Anatomy of a Gut Remodel
This is the story of a seemingly simple structural fix that turned into a three-year, award-winning LEED for Homes gut remodel. It took three years because my wife graciously agreed to let me use this project as my personal building science experiment, plus I only worked on it part time. Since I’m a LEED for Homes Green Rater, we also decided to fold in a LEED for Homes certification. This is our sad story, with special emphasis on lessons learned.
We started demolition of the building in the spring of 2008. It’s the lower floor of a two-story, three-unit residential building in San Luis Obispo, about five miles from the ocean on the central coast of California. The local climate is mild, tempered by the ocean, with warm to hot summers and rarely freezing (2,500 heating degree-days).
The existing 8-foot-high foundation wall was 85-year-old unreinforced brick. The grout was mixed, according to local sources, with beach sand, which tends to degrade the mortar after less than 85 years.
Sometime in 2007, my wife and I noticed that the wall was starting to buckle outward. The cause was not obvious. Although the building is located close to a fault line, there hadn’t been any earthquakes in the previous few years. We had done significant remodeling to the upper level, but that had been about five years before. Regardless of the cause, the buckling meant we had to replace the wall.
Since we were going to do major surgery on the wall, we decided to improve the ground floor apartment on that level. It was small (about 325 square feet) and very dark, with a 1940s vintage natural-gas hot-water tank and a small in-wall electric-resistance heater. To be frank, it was also a dump.
There was another problem: It didn’t appear to be a legal living space. The city had no record of it, but the building records had been destroyed in a fire some years earlier. However, if we could prove that the apartment had, at some point, been a legal living unit, the planning department would grandfather it in without significant headaches.
After several discussions with the local planning department officials, and several attempts to find evidence of prior legal occupancy, we had almost given up. Finally, I went to the local library to inquire about historical documents. They actually had reverse-lookup telephone directories going back as far as the mid-1940s. I was able to find a telephone number for the apartment dated 1949. We got the green light.
We finally decided on the following project scope: Remove the failing wall and construct an additional 90 square feet in a half hexagon. The addition, which faces south, would include high-solar heat gain glazing with overhangs tuned for winter solar absorption. The floor would have 4 inches of insulated concrete plus brick recycled from the removed wall to soak up the solar heat. In addition, we would remove the water- and space-heating equipment, opting for extremely energy- and water-efficient replacements. We would radically seal and insulate the envelope, and we would incorporate as many recycled materials as possible. We used basic building science and the LEED for Homes checklist as a framework for most of our decisions.
Once we started tearing into the building, it became clear that a gut remodel was the only feasible choice. There was significant termite damage throughout, the plumbing was on its last legs, there was no insulation, the brick mortar was crumbling, and the unit had clearly been modified by nonprofessionals over many years. It was dangerous.
We also discovered one possible reason why the wall was failing. One of the support beams was resting on a double top plate that was resting on the entry doorjamb (see Photo 1). There was no header. Due to the extensive framing damage, we ended up replacing all the framing except for the ceiling joists. We were able to save short sections of the removed lumber, which we had ripped at a local lumber mill and used for shelving.
Additional key project features included
- a 98% efficient tankless Navien combi boiler for domestic hot water (DHW) and space heating, using small wall-mounted radiators;
- a SunCache open-loop passive-solar thermal collector for preheating the DHW;
- a Metlund on-demand DHW recirculation pump with bathroom occupancy sensor;
- Energy Star-certified appliances, lighting, and range hood;
- 15 CFM continuous ventilation with passive inlet to meet ASHRAE 62.2 requirements;
- zero VOC paints, caulks, adhesives, and one large earth plaster wall;
- ultra-efficient water fixtures, including a dual-flush toilet, that meet or exceed EPA’s Water Sense specifications; and
- minimized construction waste, diverting 90% from the landfill.
We also tried to incorporate as many recycled materials as possible into the project:
- All new flooring and the entryway stoop are brick recycled from the wall that was removed.
- All interior shelving, other than the kitchen cabinets, is from removed framing lumber that was ripped at a local lumber mill.
- The kitchen counter is made from at least 70% postconsumer recycled material (ECO by Cosentino).
- The wall insulation and the existing ceiling insulation are blown-in cellulose, with a high recycled content.
- We reused the existing (very heavy) cast-iron bathtub and the existing interior doors.
- The new roofing, faux slate that almost matches the existing roofing, is actually a very lightweight, recyclable cool-roof formulation made from 85% recycled plastic.
- We used 25% fly ash in all new concrete.
Air Sealing and Insulation
From the beginning, we paid special attention to air sealing, caulking, or foaming all penetrations. In addition, we used a double layer of sill sealer, with caulk underneath, on all the new bottom plates. As other authors for Home Energy have pointed out, extreme air sealing is an obsessive-compulsive process. We were pretty obsessive and compulsive. For instance, Photo 2 shows the rigid insulation in the ceiling of the hexagon addition. We cut all that insulation to fit tightly, then foamed all the joints and edges. We were equally attentive to all penetrations in all directions (see Photo 3).
At the outset of this project, I was hoping to meet or exceed the Passive House leakage standard of 0.6 ACH50. I thought our obsessive sealing would get us there, but I didn’t test the leakage until after insulating. I was very excited when I had to buy Energy Conservatory’s ring C in order to get a usable fan reading. That’s the ring for measuring between 85 and 300 CFM50. The final measurement came in at 220 CFM50, about 4 ACH50—respectable, but not even close to supertight.
Lesson Learned: Blower door test while air sealing. Because the sealing was strung out over many months, I didn’t take that approach. Also, the windows weren’t installed until after the insulation was done, so I couldn’t easily test at that stage. Finally, I was under the mistaken impression that well-installed blown-in cellulose insulation is an effective air barrier, so I only sealed up obvious penetrations in the ceiling. I totally ignored cracks between the 1925 vintage floor sheathing for the unit above. Knowing what I know now, I would have put a 1-inch foam flash coating on the ceiling.
We used a mix of rigid insulation and blown-in cellulose. The exterior walls ranged from 4 nominal inches to as deep as 12 inches in some spots. The deeper spots and the ceiling were netted (we insulated the ceiling even though it was adjacent to conditioned space above). I inspected the insulation after the truck had left. The damp mix that was not netted looked good and had the proper density. However, I found a few dozen spots where they had not fully filled behind the netted walls or ceiling. I had to call the insulator back to finish the job right.
Lesson Reinforced: Don’t assume that your subcontractors do a good job. Check their work.
As I mentioned, we installed a 98% efficient Navien combi boiler, the 150,000 Btu model CC-180A. It has a separate closed loop for radiant heating. This loop was connected to two small wall-mounted radiators and a towel warmer in the bathroom. In addition, we hooked up a Metlund on-demand hot-water pump to the DHW system for efficient hot-water delivery to the bathroom (the kitchen was less than 10 feet from the heater).
During installation, everything worked fine, except that the Metlund would take more than a minute to prime the hot-water loop, which was only about 50 feet round trip. That seemed way too slow. I upsized the unit from the smallest to the next-larger size. It still took about 45 seconds. Of course, I didn’t test this until after the piping was all installed and covered.
Lesson Learned: After thinking about this for a while (much longer than I should have needed to), I realized that there were two problems. The recirculation loop has a ¾-inch diameter, and the sensor for turning off the pump is at the end of the loop. I’ve sat through several variations of water expert Gary Klein’s basics class. It took a hard-to-correct, real-world error for me to finally really understand that distance matters a lot, but volume is probably more important. I could have made the recirculation line 3/8 inch and put the sensor in the bathroom, where it would shut off the pump once the supply half of the loop was full of hot water.
There was another problem with the hot-water installation that came up only after someone moved into the unit. The Navien would occasionally shut down (as the tenant put it, from “zero to four times a week”) when he was trying to take a shower. The only solution was to go to the utility room, outside the apartment, and reset the water heater. That’s not very convenient when you want to take a shower. Fortunately for us, the tenant was a techie college student who liked fiddling with hardware and solving problems.
I ended up working with Navien tech support for several days, early on, but we were unable to identify the problem. I replaced the primary circuit board and the remote control, I upsized the incoming gas line. Nothing made a difference. Finally, after ignoring the problem for about 18 months, I called Navien again and asked if they had any factory-trained plumbers in the area—information they hadn’t been able to provide to me in the past. They put me in touch with a local plumber who, within 15 minutes, identified the problem as a faulty igniter. It would occasionally fail to fire properly, and after three attempts, it would shut the unit down.
Lesson Learned: Try to avoid new models or new technology. There’s a reason why they call it bleeding edge. Although I really like the Navien hardware, it took their tech support folks many months to develop comprehensive field diagnostics. This also applies to materials. The roof manufacturer’s ultra lightweight, 85% recycled plastic, faux-slate cool roof seemed too good to be true when I bought it. Apparently it was. The company is no longer in business, not because of the quality of the material, but because the business model was unsustainable.
Modeling, Reality, and Certifications
This project was LEED Platinum certified in late 2009. In 2010 the U.S. Green Building Council C4 chapter, serving San Luis Obispo, Ventura, and Santa Barbara counties, gave it their Energy Efficiency award, primarily because the California energy code calculations showed that the apartment would use 80% less energy than a project built to Title 24. The project was also given a sizable utility incentive for the same reason.
Title 24 calculates energy use only for space heating, space cooling, DHW, fans, and pumps. Appliances, electronics, lighting, and solar PV are excluded. Site use is calculated in time-dependent valuation (TDV), where each therm and kWh is weighted hourly, based on the estimated cost to obtain that energy. Those numbers are useless when you’re trying to compare modeling to real-world performance.
California’s certified modeling software can provide site energy usage that includes appliances, electronics, lighting, and solar PV, and is not TDV weighted. It converts every kWh to roughly 3 times the site consumption to compensate for generation and transmission losses. Therms are unchanged. Converting actual consumption to site energy, using the same electricity multiplier, lets you compare the modeling to the real world.
Using that approach, the modeling for this project turns up remarkably inaccurate estimates. Using annual utility consumption, I found that natural-gas consumption was 290% higher than predicted and electricity consumption was almost 340% higher than predicted.
Lesson Reinforced: Modeling software is not the same as reality. It can’t take into account all the factors needed to accurately predict utility usage. Unfortunately, it’s about the only tool we have when designing a building to gauge that building’s performance relative to another design—there’s nothing to measure. Many certification programs, including Energy Star, LEED, Green Point Rated, and HERS, currently use modeling, right or wrong, as the basis for predicting energy efficiency performance.
Assuming you understand the limitations of modeling software, there is still value to be gained from green certification systems. Many smart people have put a lot of thought into, for instance, LEED for Homes. That collective brain power can do a much more comprehensive job than I can at figuring out how to build a home that’s sturdy, long lasting, healthy, good for the planet, and possibly energy efficient. Most green certification systems provide a good framework to guide design and construction practices.
When I first started this project, I thought for sure I could easily qualify to meet the Thousand Home Challenge, which analyzes deep energy retrofits to see if they meet certain energy efficiency thresholds. Just as I was at first surprised at my CFM50 air leakage numbers, I was surprised to find that I failed the Thousand Home Challenge—because the house used too much electricity. That got me to thinking about the Navien problems, and the Metlund pump issues, and the occupant’s behavior.
Contact the author at firstname.lastname@example.org.
Since then I’ve disconnected the pump and fixed the Navien. We also got a new tenant who is not a techie college student but a mature working adult, who doesn’t have a house full of electronics or stay up all night. Over the next several months I’ll be following her utility consumption to see if those items make a difference. [Postscript—it appears that the new tenant’s electric consumption is roughly 40% less than the previous tenant’s, but it’s unclear whether that is due to her behavior or to the fact that I disconnected the pump. That will be the next thing to investigate.] In the meantime I’m factoring all these learned lessons into my next two projects—a 300 ft2 new building, and another gut remodel after that. Even though this project didn’t turn out as cool as I had hoped, it’s still pretty good looking (see Photo 4), and it should survive the next earthquake.
Lesson Learned: There’s always more to learn.
Enter your comments in the box below:
(Please note that all comments are subject to review prior to posting.)
While we will do our best to monitor all comments and blog posts for accuracy and relevancy, Home Energy is not responsible for content posted by our readers or third parties. Home Energy reserves the right to edit or remove comments or blog posts that do not meet our community guidelines.