing for the roof soil to end up level, which is preferable. We used silicone to glue the individual blocks together into a continuous edging.
As a light weight filler we broke up scraps of styrofoam and filled the space between the blocks and the parapet to within a couple of inches of the top edge of the block. It keeps it from the landfill and it serves as a good and light drainage medium. On top of the foam we placed a strip of the soil cloth and then pea gravel to the top of the block. All the gravel used on the roof is round stone for the obvious reason that sharp edge stone is a hazard for roof membranes. In the picture above (center top) is a perforated aluminum box we made on site to keep the drainage scupper clear of stone and debris. There is a removable lid to facilitate cleaning the area around the scupper if it ever proves necessary. Around the aluminum box we used a larger 2″ round stone, to make the area around the scupper drain even more readily.
With all these pieces in place we are ready for the soil, something that will wait until construction is nearly completed. Even so we have had a lot of rain this summer to test the roof package as it is now and it has drained very well (even during the violent storm that dropped 3″ of rain in about 30 minutes and uprooted three trees across the street). There is a noticeable delay from the beginning of the rain to when water starts coming out of the scupper, a good thing that will only get better, that means a longer delay, with the soil installed.
With the house in the dry, we were free to move ahead on the interior. Our first step was to make sure that all the penetrations through our air barrier (the plywood sheathing) were sealed before we covered them with insulation and made them inaccessible. With that done we insulated. There are many things that we do on our houses that push well beyond common practices for the area. For much of those, we do them ourselves, but there are some things we choose not to do or are unable to do, and rely on subcontractors, often asking them to move beyond their experience and sometimes comfort to accomplish our end goal. I will admit that this doesn’t always go well. This is sounding very ominous so let me say now that the insulation went fine, but dense packing 12″ walls with the quantity of insulation I needed is not something they had done before. I was nervous about this and considered trying to get a subcontractor from Atlanta that may have had some experience with what I wanted. Finally I decided to use the local contractor I had used for years and with whom I had a good relationship, giving them fair warning, though, that I would be spending a lot of time over their shoulders making sure they understood what was required.
So to keep dense packed cellulose from settling in a tall wall you apparently need to pack it to a density of 3 pounds per square foot. That is really dense, think a full belly after a large meal kind of dense. And obstructions in the wall don’t help, as the insulation fills so fast it will pile up against an obstacle and compact and refuse to flow into a deeper cavity. So having an idea of where the cavities that can’t be seen are in the wall is important to ensure that insulation gets into them . This involves much filling and pushing with hands, really until no more cellulose can be pushed into the hole. I realized during the insulation install that I need to be more careful with making sure that framing components are placed to allow free access around them, making the insulators job easier and eliminating extraneous wood at the same time.
The total volume of insulation in the house is around 2400 square feet, the insulators put 7200 pounds of cellulose in the house, pretty much our target of 3 pounds per square foot. I will do infrared imaging after we get some climate control going to see check how well-distributed it all is.
We made a choice to cover the interior walls with plywood. I am going to say up front this was not an inexpensive decision. I bought a large quantity of a decent formaldehyde free 1/2″ maple plywood, and got a great price and even so it probably cost 3 times what drywall would have been. Though it did relieve the worry of having to pick paint colors.
Cabinet plywood comes a half-inch oversize in both dimensions, so our first step is to true the sides and cut them to an accurate 48″x96″. This is fast and precise with our track guided circular saw and a jig we make to dimension the sheets. Creating very precise, square sheets lets us butt the boards to one another with virtually no gap between them, making for a largely seamless installation.
All the panels are face screwed with trim screws. Concerned about the potential of cellulose dust filtering through between the panels, we put a small bead of clear caulk along the edge of each sheet before placing the next sheet. With all the plywood in and the detailing around windows and doors finished we are moving on to cabinetry and storage.
And tomorrow the solar goes in!
As I stated in an earlier post, maybe the first, one of the drivers for the design is to end up with a house that doesn’t require the standard maintenance of most houses. No painting, no cleaning gutters, I guess we will clean windows from time to time, but since they all open in that ought to be easy.
Obviously the siding plays a key role in achieving zero maintenance. For this to work the exterior cladding has to cover and protect the house and be a material that can stay outside for a long time without deterioration. A long time meaning fifty years or more. That is a lot to ask of a building material. Early on we decided against brick, stone, or concrete, three things that can support and protect buildings for hundreds of years, pretty much leaving us with metal or durable wood for the exterior. We chose both. Sheets of a weathering steel alloy for the bottom 8 feet, and durable wood, in our case white oak, for the rest of the wall surface.
