January, 2006 (updated May 2007)

We're coming up on 2 years in the house now, so we finally have some performance data.  The major glitches in our data is that we didn't get our PV till July 2005, and for various reasons our rainwater system has only worked for about eight months. We've had a few hundred people tour the house, and won a couple of awards, but more importantly we love the house.  Of course, it's not perfect, and so here we talk about the things we like and the things we think we could have done better.

Tours

We've hosted at least five major tours of 50+ people, and numerous other smaller tours.  While we are committed to giving them, they are also exhausting.  We're currently debating how long we'll keep doing them.

Layout

In general, the house works for us.  We both love having separate offices, and use them frequently. While we used to fight over the couch a lot, we now both use the window seat far more often, especially when its sunny.  Although the dining area is small, we had eleven people around the table for a snack and a roundtable discussion and it worked fine.  Having a small bedroom hasn't bothered either of us in the least.  Bob hasn't been able to use his art studio (due to work and the house), but is anticipating using it soon.  Initially we over used the mini-office, but once bob moved his email to his laptop, this problem went away. The mail sorting area does help with paper clutter, but it doesn't solve it completely, because we didn't give ourselves enough space for "projects".   We've struggled a bit with the  "living room" part of the great room, and finally (Dec 2005) found a furniture arrangement we think we like.

Energy

Our gas bills are quite a bit lower than projected, although our electric use is not as low as expected.  In July 2005 we added a 3KW PV system (the PV compatible meter came in l Aug) and upgraded a significant amount of our lighting to CFLs.  Here is our gas/electric usage to date:

Our predicted heat loss was 9854.7 BTU/deg/day, or about 44MBTU for a 4500 degree day year, of which our guesstimate was that 25% of that would come from solar, so we'd need about 33MBTU of supplemental heat.  Our actual use of 25.7M is much lower than that, but if you add back in the internal gain from electric usage during the "heating months" (guesstimate of 8MBTU), then we match our predicted loss almost exactly.  Our gas usage also includes hot water, two dryers, and a stove, but since we get about half our hot water from solar, none of these items are likely to impact our gas energy usage dramatically...my guess would be that it increases our gas usage by 5-10%.

Another way of looking at this is in energy used per square foot of heated space, or energy use per finished space.  Our heated space is about 2900SF (about 2500 finished, and 400 of storage), so our average yearly gas usage is between 9207 and 10680 BTU/SF/year, and total energy use is between 12,150 and 14094BTU/SF/Year.  (the lower numbers are based on 2900SF, and the higher ones on 2500SF.  For gas usage, I averaged both years, but for electric I just used 2006, since its the only whole year with PV installed).

The best baseline comparison is from the dept of energy website http://buildingsdatabook.eren.doe.gov/docs/1.2.10.pdf , which lists single family residential houses as using about 44,800 BTU/SF/year.  I also found alternative values for new houses down near 30,000BTU/SF/Year, which is probably a more valid comparison.  Just based on straight utility averages for comparable homes, Seattle city light says that the average usage is 9200Kwh a year, and Puget Sound Energy says the average is 700 Therms(70MBtu), although I have no idea what the average house size those numbers represent.  In my view these comparisons aren't  really apples-apples because electric use is reasonably dependent on the number of occupants, plus our house includes a rental unit.  Ignoring these issues, we use at most only half the energy of the average house, and possibly as little as 1/3.

Yet another measure I found is from www.2010imperative.org, a site dedicated to reducing energy use in buildings.  They use essentially the same baseline usage (44,700BTU/SF/year), and say we need to cut that number in half immediately, and by 90% by 2025.  At 14KBTU/SF/year I'm  already about 70% lower, but further lowering will not be easy (click here to read about the difficulty of building zero energy/zero carbon homes)

For electric comparison, I ran the city's resource calculator: http://www.seattle.gov/conserve/homeprofile.  That showed by data for Jun 21, 2004 thru Jun 21,2005 (pre PV install), so its a good comparison of our actual use.  In that period, we used 17Kwh/day and the average comparable house used 25Kwh (single family with no electric Heat or H/W).  Since we have an ADU, our comparable usage is pretty good, although a more fair comparison would probably be based on number of occupants.  During the last year, we converted a significant portion of our lighting to CFLs, and upgraded one computer.  Unfortunately the hibernate feature on the new computer works only when it feels like it, and the hibernate feature on the computer in the ADU doesn't work at all.  In addition the refrigerator in the ADU uses 2.2Kwh/day, which could be reduced to 1.1Kwh/day by buying a new one.

