Blowing Smoke And Air And Not Smoke
Wow, things got busy. In an effort to get out and make a little money while slowly rebuilding my client base, I’ve been doing some LTE work with the DNR. It’s a pay cut, but the people are great, and the work is gratifying, outdoors, and crucially, not construction. I’ve spent the last month and a half at the state tree nursery, harvesting, grading, and packaging 5 million trees for the DNR’s big spring tree sale. It seemed appropriate for a carpenter to help get a bunch of trees in the ground. I told everyone I worked with that I was just there to secure the supply chain for my business. They all laughed politely, God bless ‘em.
But I didn’t do any writing on the weekends like I told myself I would, so I’ve got some catching up to do. I left off with the interior framing finished, and ready to start running mechanicals.
“Mechanicals” generally refer to the invisible trades that run through the walls and provide comfort and convenience to a house, mainly plumbing, electrical and HVAC (Heating, Ventilation, and Air Conditioning). Natural gas or liquid propane (LP) lines would also qualify I suppose, but that’s N/A here because we’re all-electric and really smug about it.
There’s not much to say about the plumbing and electrical, honestly. Those processes were mercifully boring, thanks to smart design and some really great subcontractors. Other than some low-voltage lighting and air-sealing any exterior wall penetrations, our plumbing and electrical are pretty conventional. Our HVAC is another story.
Ω has all 3: Heating, Ventilation, and AC. The Heating and A/C are both provided by “ductless mini-splits” which are a type of “air source heat pump”. Heat pumps do just that: they change the temperature in one place and pump that temperature differential (aka ∆ T) to another . This movement of heat requires significantly less energy than traditional heat generation methods which require capturing heat loss from combustion, or forcing electrical current through resistive metals. Another type of heat pump is a “ground-source heat pump” or as it’s often called: geo-thermal. These take heat from underground, a very constant 50ish˚F, via buried refrigerant lines, while our mini-splits take heat from the the exterior air instead. What’s incredible to me is that the equipment has become so powerful/efficient that it can extract enough heat from damn sub-zero air to add significant heat to an interior conditioned space. But it totally can, and so mini-splits can be a primary heat source as long as the building has sufficient resistance to heat loss (i.e. well insulated and air-tight). Luckily, Ω is. Well, for the most part. There was a proper polar vortex in February where the heat loss out-paced the downstairs mini-split overnight. When we got up to make tea, it was 48˚F in the living room, where we have a 25+ sq ft window next to an 80 sq ft patio door. While these both have good air-tightness and pretty good R values for windows, our walls are still a dozen times better insulated. But this was exactly as expected and why next winter there’ll be a Masonry Heater sitting a few feet away, radiating out heat from the evening fire all night long. When the sun came up, that window and patio door (which face East and South, respectively) provided enough passive solar gain to get it up into the 70s in that room. Thankfully, the coldest winter days are often the sunniest.
We sized our 2 mini splits fairly small, which partly accounts for the inability of the living room unit to keep up on the coldest couple nights of the year. That one is a 9,000 BTU unit, with another smaller, 6,000 BTU unit serving the upstairs. Part of the reason we “undersized” these units is because we knew we’d have supplemental wood heat for when it got really, really, cold. The other reason is that an oversized unit, while handling extreme temperatures somewhat better, can struggle in the shoulder seasons. When the heating (or cooling) demand is low, a more powerful unit barely gets up to speed before the conditioned space is up (or down) to temp, and the thermostat shuts it off. This “low-load cycling” can lead to unnecessary energy usage and other issues. Considering our climate, where the shoulder seasons make up most of the year (not to mention some mild winters that I’m sure aren’t ominous), I’d rather have something smaller that needs a lil help a couple nights of the year, then a big hoss who inflates my electric bill the rest of the time.