With the windows and doors fully installed and weatherproofed our next step was to fabricate and install metal trim pieces to provide a transition between the flat metal panels and the exterior surface of the window frames. After that we installed a drip flashing at the base of the wall to provide a clean termination at the bottom of the metal panels. The metal panels are screwed directly to the battens with #10 pan head screws. We installed the panels with an 1/8″ gap on all sides and at corners to provide more flexibility and since the steel is pretty thin (22 gauge) we tried to stretch the panels to provide tension in the interest of keeping them flat across the spaces between the battens. This was generally slow and hot work, we started using an abrasive blade on a circular saw to cut the panels, but moved to electric shears by the end. The shears produced a cut nearly as straight without the sparks and noise of the abrasive blade.
Everyone who touched one of those panels would have preferred to have been doing it in some other month than July. The material sitting in the sun would heat up to the point you couldn’t handle it without gloves and the sweat marks from forearms pressed against the sheets when they were installed rusted quickly to a deep purple. But as hot as the metal got it wasn’t anything like burning the wood siding in mid-summer under the relentless sun we had this July.
The treatment of the wood for the siding is something I haven’t ever done and I was still a little unsure of. I am not going to say skeptical because it made sense to me, charcoal is a very durable material, and the wood we planned on charring is a naturally weather and rot resistant wood , white oak. Still the extent of my knowledge of the process was pretty much limited to online video; that should give anyone pause. I bought a weed burning torch, got three 20 lb propane tanks filled and layed some concrete block on a barren patch of dirt and started burning wood. When we started it was hot even for July, highs approaching 100, and the spot available for the burning was on the western side of the house, prompting comments about what the heat would be like using the torch with the afternoon sun overhead. I ended up doing most of the burning and I have learned that the torch is so intensely hot that I honestly forgot about the afternoon sun, so hot that the guys installing the siding 10 feet above me could feel the heat where they were working when the wind took it their way, so hot that when the wind shifted it would singe the hair on my lower legs.
My goal was an even and complete blackness, but without much alligator skin from coals forming on the surface of the wood. This was relatively easy to accomplish; if we had wanted a lesser and consistent amount of charring it would have been hard to get at the rate I was burning the wood. Completely burned is pretty easy with the amount of gas coming out of the torch, half-burned would be hard to control.
To move systematically through our large pile of white oak, I would put three pieces of wood on the concrete blocks about three inches apart. Burning the center piece from directly overhead with an aggressive flame would char the top of the center board and the adjacent edge of the outside boards. A pass down each side board at an angle would char the top and remaining edges; it wasn’t perfect but pretty efficient with a little touch up here and there we got three charred sides on each board. Before starting I figured I could burn 3 boards in 3 minutes, the actual rate was more like three boards in 10 minutes; in the end we used 180 pounds of propane. I have done a careful analysis of the time differential to a more traditional method for weather protection, say stain or paint, but my gut says this wasn’t much different, and if it doesn’t need additional maintenance in the future, the life cycle costs will be much less. You have to like a black house though.
We have been lucky recently here in Athens that a nearby specialty wood supplier has started to buy local logs, saw, dry and mill them and offer hardwood flooring and dimensional hardwood lumber that are locally sourced. Much of this comes from residential trees that tend to be knotty. I have used the white oak flooring in projects for clients in the past and got the same company to provide the oak for our siding. The knots are generally cut out for the flooring and for our siding, since it was all burned, didn’t matter much. An unexpected happiness: the charring of the surface transformed the grain in the wood to a contrasting iridescence in tones of black, the plain sawn boards with the typical wood grain pattern and the quartersawn pieces with the rays still visible wriggling across the surface.
To add visual interest and texture I decided to use two different width siding boards, a 1×6 and a 1×3. We installed in a pattern of two 1×3’s topped by one 1×6. All the boards are screwed to the battens with exterior trim screws. Even though we rinsed the boards with a hose after the charring process, the charcoal would readily rub off when touched. At the end of the day the guys installing the siding would look like they had climbed out of a coal mine.
For most of the house the charred siding is out of reach so there isn’t any danger of smudging against the charcoal, on the front porch, though, the wood is the main wall surface and to avoid rubbing up against charcoal I had planned on staining the porch siding black instead of burning it, but after putting the wood on the wall it seemed too beautiful to cover, so within the screened porch we are now going to give it some minimal finish to provide a little protection from the elements but keep the natural wood.
I am trying to figure out what to say about the windows and doors. It should be easy, they are like nothing I have used before. Triple glazed European tilt/turns, the small ones are startlingly heavy, the large ones are so heavy that moving them requires much conversation and several people to make sure no one is hurt and nothing is broken. They close and seal so completely they feel like a door on an old refrigerator, the ones that still had latches. The doors close effortlessly and with the multipoint locks engaged are secure and airtight. I guess that’s about it, they work, well, and are pleasant to use.
There are a couple of places in the house where the operation of the windows is a little awkward. Every window opens and since they swing in an open window can stick into the room a bit. The largest windows are in the bays and open against the side walls of the bay where they are neatly tucked out-of-the-way. For the rest of them, if they are going to be in the way, we can just tilt them instead of swinging them in. This function allows the window to be hinged at the bottom and open inward at the top about 6″. So the window can vent without encroaching into the space. It has been brutally hot recently and every window in the house is open to catch whatever breeze is available and people in the house have automatically gone to using the tilt for windows that would be in the way if they were swung open.
After a very long wait the roofing arrived. A huge relief and a necessary step to move forward on interior finishing. The roof is on two different levels, the upper roof spills onto the lower roof. The upper roof will be covered in concrete pavers, the lower roof will be planted as a roof garden, but beneath both is an 80 mil TPO membrane. The roof deck framing is level, the pitch for the roof membrane is created with tapered pieces of foam screwed to the roof deck. 80 mil TPO sheets, a plastic roof membrane, are fastened through the foam to the roof deck and all of the seams are heat welded to form a continuous drainage plane.
And now with that done water no longer comes into the house, I sleep at night when it is raining, and we can start finishing the inside and building the roof garden.
We have had a couple different ideas for the second level floor. Back when one of the goals for the house involved a budget, I had conceived of the second level floor as sanded plywood. Mary was good with that, but offered up concrete as a possibility, briefly we considered cork. In the end what settled out was concrete, however using concrete as the finish floor surface on the second level of a wood frame structure is not the same dual purpose money-saving choice it is on the foundation level. On the other hand we do have significant southern glazing on the second level and the 2″ of concrete provides a nice heat sink for the solar gain of bright winter days. Additionally, and this is a benefit I didn’t anticipate, the concrete blocks a lot of the sound transfer through the floor. With the joists open below had the floor only been the plywood, every footstep, the murmur of conversations, and just about any sound would have transferred readily through the floor. Very little though is clearly audible directly through the 2″ of concrete, and it does look good.
We put 15 pound felt on the plywood, and formed around the stairwell so the concrete edge would serve as the finished nosing at the floor edge. A 2″ strip of 1/2″ plywood was nailed around the perimeter as a form for the concrete at the walls. Getting the concrete up to the second level required a conveyor truck and a careful operator to direct the belt through the porch door, from there we used wheelbarrows to distribute it throughout the floor. Not so bad for only 2 yards on a level surface. I chose to use a lightweight aggregate in the mix to reduce the load by about 20%. It is the first time I have ever used this type of mix and it was an odd thing. The stone is light enough that it floats in the concrete, a property that confounds much of the finishing process. I used a vibrator to consolidate the edges at the stairwell, but agitating the concrete would cause the stone to rise instead of lock into the other concrete components. In the end it proved difficult to get the surface flat, but the finishers worked hard to keep it looking good.
While we were lifting concrete to the second level we poured the screened porch floor as well. This is something I really like. The porch floor is supported at two places, it rests on a steel angle bolted to the wall of the house and 6 feet from the house the porch slab passes over a 4×8 steel tube and then extends a bit more than a foot further. Other than that it is a self-supporting 4″ slab, 5/8″ rebar placed on a one foot grid provides the reenforcement.
In the picture above the side forms are not in place yet. The plywood bottom is 3/4″ MDO, a concrete forming plywood with a smooth skin on one side to provide a smooth finished concrete surface. On the top of the beam on the left the steel headed studs that will lock the slab into the beam are clearly visible. There are also studs welded to the steel angle on the house to lock the slab to the angle.
Under the plywood there was an extensive forest of support to keep everything in place untill the concrete was cured enough to support itself. When the concrete was placed we made a concerted effort to vibrate every surface (since they all would be removed and expose the concrete on the other side) to remove any air bubble against the forms. After a week we pulled the forms saw how we did.
My daughters are home for the summer and Lois had recently been doing wood block carving in a printmaking class and had done some wood carving during a Maymester study abroad in Bali, so I asked if she would be interested in carving into the plywood form, which would produce a relief on the exposed bottom surface of the slab. She agreed, and made this:
p.s. in the photo looking down from the roof at the porch forms you can make out Lois’ carving in the lower left corner.
WIndows arrived, clients moved in, weather’s been good and we have gotten a lot done on the little house. Here is a quick summary of the latest progress. The window bucks are in (they are the rough openings within the rough openings, they bridge the space between the two walls so we can seal around the windows and keep insulation from later leaking out), interior walls are built, at least the one’s that don’t have some finish cabinet detail that integrates with them, temporary stairs are in all the way to the roof, so it isn’t such a chore or a hazard to move between floors. The loft is framed, but due to its interaction with the finished walls the deck on the loft is partial and temporary. Most of the housewrap it on, we are still waiting on the roofers to put the membrane in before we finish the wrap. The grading around the house is done, making the future work of siding easier and safer, and opening up the exterior site work like driveway, walkways, bottle wall and pizza oven.