Update: Jan 2007: we've updated the computer in the ADU (hibernate works now), and installed a couple of plug strips on the tv/dvd to eliminate the phantom load.  We now have a full year of solar data, but because we're away a lot, its not clear how to year by year comparisons. 

On a more subjective basis, we find this is the most comfortable house we've ever lived in: there are no perceptible drafts, and when the thermometer says 70, it feels relatively warm even when its fairly cold out.  The one exception is in sitting by windows, which alas still have relatively cold surface temperature.  One other thing, is that the upper floors are no warmer than the lower floors.  On cool sunny days, the main floor gets a bit warmer than any other floors because it has more window area per square foot of floor.  In early 2005, we had unusually long sunny spell, and as a result we were often able to go days with little or no supplemental heat.

One positive surprise is how well the house performs on hot days.  Even when the temperature was in the 90s out, the house never got warmer than 76, provided we kept the windows open all night and closed them by 10am or so.

HVAC

The main annoying downside to our HVAC system is that it only works with very hot water, so we have to keep the tank at 140 degrees.  To make matters worse, we can't turn the tank temperature down because plumbing code (and possibly sensibility) require a tempering valve if the tank is above 120 degrees.  The problem here is that tempering valves lower the hot water temperature by at least 20 degrees, so if you try to turn the tank down to 120 in the summer, the tap hot water is only lukewarm.  This problem is further exaggerated by the fact that our "utility closet" in the basement is the only part of our house that isn't super-insulated. The result of this is that our standby losses are probably greater than they should be, although we don't really know what they are.

One side effect of the fan/coil heating system is that it delivers heat very slowly, even with the tank set at 140F.   Our tank will go up to 160F, but that would also increase our standby losses, so we haven't tried it.

In retrospect, I wish we'd found a cost effective way to do radiant heat because it can transfer the same amount of heat with lower temperature water. While radiant heat is typically installed in either slab-on-grade or in a thin slab, I'm not convinced this is a good idea, because it means the temperature of the house can take a very long time to change.  The idea of the slab is to distribute the heat better, but I could find no evidence that its really any better than using aluminum heat distributors under the floor.  In my view, the conventional wisdom about radiant doesn't apply to a super-insulated house: (1) you're not likely to save energy (2) most surfaces are already near room temperature and the floor doesn't need to deliver as much heat so there really isn't a lot of increased comfort (3) there is little heat stratification in a super-insulated house.  If I had radiant heat, I could downsized my duct dramatically since ventilation needs much less air than heating does.  My preliminary calculations are that 120F water would be more than sufficient to heat the house even on the coldest day.

Solar H/W

The thermomax system works like a charm. In the summer the pre-heat tank runs between 110 and 160 degrees, in the winter it often hovers around 60 and in the spring/fall its usually between 80 and 120.  If we didn't have to keep our main tank at 140, we'd get a higher percentage of our H/W from solar.

Tankless H/W

I'm still waiting for a tankless hot water unit that will serve a whole house, has no pilot, is closed combustion (those are available) and also will work with pre-heated water from a solar collector (at the same time, it ought to deliver hot water at less than 1 gal/minute).  The problem with all of them is that they have no equivalent to "simmer" on the stove...even on "low" they put out a lot of heat.  Apparently the solution to let the unit overheat the water and then use a tempering valve to limit the temperature.  Some places advertise that their tankless unit modulated down to zero, but as far as I can tell this is a false claim unless they modulate their burner, since the minimum burn rate on every unit I've ever seen is much too high to deliver 120 degree water when the input temperature is 100 or 110.  The lowest minimum burn rate I've seen is on a Rinnai, at 15K BTU/hr, which is still about three times too much (5k BTU/hr is more like it).  While it is theoretically possible to modulate the burner (ie turn on and off) to lower the effective output, doing this would undoubtedly wear out the burner fast, or even damage it.

