GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Uneven Temperatures Between Floors

nateflanigan | Posted in General Questions on

Hi – Kind of a boring question.
Zone 4a
Old craftsman – with recent air sealing and extra attic insulation
The house is new to me – this is my first spring here

The weather has been nice lately – a little warm during the day but cool at night. However it’s been a little uncomfortably warm for sleeping overnight yet when I come down stairs to make coffee in the morning the first floor feels nice and cool. I checked the upstairs vs downstairs thermostats this morning and was surprised to see a 3 degree difference 68 and 71. Bedroom window was open with a ceiling fan on, downstairs was closed up. Not a big deal right now but, in a month or two might get unpleasant.

I’m a bit confused about what’s happening here, how to diagnose it and what to do about it.

– Is the attic just holding a ton of heat and slowly radiating through the insulation and ceiling all night?
– The roof just has ridge vents, would a powered gable vent be a good idea?

What other intel can a provide to help isolate the cause?

Thanks!

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.

Replies

  1. user-723121 | | #1

    One of your questions is likely right on.

    – Is the attic just holding a ton of heat and slowly radiating through the insulation and ceiling all night?

    Solar warming of the roof and ceiling insulation during the day is showing up as a temperature differential between floors. Any nightly supplemental heating is adding to this because of the stack effect.

    1. Expert Member
      Dana Dorsett | | #3

      >"Solar warming of the roof and ceiling insulation during the day is showing up as a temperature differential between floors. Any nightly supplemental heating is adding to this because of the stack effect."

      >"Bedroom window was open..."

      Stack effect air leakage & infiltration is likely to be your biggest problem, not shoulder season solar gain.

      With the upstairs window open the stack effect is operating BIG TIME. By lowering the pressure upstairs the depressurization downstairs is even greater, driving even more infiltration into the downstairs.

      The way to control this is to air seal both the top of the house (close the window already!) between the attic & living space (whether kneewalls or attic floor), AND to air seal the BOTTOM of the house, starting with the basement rim-joists & mudsills, and any basement windows & doors, and any flue /electrical/plumbing chases that might run from basement through to the attic. These air leaks are sometimes hard to chase down without using blower doors & IR cameras, and kneewall attics are notoriously hard to air seal without paying very close attention. (Floor joist bays running beneath the kneewalls need real air dams on each bay- you can't just stuff them with insulation and call it a day.)

      By way of example...

      I live in a 1.5 story 1920s arts & crafts bungalow with kneewalled attic. The air sealing at the top wasn't anywhere near under control until it was insulated at the underside of the roof deck with spray foam, and even then there was some "mystery leakage" that took a bit more to discover. It turned out to be a 3" x 15' gap under a beam that separated the main house roof from a blind attic over the porch. The foam crew had insulated down to that beam, but not the gap, probably due to how difficult it was to access or even see that gap. (The beam and roof deck above was obvious, and could be sprayed from several feet away.) Getting to that leak required contortion crawls through one end of a kneewall attic, then dropping about 2' to a (rock wool insulated) porch roof deck level that was also at ceiling level of the entry/mudroom. I ended up sealing that gap with corrugated cardboard caulked to the beam and attic floor, taping the seams with housewrap tape, then dense-packing the porch-attic space with cellulose.

      My foundation is (rare for the era) poured concrete, which I insulated with 3" of reclaimed roofing polyiso (labeled R17, performing probably ~R15), seams taped and sealed with 1-part can foam. The rim joists & mudsills are similarly sealed with cut'n'cobbled foam board + can foam. The antique steel framed triple-light single pane basement windows have been replaced with U0.5 vinyl double panes, and the leaky unrepairable paneled basement door was replace with a U0.25-ish weatherstripped insulated steel door. The utility & electrical penetrations in the basement ceiling have all be sealed with foam or caulk.