The bottle wall and pizza oven are just teasers, we’ll talk about them later. Since I wrote that first paragraph the weather has changed for the wet, and really what I feel like I have been spending my time doing is sweeping the water off the plywood roof deck after each rain and shopping for obscure plumbing parts and energy-efficient appliances and fixtures. The plumbing parts and efficient appliances are actually related, and I will get to that, the sweeping of the roof is just my 15 minutes of anti Zen, where I scream in my head about roofing material that is apparently lost in Utah.
The master bathroom in the house we left last year had a bidet. If we were to compile a list of the things that have been missed living in the rental and especially those that have garnered comment the bidet would be very high. The small house doesn’t have space for a free-standing bidet, leaving a bidet toilet seat as the only option. Bidets that are added to toilets have a wide range. The simplest are essentially kitchen sprayers tapped to the supply for the toilet that one uses manually to wash, with cold water. At the other end are add-on bidet toilet seats that cost over a thousand dollars and have remote controls to direct such things as the opening and closing of the lid, the desired style and temperature of spray, the temperature of the seat and a fan to blow warm air to dry one’s self. I am sure it is amazing, and a level of luxury that would make me feel both sublime and sheepish at the same time, but the sticking point for me with the high-end bidet seats was that the warm water for the wash is generated in the seat in a little boiler, a little on demand electric water heater dedicated to washing your bottom. Since the design for the bathroom where the bidet top is going would have me sitting literally 18″ from 100 gallons of water heated by the sun the idea of heating cold water with a 1800 watt boiler was unconscionable. Simpler bidet seats are available, non-electric and typically cold water only. Cold water only, though, was, for some in the household as much a non-starter as little boilers. The solution came with a mixing valve placed in the wall near the toilet. Hence the search for obscure plumbing parts. Hot water from the water heater, is mixed with cold to a temperature readable in centigrade (by us it will be done by feel, how hot is 45 deg celsius anyway?). The top is made in Italy, but available (kind of) in the US, is reasonably attractive, and best of all doesn’t require an electrical rough in for the toilet (how did we get there, unless, of course for composting toilets).
This is the point in the job where any decisions about finishing items that haven’t been made need to be thought about seriously. Being contractor, designer, and client gives us a little leeway, shortening the communication loop considerably. But even with these added advantages, now is the time so we have been busy finding appliances and fixtures and tile and other stuff. With one of our goals being achieving net zero, and the efficiency through insulation and other building envelope components set, my focus has shifted to the items that use electricity in the building. Here is an embarrassing admission, I have bought a lot of things for a lot of houses and as much as I have considered the efficiency of the structure itself I haven’t paid much attention to the efficiency of what goes in the house. The rubric on this project though has forced me to consider every use of power, not just the big things that are built into the house, but the many things that plug into the wall or hang from the ceiling and use power day in and day out. Things like bidet seats and refrigerators.
Early on in the design our strategy for refrigeration was a fridge only upright placed in the kitchen like any standard fridge and a chest freezer hiding under the island to provide freezer space that would be accessed by rolling the freezer out and sliding it back when we were done. When I type this out it doesn’t sound like a great plan, but it has its merits. Using standard appliances the combination of a 20 cubic foot fridge only and a 7 cubic foot freezer provides a lot of cold storage, especially freezer space, with an annual energy consumption of around 500 kWh. Compare that to an efficient 25 cubic foot side by side which uses about 650. If you substitute the standard fridge for a super efficient one like a Sunfrost, you can get the total consumption down to about 400 kWh, but that combo costs about $2500 more, ouch. For that same 400 kWh of energy a 21 cubic foot fridge with freezer will provide a little less storage for about $900. No ice maker, no through the door water. After considering all these different options we chose the 21 cubic foot fridge with a top freezer, adding to the energy and cost savings of that choice, eliminating the chest freezer under the island allows us to wring a little more storage out of the kitchen.
Why are we are quibbling about 100 kWh of electricity, which costs about $12? Because our goal is to balance our electrical generation with our consumption and the small PV array that will fit on the roof will only give us about 3000 kWh per year. So 100 kWh saved is roughly 3%, but going through the house item by item the difference between an efficient appliance or fixture and less efficient ones really starts to add up. 100 kWh for the fridge, 100 for three ceiling fans, 400 for a dryer (in this case a heat pump condensing one), bath fans, a TV, vent hood, washing machine (where you have the opportunity to choose to save water as well). It is pretty easy to shave 1000 kWh a year off the household energy use, and in most cases it doesn’t cost any more. The heat pump dryer is an exception, that costs more, but like heat pump water heaters the jump in efficiency is so huge it is hard to justify not doing it. The Energy Star website makes quick work of looking for efficient appliances; it’s not perfect but is a good place to start, and it has listed some things that I would never have found otherwise.
For us the 1000 kWh saved might just let us squeak by in meeting our goal, but 1000 kWh saved for say, a million US households, would shift the arrow in the right direction to meet goals way more important than our little flight of fancy.
After a second round of studs, hundreds of gussets, several rows of blocking, and the completion of the roof dormer, we are ready to put on the sheathing, and the house will go from a skeleton to having skin. The sheathing for this house will be 1/2″ CDX plywood and will serve multiple roles. The plywood skin braces the house against lateral loads like wind, on this project it will act as the air barrier sealing the house against air moving between the inside and outside of the structure, and it is the substrate supporting both the weather resistive barrier (house wrap) and siding or in our case the battens that support the siding.
Before, in a different post, I discussed the let in bracing that we used on the interior walls and the reasons we chose to build those walls that way. What I haven’t mentioned is that as the house got taller and more levels were added it was apparent that there is a lot of flexibility in tall walls with let in braces. I mean a lot. Finally on the roof finishing the framing details at the top of the house a couple people moving around up there would produce enough swaying everyone anywhere in the house could feel it and it was likened to a boat at sea. Very flexible. As the plywood was nailed on, each completed wall section made the house more rigid, and when every surface was finished the motion was gone. Very rigid, just like houses ought to feel.
We started putting on the plywood at the top of the house and worked down. This allowed us to utilize the open wall studs as support for our scaffold jacks. Moving down in 8 foot increments, we could do two courses of plywood before moving the scaffold jacks. At the bottom of the wall where the wood meets the concrete foundation, our last course is a 2′ strip of pressure treated plywood as added insurance against rot for the sheathing that is closest to the ground.
One of the goals in an energy-efficient house is to minimize the transfer of air between the space inside the house and the outside. This requires a clear distinction for where inside air stops and outside air begins. I have generally used the sheathing to define the separation. I find this line to be cleaner than using the drywall, which is another option, to insure a complete barrier.
For this house, without a pitched roof, we just need to seal the plywood box composed of the sheathing and the roof decking, tied into the foundation wall, which has already been sealed, to create our air barrier for the walls. I chose a sheathing tape for the job on this project. On other houses we have used caulk and duct mastic to seal the plywood joints, and it seems to work well, but with two trips around the house the time required was unattractive. On a house previous to this one we switched and tried Siga Wigluv tape, and I have used the same tape on this project. It is very expensive tape. 70 sheets of plywood for the sheathing and about 18 for the roof deck is $700 in tape to seal the joints, but we can do it when the sheathing goes on, one trip for both nailing plywood and taping joints, which speeds things considerably. For us I think the caulk and mastic is probably cheaper, including the labor, but the lost time for the extra steps has a cost that better business people than me could likely quantify, and my confidence in the tape is much higher than the caulk and mastic. The Wigluv is crazy sticky, and flexible, lifetime of the house kind of sticky.
I plan to get the house sheathed, housewrapped and get the windows and doors installed and get a blower door test to see how we’ve done and identify places we need to seal, before the insulation and siding get in the way of corrective measures.
The front photo shows some things that depart from traditional framing practice, and are a nod to establishing a continuous air barrier. First the parapet for the front roof is already installed when this photo is taken, but the sealing tape is visible up to the parapet, but not on it. We sealed the wall sheathing to the roof deck before the parapet was built, so that the parapet is outside the air barrier where it ought to be. If we had run the exterior wall to the parapet height and sheathed continuously to the top of the parapet, air sealing the area on the wall above the roof deck would have proven to be much more complicated. Also, there will be a porch coming off the second level on the front of the house. The opening in the center of the house in the last photo is the door that enters onto that porch. We haven’t framed that porch yet in order to maintain the air barrier formed by the sheathing and tape. In fact the porch won’t be framed until the house wrap AND the battens for the rain screen are complete, allowing for the air barrier, rain barrier, and the air space behind the siding on the house to be continuous and not interrupted by the porch framing. Something we will talk about later.
The windows have arrived as well as all the products for flashing the openings. Next post I will discuss prepping for the windows and the installation, the roof is going on soon as well, but that might need a separate discussion.
Where we are now on the project, with the roof framing on and decked, most houses would be ready for sheathing or likely sheathing would be finished and the framing pretty much done, but for us there is a second wall to build outside the first one.
It is always good to get houses dried in and safe from weather in as short a period of time as possible. With less swelling and movement, and less effort to dry the framing after the house is enclosed, houses built with less exposure to rain are favorable. Sometimes the weather and scheduling cooperate for this to happen, this spring has not been a cooperative time. So when we start on the second wall when we could be getting the house in the dry and 3 more inches of rain fall over a weekend my sense of urgency heightens, and I start to regret the complexities of my little house. In a month, typically one of the dryest, where twenty days in we’ve gotten about seven inches of rain, a month that averages three, I force myself to remember the house will be covered soon and it will dry out and the second wall is integral to the whole concept, lots of insulation, little foam. Here’s a spoiler alert, building two walls is about twice as much work as building one. Maybe more.