The problems with oversized HVAC aren’t specific to mini-splits or heat pumps, either. Furnaces and A/C units can have similar efficiency issues in “low demand” situations and can sometimes fail to perform other functions like humidity control if they only run in short bursts. Unfortunately, oversized equipment is fairly common, if not the norm. A “heat-loss calculation” or “Manual J” is a fairly simple process (hell, I was able to grok it) for determining how much heat a specific building will lose at a certain temperature differential over time. It accounts for everything about the space: its local climate, it’s wall insulation, windows and doors, air leakage, even how many people are in there and how much passive solar gain it gets. Once you know how much heat the space loses, it’s a short hike to how much heat you need to add, and therefore the size of the equipment that should be installed. The problem is, these calculations take time (not much time though!) and most clients don’t know to demand them. As such, many installers just use a rule of thumb their uncle taught them to size the equipment. Naturally, this method errs on the side of bigger, because while you may get a complaint (or god forbid called back to a site) about a cold spot from poor framing or insulation work, no one’s gonna call you back about the electricity or gas they’re not saving.
Heat loss calculations are a great tool, and literally anyone can do them. There’s no special knowledge or tools or software required. All you need to know is the insulation values and sizes of your walls, windows, ceilings and doors. And how air tight your house is.
OK, that last part does require some special equipment if you want to plug an accurate value into your spreadsheet. The amount of heat loss due to infiltration, aka air leakage, can only be accurately measured with something called a blower door test. A blower door test involves closing up the whole house except for one exterior door, which has a large fan that gets sealed into the opening. It’s pointed outwards and acts as an exhaust fan, depressurizing the house to an exact pressure, 50 Pascals. Once the house reaches that pressure, a meter on the fan tells you how much airflow (in Cubic Feet per Minute) is required to maintain the 50 Pascal pressure (or CFM50). That airflow should exactly match the amount of air leakage in the building envelope. If you multiply that CFM50 airflow by 60 (minutes in an hour), and divide it by the house’s total volume (in Cubic Feet) you get the Air Changes per Hour @ 50 Pascals (or ACH50) of your house. You can use that number to calculate heat loss due to air infiltration.
We ran a couple heat loss calculations based on different benchmarks: 3 ACH50 based on International Residential Code (IRC), and our personal goal: 1 ACH50. Because this was 6 months beore we even broke ground, all we could do was guess. Through all the framing, foaming, caulking, taping, hanging doors, windows and plastic, I crossed my fingers and did my best. All I could do was hope it was good enough to meet our air sealing goals so that our “undersized” HVAC equipment would be big enough. On April 1st, 2023 we got our first and only blower door test and finally got some damn data.
Ideally, we’d have gotten 3 tests at different stages of the build. 1st (when we did): after the air barrier was complete, but before insulation was installed, so that any gaps and cracks could be identified and fixed. 2nd after drywall, and 3rd after the house was all finished. Unfortunately, there’s basically no energy raters in our area so we were lucky to find a guy who’d drive an hour and a half out to do a test, even if he’d only do it once. We each guessed what our test results would be. We’d discussed our goal of 1 ACH50, which would equate to a CFM50 score of about 450. I was the most optimistic and guessed we’d beat our goal with a score of 420 CFM50 (lower score = tighter house). Danielle guessed 450, and Cory the energy rater guessed 500.
He turned it on, said, “Oh boy” and started making adjustments to the fan.
These fans need to be fairly large so they can move enough air to balance all the infiltration in a large, leaky house. That means that their minimum CFM is fairly high; they’re good at moving lots of air, not so good at moving just a little, like you’d need to on a tight house or to test an individual room. That’s why blower door kits come with different sized rings (size A, size B, C, D, etc.) that you can use to cover up parts of the fan. By restricting the size of the fan, you lower the range of CFM that the fan can accurately measure.
You need the Size C ring to measure a house that’s between 300CFM50 - 85 CFM50. Cory didn’t own a Size C ring, because he’d literally never needed one. Until that day. I still don’t know what our actual CFM50 score was, because our house was too tight for his equipment to get an accurate reading. Here’s what I do know:
It was at most 300 CFM50
Based on the rough numbers we were getting, it was probably in the 150 CFM50 - 250 CFM50 range
Everything we’ve done since: insulation, drywall, plaster, hell even painting, has only made the house tighter
Our goal was 450 CFM50 for an ACH50 of 1. I’m confident we did at least twice as well as our goal.