The most promising solution I've seen it to use a 5gal (or so) holding tank and hook it up an on-demand unit to it driven by a pump.  A thermostat in the holding tank is used to turn on the pump, whose flow causes the tankless unit to turn itself on.  In order to be able to deal with water at 115, the tankless unit will have to be set at 140, to get it to turn on at all.  Now even a  1.5gpm pump (90gal/hr) will push enough water thru to keep the unit above its minimum burn rate.  The tankless unit will then produce 140F water which will mix with the 115F water (hopefully very well!), until the holding tank reaches 120, at which point the thermostat will turn the circulating pump off. (note: obviously you need two thermostats: one set at a "turn on" temperature, and the other at a "turn off temperature", and those have to be far enough apart to keep the two from trying to trigger at the same time, or causing the pump to go into oscillation).

This system is really effectively just a small tank unit that uses a very large burner (supplied by the tankless unit).  At least the small tank can be super-insulated, reducing the standby losses to very small.

Since my current system works, I'm not likely to try this anytime soon.

PV

We had 3KW of PV put on our roof in July 2005.  To get net metering going, you have to fill out a bunch of paperwork, and once we did,  the utility came out in Aug 2005 to give us a meter that will actually run backward.  The installed price is around $18,000, which as far as I can tell has a payback that is around the life of the system. Based on our estimate of around 3000kwh generated a year (I have only 18 months of data), and assuming its worth 10 cents/kwh (even though our summer rate is more like 4 cents/kwh), the PV generates $300/year of electric.  At that rate the payback is around 60 years.  I'm assuming here the the discount rate on my $18k investment cancels out the inflation rate for the cost of electric.  If the cost of electric goes up significantly faster than the amount I could have earned keeping the $18k invested, then the payback time will shorten.

The state of Washington passed a 15cent/kwh rebate law in Jul 2005, but due to some internal political battle, it wasn't actually implemented  till Jul 2006.  They did actually make the rebate retroactive, and as a result I got $382.50 for the PV I generated from the time I installed thru July 2006 (2550kwh).   Even with this rebate, and using 10 cents/kwh avoided cost, the payback is still 24 years--a long time.  I'm declaring myself neutral on the issue of whether subsidies are a good idea.  If I had a choice, I'd be putting funding into designing more efficient cells and increases in manufacturing efficiency.  More info can probably be found on www.solarwashington.org or at the American solar energy society's site: www.ases.org.

We now participate in Seattle City light's green power program send $72 back to the city to help fund their purchase of green power.

May, 2007: Now that the production program is in place, they'd decided the production meter must be located outside, so they can read it.  Unfortunately, we put it in the attic because it was easier, assuming I'd be reading it. It cost a bit over $700 to move it, and then I have to pay a $47 fee for a new meter.  So while someday I still expect to save money, that hasn't happened yet.  Such is the life of a pioneer.  We are still saving carbon, its just that its cost us to do it.

The other issue that has come up is the effect of shading, which reduces output dramatically more than I understood.  (I've now updated the "theory" page to explain it, click here to read about it.) We had one tree thinned in Jan 2007, and although it should reduce the problem, we probably took off too little to effect it much...or keep it from recurring next year.  I really need to top the tree somewhere near the roof line, but its not my tree.

Ventilation

Although we installed an HRV, we've never used it.  We installed a separate ventilation controller to run the HRV directly below the thermostat, but it turned out as dumb luck would have it, the heat rising off the ventilation controller's power supply was heating up the temperature senor in the thermostat because we installed the two really close and the vent for the controller lines up exactly with the inlet for the thermostat sensor.  As a result, I snapped the controller off its base to fix the problem, only to discover that the installer didn't leave any slack in the wires, so I'm going to have to break some sheetrock to fix the problem.

In the meanwhile, I looked more into ventilation requirements and bought a humidity meter to try to get any idea of whether the lack of ventilation in our very tight house was a problem.  In fact our humidity stays in the 45-50% range all the time, which is considered good.  I also tried to reconcile why commercial ventilation is set at 15CFM/person, but residential is expressed as .35 ACH, which in our house translates to around 90 CFM.   As it turns out, even the 15CFM is high if you're looking at only the Oxygen necessary for life. The addition ventilation is for diluting other pollutants.

I have no real way to measure how clean our air is, but I can say that we avoid most of the traditional source of indoor pollutants, and use both the bath fan and kitchen fan religiously.  The blower door test measured our house at .14ACH, which corresponds to about 37CFM of "natural ventilation".  I've yet to determine even approximately what conditions "natural" ventilation represents, but I'm pretty convinced that typical winter weather causes significantly higher passive ventilation than that, and as a result, I'm not convinced that mechanical ventilation beyond bath/kitchen fans is necessary in winter.  While kitchen & bath fans aren't a huge ventilation contributor, I'm guessing they still add 8-10CFM when averaged over the day.

Rainwater

Since almost no one in Seattle has done a rainwater tank (including us), this project has suffered from a lot of glitches.  Our small lot dictated that it had to be under ground (or under the house),  so we still feel that cast in place concrete was the right choice.  We use the patio on the tank surface whenever the weather allows.

Sealing it required a lot of futzing around, but like anything else one you know exactly what product to use and how to use it, everything becomes much easier.  Had we sealed it during the summer, our job would have been easier, because we had a steady stream of groundwater coming in the tank in the seam between the floor and the walls, which complicated sealing the tank significantly.

While the tank itself was ready to go when we moved in, we didn't get our permit to use it flush our toilets till many months later.  When we finally did, we promptly burned out the pump, because even though it says that its self-priming, they don't really mean that.  Once we found someone to repair it (a few more months: since there are no wells around it was hard to find someone who knew how to fix them), the system run great for about six months and then burned out again.  It also turns out that adjusting the turn on/turn off pressure switch is easy to get wrong, and when you do the pump goes into on/off oscillation.

We don't know exactly why the pump burned out, but our theory is that it was because the filter was clogged, creating too much back pressure.  We installed a "motor check" device, which was supposed to prevent this, but again, we may have had it adjusted wrong.

We rebuilt the tank inlets so that the water disperses upward to try to keep it from stirring up the silt, and we converted our fixed intake to a floating one, because the cleanest water is generally about 6" or so below the surface.  We're also going to build a dam around the hatch cover to stuff from entering that way.  I'd recommend raising the hatch cover at least 1/2" above the tank surface for this reason, even though it means you now have to worry about tripping over it. We're replacing the shallow well pump/pressure tank with a motor that is made to cycle often and has its own built in pressure switch, and replacing the filter with one that has a cartridge about three times as big as the old one.

When its all done (spring 2006), you can find the updated info here.

Plumbing

We're sold on PEX.  We get hot water everywhere in the house really fast.  We love our ASKO dishwasher, not only because its so energy & water efficient, but because its so quiet.  We like our Fisher/Paykel washer also, although it's not so quiet.  Having a washer with a spin speed makes the clothes come out much drier than a conventional washer.  Someday we'd like to get a genuine front loader, but in the mean time we're happy with what we have.

The rainwater tank lasted for most of the 2005 summer watering season, so we didn't have to use too much city water.  In spite of that our water use, remains very low.  Here is a chart of our water usage:

Our usage is nominally for four people, but we've been traveling about two months a year.  Countering this is the fact that we work at home, so we're home more than most people.  I'm  guessing that the drop in usage for the Mar 2005 billing was due to the rainwater tank working, and the rise in Sept 2005 was due to the pump breaking.  The low number for Nov 2005 is probably due to all of us being away a lot in Sept & Oct.

The low number in Feb-mar 2006, is again due to travel, and the higher number in Aug-Oct is because we ran out of water in the rainwater tank and had to hand water plants for a three weeks or so.

Materials

We've have done a lot with materials since moving in, although I've spent a lot more time researching salvage materials.  There is no longer anyone doing full deconstruction in Seattle, but there is a move to get it restarted in the spring of 2006.  For more information on a working model for full deconstruction visit the website of California based non-profit The Reuse People ( www.thereusepeople.org).

One positive piece of news is that  FSC lumber is now more readily available, with at least one supplier stocking some materials (see www.dunnlum.com). (Aside: I wrote to WA state lands commissioner Doug Sutherland that they should consider FSC certification, and the letter I got back was that they didn't think anyone wanted it or would pay the price premium for it.  Obviously Doug never actually tried buying any of it.)

Salvage materials are becoming more popular, but as stated elsewhere, there is a learning curve.

• Dimensional lumber is still generally not available, although its fairly easy to work with as long as it's been cleaned well.
• Re-milled flooring is a drop in replacement for new and is readily available.
• Re-used flooring needs to be free of nails and excess finish in the tongue & grove area.
• Re-milled finish lumber is available, but takes some practice to use effectively.  In retrospect, I don't regret buying "character grade", but in the future I would want to see a large sample, like 10 full size boards of whatever size you're buying.
• Re-used trim is readily available, although not necessarily in the size and shape you want.  It will require significant processing to make it useable.

Once you really get into reclaimed materials they become almost like pieces of art.  This fact makes them both expensive and the most loved pieces in any house.   For more information (and lots of pictures!) on how others have used salvage materials, see http://www.ci.seattle.wa.us/sustainablebuilding/greenhome.htm

There is still a large pile of leftover wood in the basement, although we've used some of it.  Although its the wood that had the most "defects", most of it will be useable by cutting it into smaller pieces and gluing them together.

Adaptability

Our adaptable wiring system works, but leaves much to be desired.  We're calling it a work in progress.

Sound Insulation

While the sound barrier between us and the ADU is very good, low frequency noise still travels thru the floor because we didn't isolate that.  Its not a problem, but takes some getting used to when you've lived many years in a detached home.  I suspect there are good solutions for this, but they're not obvious since the two units sit on the same load bearing wall.

Cost

The whole project ended up more expensive than we hoped, but this is almost always the case for any custom construction, and we don't attribute much of that to the various green features.  Our final construction cost was around $575k, which comes out to about $230/SF.  This doesn't include cost of the land, the original house, landscaping, architects fee and all the various other permits and design fees we had.  While this is very expensive, its not out of line with comparable non-green projects.

While we have only ballpark numbers on the cost of many of our green features (see Costs in the design section), we believe they're fairly accurate.  Based on these, the majority of our "green" features added up to $22k, or less than 5% of our total cost.  If you add in the PV, active solar and Cistern, it goes up to $58.5K or closer to 12% of our total cost.

Landscaping

While turf grass is cheap and easy to put in, a useful outdoor space filled with perennials is neither cheap nor easy.  The good news is that you can put it in slowly, and as long as you keep up on the weeds, it probably could be lower maintenance than turf grass. However, in the process I became a bit of a plant nut, and now have over 200 different perennials.  It would have been much easier for me to restrict myself to a much smaller number of only the tough ones, but like everyone who gets hooked on gardening, I wanted to have flowers or something of interest all over the yard all year long.  In Jan 2006, we took over the partly shaded parking strip section, filling it mostly with tougher plants that are largely scavenged from neighbors or other parts of the yard.  Some day I will add an entire gardening section.

Spring 2007 update: much of the garden has filled in, and we now have more than 250 different perennials.  Documenting them has gotten to be a time consuming task since they die, they get moved, I loose the tag or get a plant as a gift and don't know what it is, so I'm no longer sure of the exact number.

As with many other gardeners, I'm now running into the problem of plants got bigger than expected and plants that spread more than desirable.  In our initial installation, I planted about eight 4" pots of Sweet Woodruff (Galium odoratum) and am now ripped out many square feet of it.  Its an attractive plant, but much too aggressive for  good garden soil and regular water, so I've moved chunks of it to difficult spots and am ripping the rest out.  All the old english ivy (Hedera helix) and St Johns wart (Hypericum) are also gone.  I also had to move the Shasta daisy's as they were starting to take over.

Some day I will post a complete garden description, although generally I'm finding reasonable resources for northwest gardens on the web already.  Here are a few I found useful:

www.greatplantpicks.org

www.rainyside.com

www.paghat.com 

We ran out of water in the tank last summer due to drought and new plants that needed water.  In the two years of using drip, I've finally discovered its limitations.  Everyone says to water deeply, but I've found at least two problems with this wisdom.  The first is that in well drained soil, a drip system will produce a column of water that can easily be much narrower than the root system of your plant.  The second is that many plants produce only shallow roots. To correct for this, I'm going to try installing more drippers on the bigger plants, and running the system for less time.  My idea now is to try to get water into the top 4-6" of soil, rather than all the way down to 12".