The whole purpose of building two walls is to fill the space between them with insulation. In our case it will be twelve inches of cellulose giving us a wall with an R-value somewhere in the low 40’s, but super insulated houses are put together in a variety of ways. A standard stud wall covered in a bunch of foam can yield walls with R values approaching 60, but the building has to be covered with 8 inches of foam to see performance like that, and I have already written about why that approach wouldn’t work for us, mainly we don’t like foam. Another method also starts with a traditional wood frame wall and after sheathing another frame is added. I have done this with a 2×6 wall attached to the sheathing of a 2×6 wall, an even more insulated version attaches a 12 or 14 inch I joist to the sheathing of a framed wall. I can see the value in that approach, especially in time savings, but we would have had to sheath both walls, which I think would cost as much as the time saved, and I joists aren’t cheap. The method we are using does two things that are important when it comes to insulation, first it provides a lot of space. Since the insulation performs generally in a linear relationship to its thickness, more is better. Doubling the R-value of a wall reduces the energy loss by half, with caveat that the rest of the components of the building envelope, namely windows and doors, need to keep pace. Secondly, it separates the interior surface from the exterior. With the two walls, a continuous blanket of insulation rests between the interior framing and the exterior framing. This differs from a standard single frame wall where the stud touches both the interior surface (usually drywall) and the exterior surface (usually osb sheathing) and bleeds energy from one to the other, because wood is not nearly as good an insulator as, well, insulation.
On our site we were severely limited in the footprint available for building. Building to the full extent of the available space we have a structure that is 16 feet by 29ish. The twelve inches of wall thickness versus say a more typical 6 inch wall represents more than 100 square feet lost to insulation. I only say this because for us it was a balance between sufficient insulation without losing too much floor space. Going through the trouble and expense of building two walls, it is likely better to get more insulation out of the prospect; expanding to 16 or 18 inches of thickness would involve the cost of some more insulation and some exterior surface, small expenses relative to the second wall. Again make sure the windows keep up.
Our walls are built with an inner load bearing wall that uses 2×4 studs on the lower floor and 2×6 studs on the upper floor (because of the wall height). The outer wall is 2×4 studs. Because it sits on the foundation wall the outer walls starts about 3 feet higher than the inner wall. The two walls are staggered with the inner wall breaking at the floor system and the outside wall extending 5 feet or so past that point breaking at a plate and then a second level of studs extending to the roof deck.
The roof deck ties the outer wall back to the structure at the very top of the building and of course it is tied to the plate on top of the foundation wall. Between those two points we used a lot of 3/4″ plywood gussets spaced 2 foot on center vertically to tie the two walls to one another.
There are a couple nice things about the two wall system. First only one side of each wall receives a finished surface, so if there is a width discrepancy between a stud and the plate they just have to be flush on the finished side, the other side will only see insulation. Similarly when we needed to switch to 2×6 for the upper inside wall we attached an extra band to the floor system and extended the additional stud width into the insulation cavity. We didn’t need to use 2×6 on the lower wall just because the upper wall needed them. Also the walls could be plumbed and straightened independently. So small errors through the floor system, for example, didn’t need to transfer through to the outer wall, we just set the plate in the middle of the outer wall to a string and tied it back to the inner wall that had been plumbed and braced, doing this again at the mid-point of the studs we were able to achieve a very flat outer wall surface.
To tie the outer wall to the inner wall we used 3″ x 11″ x 3/4″ plywood gussets nailed to aligned studs in each wall. There are a lot of gussets. We had form material from the foundation walls we were able to get the gussets from, 800 in all.
With the outer wall completed (except for some blocking) our next framing task is the roof dormer for the stairway access to the roof and the sheathing. The forecast calls for few dry days. Sometime next week we ought to be ready for some roofing, finally a dry house!
With the second level walls framed, plumbed and braced it is time to frame the roof or actually roofs. On top of the two different height walls are two separate roofs, framed with different materials. The lower roof will be the planted roof, built up with multiple layers of waterproofing and drainage material, topped with 12″ of soil. The upper roof is the roof that supports the dormer that covers the roof access stairs, with the remaining area, about 8′ x 14′ serving as a patio. The different uses and associated loads of each roof dictated the size of the framing material needed. In the case of the planted roof, covered in saturated soil, the loads required a 16″ I joist, for the other are, just serving as a patio, only a 12″ I joist was needed. The deeper the joist, the more insulation that can be packed in, which is a good thing, and I would have prefered to use 16″ joists on both roofs, but that additional 4″ would have put the rear section of the house beyond the maximum allowable building height. So we have to be satisfied with an R value of about 50 including the foam tapers on the top of the deck for that roof, where the planted roof will be more like R70.
Since the bay windows project from the building without bearing directly on the foundation, their weight being carried by cantilevered flitch plates at their base, they are not able to carry the loads from the planted roof. So to support the load from the portion of the roof that sits above the bay window we employed double 16″ LVL’s cantilevered over the bay. The double LVL’s served as the rim along the entire front and southeast side of the roof.
The beams join at a corner of the building that is suggested by the rest of the walls, but not realized until the roof comes together over the bay window. This creates an interesting play of triangles: the two corners of the window projecting from the main building and the roof projecting over the center of the window like a ship’s prow. This will become clearer when the second outer wall increases all three projecting points and even more so when the sheathing solidifies the mass of the building.
We also needed a set of double LVL’s across the transition between the lower roof and the upper roof. Because of the orientation of the dormer for the roof stairs, the joists for the upper roof had to run back to front. This necessitated a beam across the building at the roof transition. The lower roof joists don’t bear on this beam, they run side to side with both ends landing on a wall, with the exception of the joists over the bay that hang from the previously mentioned cantilevered beam.
It is easy to write about the process of building the roofs, placing framing members and revealing the puzzle pieces that will go together to make the whole. In the end this is what building a house is, a series of steps and layers that culminate in a place where people live. It is also easy to get lost in those steps, to substitute process for purpose. In our case I think it is worthwhile to discuss the purpose for the planted roof.
Not everyone loves green roofs, the value that they add to a sustainable design is debated. Engineers tend to point out that the benefits they provide are available through other means. Their contribution to reducing urban heat islands can be accomplished with high reflectivity roof materials. Their mitigation of storm water run off can be handled through infiltration systems on the ground. Because dirt is not a very good insulator, planted roofs don’t really add to the energy efficiency of buildings, except that the plants on the roof, supplied with sufficient water, will provide evaporative cooling through transpiration. Better and cheaper to increase the insulation on the roof to improve the efficiency of the building, and insulation will perform during both the cooling and heating seasons. And using potable water to irrigate plants on the ground is wasteful enough without adding irrigation to the roof.
I agree with all that I just wrote, and there are good studies to back it up, but there are a couple of reasons, general to building and specific to this project, why we decided a planted roof was purposeful and included it in this house.
It is hard to assess how much of building is directed toward managing water, water from the sky, water from the ground, water in the air. Maybe it is just in the Southeast, but I don’t think so and here we are mostly spared the problem of frozen water, thank God. Anyway it is a lot. The conventional wisdom and vernacular architecture in the South embraces steeply pitched roofs to shed rain water quickly off the house. I have found, though, over many years of repairing houses it seems more damage is done by the rain water after it leaves the roof than from the time it spends on the roof. Overflowing gutters, splash from the eaves, flooding foundations, all lead to problems for houses. When I built my first flat-roofed house I never noticed the water that left the roof, it was collected behind a parapet, drained into a downspout, then through underground pipes and finally flowed onto the ground forty feet from the house. The roofing material was a heat sealed membrane that was user-friendly and durable. The one drawback was that the leaves from nearby trees tended to collect at the screens for the roof drains, but at least when you have to clean them you are standing on a (nearly) level surface. So the only thing better: a flat roof where the leaves don’t matter. A filter of soil and the other layers of the planted roof negate the ability of accumulated leaves to clog the outflow for the roof. As opposed to a traditional roof that encourages the rain water to flow off the roof as quickly and directly as possible, dragging with it any debris that has landed on the roof, the planted roof slows the rain, absorbs and retains some of it, and filters it through a broad area before releasing it slowly to the roofing to convey it away from the building. It may seem counter-intuitive, but part of my no maintenance scheme included the planted roof. I was going to garden somewhere, why not accomplish roof maintenance at the same time I was planting or weeding.
My musings on water management made the idea attractive, but the reason for the planted roof is the desire for garden space on a piece of land so small its hard to escape our own shadow. We have always had gardens, frequently unruly, sometimes very productive, sometimes less so, a garden is part of who we are and I have nearly arrived at the place that I believe if land isn’t going to be left to its own and return to what here would be a subtropical deciduous jungle, it ought to grow food. The house dominates the lot, add a parking space, walkways, space for our aging dog, consider the north side lost in the shade of the house, bamboo and a struggling pecan tree on the adjacent property shade some of the narrow strip to the south; even so we will plant the front hill with blueberries and strawberries, we will put some blackberries on the fence line. But the roof offered an unobstructed view of the sun, and though small, plenty of space for a generous kitchen garden. The roof is where we will have our cooking herbs and spring lettuce. Every garden we have ever had has been an evolutionary project and this will be as well. I see some years with sweet potato vines cascading down the walls, while beans grow on a trellis above them, maybe our virgin roof top soil will even produce a zucchini before the squash borers kill the vine.
One drawback to planted roofs is that they almost certainly need irrigation. Maybe not always and not as much with the carpet of sedums that often make up green roofs, but with plants like tomatoes that are thirsty in the ground, watering will be a necessity on the roof. I didn’t want to take an idea that is borderline sustainable and throw it clearly on the wrong side by watering our roof with potable water. To avoid this we needed to get our irrigation water from somewhere other than the tap. My first choice is stored rainwater, and we have some roof area that will not be planted that can be collected during rain storms and used during the hottest months. Several hundred gallons of stored water can last a few weeks without rain as long as the irrigation is efficient and there is plenty of mulch. We will have a 300 gallon storage tank under the porch with a pump to lift the water to the roof, and a subsurface irrigation system to put the water where it is most needed. The water keeps the plants alive and makes the garden viable, but during the hottest months it also provides the one energy benefit of a planted roof, evaporative cooling.
We did get the roof framing done and moved on to building the outside wall, things are wet now, but next week looks clear. A good week’s work and we’ll be close to the end of the framing
With the floor system finished and the subfloor (and finished ceiling) installed it is time to make this thing look like a house, or at least look like the house that it is.
As I have said earlier, raising walls is one of the absolute joys of construction. Tilting a wall up, looking at the top plate overhead and a cloudless blue sky beyond, is a singular moment, emblematic of the entire effort, and makes me feel that I am lucky to have a job where I get to work outside. And makes me forget about carrying buckets of gravel in the rain a month before.
Our plan is to get the second floor interior walls up and straightened before working on the outer walls. There are two different wall heights on the second level, 10′ 4 1/2″ in the front and 14′ 10 1/2″ in the rear. This is for a couple of reasons. First the higher wall in the back is needed to allow for the loft. With 7′ being the minimum ceiling height allowed, 14′ 10 1/2″ gives us enough for two levels plus a 2×6 floor system and a little extra for ceiling covering and such. Because the planted roof is on the front section and that overall roof package, framing, insulation, and planted roof paraphernalia, is 30″ thick, the wall height below it needed to be lower. The stairs from the loft to the roof also open on to the front section so having it lower than the back requires fewer steps to get up onto the roof.
Here is a section showing the floor levels, roof levels, and stairs between them.
There was one difficulty in building the tallest walls on the deck and raising them into place, the walls are about as tall as the building is wide. And once one side was in place there wasn’t enough room for the other side to be laid flat. The last of the tall walls we framed were at least not long so we built them where we could, stood them up, and wrestled them into place.
We worked last Saturday because the weather was good and we hadn’t gotten much done on site for much of the week. I worked the full day, everybody else split half shifts to preserve some of their weekend. In the afternoon Woody came in after we had gotten the rest of the second level walls up and started to plumb corners in the AM. He and I finished plumbing corners and framed in window rough openings. At some point during the day, in the course of conversation, he pointed out that one of the things he liked about the house was that though it is a fully modern design it uses old framing methods to get there. It was something I hadn’t thought about, but he is right, let in 1×4 bracing disappeared with the advent of plywood, and balloon framing with fire codes and pre cut framing and small trees, and we’re using both in this house, and I don’t think I have ever used either method before, mostly because I’m not old enough.
I have built another double stud wall house, but that one was sheathed twice, once to brace the walls and then again to hold the siding. I didn’t want to do that and so needed another way to brace the interior wall and 1×4 braces let in to the studs satisfied. We can insulate around them so the stud cavity between the two walls is uninterrupted, they brace the walls against lateral loads, and they use a whole lot less material than plywood sheathing. And we are going to fully sheath the exterior wall with 1/2″ CDX plywood, mostly to hold siding, but also the exterior wall has a full foundation under it and though it won’t carry the bearing loads of the building, it will be subjected to other forces on the building like wind loads and it will act in concert with the interior wall (because they are fully tied together) to handle these loads.
The balloon framing is really only used on the loft framing, though the double wall acts a little like a balloon framed wall. For the loft floor we are going to use a let in 1×4 on the back wall of the house that the loft floor joists will rest on. The joists will also be nailed into an adjacent stud. Balloon framing multi story houses disappeared from common building practice long before framers quit using let in bracing, I guess because the studs had to be so long. It made sense for the loft floor because it simplified the framing of the back wall by allowing the entire rear section of the house to be framed to the same height, and again serves to allow the insulation to flow fully through the framing cavity. The main floor system was platform framed and we placed foam blocks against the band joist so the amount of insulation wouldn’t lessen at the floor system. That won’t be needed at the loft floor which won’t create an obstruction within the wall.
Balloon framing does create the need to address a common problem of the past, the potential for a fire within the wall to jump floors because the framing doesn’t create a full barrier between the floors. This is the aspect of the double stud wall that behaves like a balloon framed wall. The goal of the double stud construction is to provide a continuous insulation layer, with no bridges of framing for heat to move through. In order to achieve this there is no framed fire block between the floors, a block that would also be a thermal bridge. Instead the wall cavity is completely filled with dense pack cellulose, which in addition to the many other wonders of cellulose serves as a fire block, partly due to the fire retardants in the insulation (borates mostly, not like fire retardants in pillows and furniture) and also due to the lack of air flow through the material.
So fully modern, yes, but leaning on older techniques that were properly suited to accomplishing our overall goal
small house small lot 