      With only full dimension 2x6 rafters the insulation level at the attic is well below code (I'm in zone 5A), but after air sealing the upstairs temps during the shoulder seasons remain well below the temperature of the main floor below, unless actively heated. The loft-office at the top of the open stairwell DOES heat up from convection and direct PM solar gain through the windows, but during the shoulder season that can be minimized by simply opening the door between the loft office to the main upstairs rooms (a large bedroom with a full bath off a dormer), to allow the cool air from upstairs to convect freely with the warmer air coming up the stairs. During the summers the main space stays pretty cool most of the day, but the upstairs can get uncomfortably warm without running a window-shaker AC. This is due to the high solar gain on the south facing dark shingled roof pitch. The main ridge runs east-west, and the majority of the window area is west-facing, for a hot-afternoon booster to the solar gain. Even so, with the door to the office closed a 5000 BTU/hr window shaker can and does keep up with the cooling load for that floor.

      Last year we used a 1-ton window-shaker (a U-shaped Midea, now also sold under the Mr. Cool label) in the loft office, with convection in the open stairwell it keeps up (with margin) with the entire house (2400' total, not counting the basement) at the low-80sF 1% outside design temp. The grotesquely oversized 5 ton central air doesn't kick on until it's north of 90F outside, and the duct capacity to the upstairs isn't up to snuff once it's been heat-soaked for several hours. Without the window shakers the direct gain from the west facing glass in the office would make it unbearable during the dogdays. even though it's a comfy 72F this morning (before the sun is on the windows) exactly the same temperature as the living-room at the bottom of the stairs. It's 45F and sunny right now. With the upstairs heating zone off (nobody is using that room right now), it's coasting along at a 65F well below the downstairs temp. Yesterday it hit 70F outside, 73F on the main floor, yet it didn't get above 75 in the main part of the upstairs, even though it hit 78F in the loft in afternoon when the sun hit the loft windows, but dropped to comfortable mid-70s temps after I opened up the door to allow convection.

      Clearly YMMV. But you'll never get it under control until you air seal the house, starting with the top AND bottom of the stack effect "stack".

      It's worth reviewing Nate Adams' "Home Comfort 101" guide, including the short narrative videos found on this page:

      https://www.natethehousewhisperer.com/hvac-101.html

      Most of his case studies and active projects are in zone 5A, but funny thing- physics works the same everywhere.

      BTW: A powered attic vent is almost ALWAYS a bad idea. Actively depressurizing the attic increases the amount of air drawn from the conditioned space up into the attic. Even a ridge vent can create an infiltration drive unless there is more free air cross section of soffit venting than at the ridge. See:

      https://www.greenbuildingadvisor.com/article/all-about-attic-venting

      The correct way to think about attic venting is as a method of purging MOISTURE (from the attic, to keep the rafters & roof deck from rotting), not so much about purging heat.

      1. nateflanigan | | #5

        Thanks Dana - I did mention in my initial post that the house has been recently air sealed. I didn't go into great detail but most all of what you outlined has been done. I know it's not perfect, and I'll continue to work on little tweaks but for what the house is it's been taken about as far as it can go without major renovation work.

        To clarify I'm complaining about it being too warm upstairs - it is by my standards, at night right now, nice and cool downstairs and nice and cool outside. If we're talking about stack effect and air infiltration, where is that too warm air coming from if not the attic?

        Note: the attic is not a finished space, there is r-19 batts in the bays and 8" of cellulose on top of that.

        Other note: there is no mechanical heating or cooling going on right now.

        1. Expert Member
          Dana Dorsett | | #9

          > "I did mention in my initial post that the house has been recently air sealed."

          What was the final cfm/50 number on the blower door? As a crude rule of thumb (with a few notable exceptions) if the blower door comes in under 1 cfm/50 per square foot of fully conditioned space (not counting basement) it is probably not a direct air infiltration issue. If it's above that stack effect infiltration is still on the suspect list.

          >"If we're talking about stack effect and air infiltration, where is that too warm air coming from if not the attic?"

          ---

          >"Note: the attic is not a finished space, there is r-19 batts in the bays and 8" of cellulose on top of that."

          Warm air rises- it's not the warm AIR in an insulated attic heating up the spaces below. A hot roof deck can radiate significant amounts of heat to the ceiling through crummy low density R19s, but not through 8" of cellulose. Unless there are big gaps in the insulation or significant air leakage, look elsewhere for the source of the heat gains.