As I’ve mentioned in a previous post, the idea that a house can be too tight is false. A house needs fresh air, yes, but intentional, mechanical ventilation is a much better way to get it than through random leaks in the building envelope. In other words, it’s not that a house is too tight; it’s that it doesn’t have adequate ventilation. The other concern with a tight house is if combustion equipment (gas or LP water heater, furnace, or wood fireplace) is “draft vented” (uses a chimney rather than a sealed exhaust fan to vent fumes). These can back draft if say, a powerful range hood is running exhaust when the water heater kicks on. This can cause carbon monoxide to build up to dangerous levels. It’s worth noting, a more common version of this horror story comes from draft vented equipment using a bad or leaky chimney (see my last post re: chimneys) instead of drilling a couple new holes and doing it right. It’s also easily fixed by installing power vented equipment that uses a sealed exhaust system for a few hundred dollars more.
We use an ERV system (Energy or Enthalpy Recovery Ventilator, I’ve heard it both ways) to get our fresh air. This thing is a nifty piece of business. It replaces the air in our house while exchanging the energy in that air so we don’t lose it. Winter, when opening up a window is least appetizing, is when ventilation is needed most. With the house closed up, humidity levels rise thanks to all the mammals and their shenanigans. Other contaminants, oils from cooking, methane from dog butts, and even off-gassing from plastic products like furniture upholstery build up in the space, nevermind the fumes from gas/LP appliances. The ERV fan exhausts that stale air but before it goes, it leaves most of its heat and some of its moisture on a heat exhanger in the ERV. After about 20 minutes, with the heat exhanger warmed up, the ERV switches over and draws fresh (albeit cold) air in and over that heat exchanger, bringing the fresh air back to about 75% of the stale air’s temp and adding some moisture so it doesn’t get uncomfortably dry. The spruced up fresh air is then run through a HEPA filter and gently blown back into the living space through strategically placed vents. The ERV control panel in the living room/kitchen acts as a thermostat but for humidity. There’s multiple settings and speeds so that if the neighbors and their pups come over or I bake of bunch of pizzas, the humidistat can kick things into high gear and keep up with the extra moisture load in the air.
Danielle ran all the duct work for the ERV with minimal help from me. To make the most of the ERV’s power, she sealed up all ductwork joints with mastic so all that air went where we wanted. The ductwork is the same you’d use for venting a bath fan; small, economical and easy to work with. She made beautiful grill covers that blend perfectly with the walls and ceilings.
She used the extra cover material for our “return air pathways” the final piece of our HVAC system. These are literally just holes in the wall. As I mentioned, the vaulted ceiling in the living room hits the adjacent second story above the second floor deck. When we were designing the house and planning to load up on heat sources in the living room, an obivous concern was “how do we get heat from that room to the far end of the second floor?” Our solution was to put “passive” vents in the upper wall, above that second floor deck. Warm air in the living room rises, hits the vaulted ceiling and “rolls” up towards the wall. It naturally floows through these passive air vents and into the second floor living space. Cooler air sinks down the stairway to be reheated in the living room, thus completing the cycle. The passive vents have cardboard baffling in them to deaden noise exchange between the floors, without inhibiting air exchange.
On those two bitter cold nights, the smaller, upstairs mini-split had no trouble at all keeping up with the heat loss, and the bedroom stayed perfectly comfortable. The living room got cold because the heating part of the cycle was shut down, and the temperature layers stratifyed between floors. A 35lb Hickory fire in the Masonry Heater right after dinner will keep that engine turning and the whole house cozy and comfortable til the morning fire.
This is what’s possible when a building is designed, not just to stand up and stay warm but, to perform.
With mechanicals in and our blower door test aced, we were almost ready for drywall and plaster. But first: insulation. This was maybe the most stressful project to prep, and my favorite project to execute. It was also as Macgyver as I got on the whole build, which if you know me, is absolutely my element.