          Direct solar gain through windows and walls are often much higher on second floors than first floors due to less shading factor. The shadows of adjacent buildings/trees/hills reach the lower levels first.

          Are the walls insulated? Is this house balloon framed? Convection within empty wall cavities can create surprisingly high stratification, leading to warmer walls.

          If you have high confidence that the house is tight (based on a blower door measurement, not a WAG) it's time to stop guessing and take some direct measurements of wall & ceiling temperatures (in several places on each wall/ceiling area) over the course of a sunny day as the difference in first & second floor temperatures increases.

          A <$100 pistol grip IR thermometer can be a very useful instrument, but a $230 FLIR One (see: https://www.flir.com/products/flir-one-gen-3/?vertical=condition-monitoring&segment=solutions ) can find the hot spots quicker. Any gaps in wall or ceiling insulation become obvious, and with even a medium sized window fan depressurizing the room obscure leaks of any size can become obvious when there's a 15F+ difference between indoor & outdoor temps.

          If there are large open air pathways between the first & second floor convection within the conditioned space can create 5F or greater stratification between first & second floors, but it has to be pretty big. The loft office at my house at the top of the open stairway to the living room also overlooks the kitchen, but the door between the loft and the main rooms upstairs very effectively blocks convection between those spaces and the first floor. The higher pitched cathedralized portion of the ceiling in that room pitches toward the east, the flatter part (it's a shed dormer) pitches west, the ceilings have only R21 high density fiberglass for insulation in that area. The main mid-day gains that room are the section of south facing wall. By mid afternoon the west facing window gains are the main driver of heat gain in that space, despite the relatively low-R ceiling.

          >"Other note: there is no mechanical heating or cooling going on right now."

          If the house is leaky it might be more energy efficient to cool the upstairs with window AC rather than direct ventilation with a fan. (I pass several 2 story houses with AC running in upstairs windows on 50F sunny days while walking the dog in my neighborhood). A window AC won't pressurize or depressurize the room (unless it is running in a "ventilation" mode). When cooling the upstairs by ventilating with a fan, open a window on the opposite end of the upstairs and keep the doors open so that the rooms don't become too pressurized/depressurized relative to the already comfortable downstairs.

          1. nateflanigan | | #10

            Thanks for the insightful reply -
            Blower door # was 4000 - which puts me around 5 ach with
            my rough volume estimates.

            > "Are the walls insulated? Is this house balloon framed? Convection within empty wall cavities can create surprisingly high stratification, leading to warmer walls."

            - Depends on the wall - the too warm bedrooms are an addition from the 90's insulated with r-19.
            - The house is balloon framed - the contractor that did the air sealing and insulation work blew cellulose into every wall cavity they could reach from the basement and sealed the cavity with a block of rigid foam.

            > "If you have high confidence that the house is tight (based on a blower door measurement, not a WAG) it's time to stop guessing and take some direct measurements of wall & ceiling temperatures (in several places on each wall/ceiling area) over the course of a sunny day as the difference in first & second floor temperatures increases."

            - I'm confident the house is NOT tight - but that there's not a lot more I can do right now. I can refine and continue to patch little holes but the biggest culprit I noticed over the winter was the windows. They're vinyl replacement windows, maybe from the renovation work that was done in the 90's.
            - It's been cool the past week or so, but I'll definitely take temperature readings once it starts to warm up.

  2. user-6623302 | | #2

    You may find that the ceiling fan is disrupting the natural venting of the warm air out the window. Try not using the fan.

  3. nynick | | #4

    Open a window downstairs and a second upstairs.

    1. nateflanigan | | #6

      This makes a lot of sense to me. It's basically creating a cold weather stack effect.

  4. avery_law | | #7

    I installed a whole house fan last summer to reduce the afternoon heat in the upstairs of my 1.5 story cape house. Once the temperature outside drops below the temperature inside, we fire up the whole house fan and crack the windows downstairs. Air sealing is probably a more scientific and proper solution but my whole house fan leaves us cool at night. We also cover our south facing skylights on the outside to reduce solar gain.

    1. nynick | | #8

      Whole house fans work. There's another idea for you.

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |