GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter 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

Ramblings on thermal mass, AC and window fans

user-4053553 | Posted in Energy Efficiency and Durability on

This is isn’t really a specific question so feel free to offer random thoughts

I live in a century old home (700 sq ft) with plenty of thermal mass (vinyl siding, wood siding, true 2×4, 1.5-2in thick barnboard, plaster, lath, more plaster), traditional attic, little insulation (no basement, loose fill cellulose and single batts in attic rafters), the windows are double pane ~7 year old vinyl but i could not find out if they have a low SHGC coating or R value (e-mailing the company was not helpful, they could not tell me) and the house has lots of air infiltration (~15ACH50, much from half basement/half crawlspace). I’m upgrading as funds and ability allow.

I try not to use the central air because its a decade old 1.4-5(?) ton 10SEER unit, but being near Toronto Ontario we have had a heat wave the last few weeks. So what i try to do is cool down at night, since our electricity is near 20c/kWh even off peak. What i have noticed is that it takes about an 1-1.25 hr to reduce the temp by 1C (2F) which also reduces humidity 5-10%. It never seems to go below 50% humidity even after 4 hours of A/C. So I’m assuming that 1C has 20,000 btu or about 6kWh (rough calculation). Agree?

The A/C unit uses about 1.8kW while running according to the whole house energy meter, of which 0.4kW is the ECM furnace fan (running on full, when heating its 0.1kW on low fire, 0.2kW on high fire and 0.4kW AC)

If starting at 25C on the hottest day of about 35C outside the temp will go up to about 30C by the end of the day. With 32C which we have had multiple days 24-25C going to 29C
If 6kWh/ºC is correct that suggests the house is gaining about 30kWh of heat on very hot days

I have found window fans work poorly, but they also seem to be low CFM. At a 5C differential 2-3 hours gets me 1C of reduction. They don’t have a rated CFM, I would hope my ~10 000cf house would be coolable with a powerful enough fan, but to match the 1C and hour how much air flow would i need?

If someday i got an HRV could it somehow work at reducing summer temps, because at night i would want lots of cfm coming in to cool the house with no heat exchange

Your thoughts?

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.


  1. GBA Editor
    Martin Holladay | | #1

    Q. "If someday I got an HRV could it somehow work at reducing summer temps? (Because at night I would want lots of cfm coming in to cool the house with no heat exchange.)"

    A. No. An HRV is not designed to lower indoor temperatures. It has a fan with a low cfm rating, and its purpose is to provide a little bit of fresh air for occupants. For more information on this issue, see Misconceptions About HRVs and ERVs.

    If you want a fan with "lots of cfm" to help cool your house at night, you need something called a whole-house fan. For more information on whole-house fans, see Fans in the Attic: Do They Help or Do They Hurt?

    The main problem with your house, and your attempts to keep it cool, is that your house has a lousy thermal envelope. It is leaky and poorly insulated. If you can gradually improve your home's thermal envelope, you'll eventually find that the house is easier to cool.

  2. Dana1 | | #2

    The building structure as-described is a fairly low thermal mass house.

    Low SGHC windows would usually have to be special ordered in Canada. They may or may not be low-E, but if it has a low E coating is't probably indium tin oxide, which is far more about retaining heat than rejected solar gain.

    The inability to drop the relative humidity to under 50% @ 25C is likely to be an indication of high infiltration rates. (Ether that, or you're boiling stuff on the stove all night without a lid on the pan. ;-) ) How are you measuring the RH?

    When beginning a cooling cycle at high humidity (60% or more) a large fraction of the cooling load is the heat of vaporization of the water being removed from the air. This is part of why it takes a long time to lower the temperature 1C. The phase change of water behaves somewhat like a large thermal mass. If more humidity is constantly being brought in via outdoor air infiltration, it further slows the rate at which the temperature can drop.

    So, there NO insulation in the full-dimension 2x4 walls?

  3. user-4053553 | | #3

    @ Martin
    Thanks for the link to the whole house fan article, though i always figured blowing cold air into the house would work better (in theory there should be no difference either way, but with a window fan i found depressurizing somehow didn't work as well, not that i can explain why). I would like a high powered fan for a window or two while the other windows are open so there is no attic involvement, but 15ACH is very interesting. I would need a 2500cfm fan to pull that off.
    I completely agree with the house having a "lousy thermal envelope", i call it the unmitigable energy efficiency disaster. I am only partially joking.

    @ Dana
    Most of my air infiltration is from the basement/crawlspace which has a dehumidifier set at 60% running 24/7.
    Fixing that problem is a huge job :(
    The door is closed from basement to upstairs, though its not airtight.
    I am surprised your saying its a low thermal mass house, the walls are almost a foot thick!
    If low SHGC windows have to be specially ordered then it unlikely i have them, i wish the manufacturer was more helpful. Perhaps i can contact them again and hope i can get more information.

    Nothing is boiling all night, i try to do less cooking in summer because the heat is indoors, and the BBBQ is certainly not indoors or the monoxide would have killed many brain cells by now :P
    No other water sources either, no standing water, steam baths or anything like that.
    RH is measured with a cheap meter stuck on the fridge, a $3 home depot device.

    "When beginning a cooling cycle at high humidity (60% or more) a large fraction of the cooling load is the heat of vaporization of the water being removed from the air. This is part of why it takes a long time to lower the temperature 1C. The phase change of water behaves somewhat like a large thermal mass. If more humidity is constantly being brought in via outdoor air infiltration, it further slows the rate at which the temperature can drop."

    I wondered about this, is there a handy calculation about how much this is contributing?
    Would it be cheaper to dehumidify the entire house with a 1.85L/kWh dehumidifier?
    I seem to start at about 75-80% humidity and get down to 60%, i'm averaging 2-3 hours of central air (to balance temp with cost)
    There is loose fill cellulose in the walls

  4. Dana1 | | #4

    Being a foot thick doesn't impart thermal mass to the wall. The material types and masses matter. Typical soft wood species used for framing barn wood sheathing run about 0.5-0.6 BTU/lb per degree F, and the densities run 25-40lbs per cubic foot (2- 3.5 lbs per board-foot). The amount of "extra" wood in your house compared to current practices is substantial but still less than 2x and nowhere near the thermal mass of a foot thick log wall. Yours is roughly equivalent to a solid wood wall 3" thick. At 3lbs per board foot would have a thermal mass of 3" x 3lbs x 0.6 BTU/lb per degree F= 1.8 BTU per degree F per square foot.

    Concrete runs ~0.18 BTU/lb per degree F but has much higher density, about 150 lbs per cubic foot, or 12.5lbs per board foot. A barely code legal 4" concrete wall would run 4" x 12.5.lbs x 0.18 BTU/lb = 9 BTU per degree F per square foot, about 5x the thermal mass of the napkin math estimate for your house.

    That's why concrete & masonry houses are considered "high thermal mass" construction, and all timber framed wood sheathed houses are considered "low thermal mass", in relative terms. Thick solid wood construction has substantial thermal mass but still not as high as typical concrete.& CMU walls.

    Psychrometric charts have the scale for of the mass of moisture per unit mass of dry air on the right edge of the scale. The volume of your house. At 25C there is about 1.5x as much moisture in the air at 70% RH as there is a 50%, but you have to estimate the volume and density of the air in your house (the density varies with altitude) to come up with the total mass of water being removed. To convert that to BTU, the latent heat of vaporization is 970.4 BTU/ lb of water- it adds up. With air-handler driven outdoor air infiltration there is the additional latent heat of the moisture in outdoor air when the outdoor dew points are above 14C (= the dew point of 25C air at 50% RH.)

  5. GBA Editor
    Martin Holladay | | #5

    Dana Dorsett is correct. More info here: All About Thermal Mass.

  6. user-4053553 | | #6

    I'm at about 650 ft above sea level. I don't quite understand how to read that chart but i see what you mean about thermal mass.

    I am now curious how much energy is being used to dehumidify the air vs cooling done, i can collect and weigh the water for an hour and calculate the heat of vapourization.

    I am also curious if my house were drywall/2x6/roxul/sheathing/rigid foam (say R20 worth and 1ACH50) how much more quickly it would cool down and how much energy would be saved if both started at 28ºC for say 1000sqft.

    If the house were airtight and starting at 75% what should i expect for a final humidity after say 2 hours of central air operation? What would i expect it to settle at if i let it run 24/7? Not sure if this is fungible information or not.

  7. GBA Editor
    Martin Holladay | | #7

    Q. "I don't quite understand how to read that chart."

    A. For more information on reading the psychrometric chart, see How to Use the Psychrometric Chart.

    Q. "I am now curious how much energy is being used to dehumidify the air vs cooling done."

    A. Under peak conditions, the typical residential air conditioner provides about 70% of its cooling capacity as sensible cooling and 30% as latent cooling. In a hot, humid climate, however, it’s possible for 40% or 50% of a home’s cooling load to be a latent load. For more information on this topic, see Air Conditioner Basics.

    Q. "If the house were airtight and starting at 75% what should I expect for a final humidity after say 2 hours of central air operation?"

    A. The answer depends on so many factors that it's impossible to provide an answer with the information you've provided. You may want to read this article: All About Dehumidifiers.

  8. morganparis | | #8

    Just a footnote to all the good advice above but as a general point, just in case you're tempted, under no circumstances should you open windows at night and run a/c later in the day. Yes, it's that latent heat of vaporization thing. Whenever humidity is a factor in your cooling concerns air sealing is key.

  9. Expert Member
    Dana Dorsett | | #9

    +1 on James Morgan's comments!

    I usually look at the outdoor dew point readings from a few nearby weather stations (they're often way off, especially the amateurs) before deciding if it's dry enough to make opening the windows worthwhile. If the outdoor dew points are 13C/55F or less it's fine to open the windows. But when the outdoor dew points are running 15C/59F and higher it becomes a real latent load if your goal is to keep it under 50% RH @ 23-25C.

    According to Wunderground weather station data the current dew points at this hour in Toronto are running in the 20C/68F range: (Try checking a few stations using their "Change Station" pull down.)

    It will change over the course of the day, but if it's not under 60F, keep the windows closed even if it's cooler outside than indoors.

  10. user-4053553 | | #10

    @ Martin, thanks for your insight and links, and you have confirmed its not fungible (unfortunately) :(

    @ Dana I don't understand how dewpoint relates to this, more reading i have to do

    I checked my files, its a 10 SEER 1 1/2 Ton unit. Here is some unscientific data, measured with the $3 humidity meter, thermostat set to Celsius (2F = ~1C) and water weighed on a scale

    72% humiidty
    30C, feels like 37, dewpoint 19C - sun/cloud
    52% humidity

    1 hour
    65% humidity
    1620 mL water
    29C feels like 36, dewpoint 19C - sun/cloud
    55% humidity

    2 hours
    62% humidity
    27C, feels like 35, dewpoint 20C cloud
    66% humidity

    3 hours
    60% humidity
    26C, feels like 34, dewpoint 20C cloud
    70% humidity

    4 hours
    1394 mL
    60% humidity
    25C, feels like 33, dewpoint 20C cloud
    74% humidity

    4:40 minutes rain started

    5 hours
    60% humidity
    20C, feels like 28, dewpoint 20C thunderstorm
    100% humidity

    6 hours
    60% humidity
    21C, feels like 28, dewpoint ??? light rain
    92% humidity

    Typically i can get down to 23C in about 4-5 hours but not today. Murphys law that the first real rain in several months is on the day of testing.

    No AC run all day till the test, no water production (showers, cooking, dishes, windows opened etc) and the dehumidifier was set at 60% running in the basement (60% full basement, 40% crawlspace) and most of the gaps to outside are in the crawlspace. The AC ran nonstop for 6 hours before i gave up (using about 10kW)

  11. charlie_sullivan | | #11

    Alan, your initial indoor conditions at 7 pm had a dew point of 23.4 C, while the outside dew point was 19 C. That means you could have used open windows and/or a whole-house fan to do the initial work, for the first few hours, and then had less work for the A/C to do. Dana's suggestion, to open windows when the outside dew point is lower than the inside dew point, works well to maximize your opportunity to reduce the load on the A/C through the use of outside air. Understanding why that works is harder than doing it, so it's understandable that it's not immediately obvious.

    To find inside dew point, you can either:

    * Read a chart
    * Enter the temperature and humidity readings into a psychrometric calcuator app on your smart phone (there are some free ones that work just fine).
    * Enter the temperature and humidity readings into a psychrometric calcuator on the web
    * Get a fancier indoor weather station that displays the dew point reading directly.

    As for how it works, the dew point also corresponds to the "humidity ratio" which is sometimes called "absolute humidity" to distinguish it from relative humidity. A 23.4 C dew point corresponds to 1.8 kg of water vapor per 100 kg of dry air, whereas a 19 C dew point corresponds to 1.4 kg of water vapor per 100 kg of dry air. So flushing out 23.4 C dew point air and replacing it with 19 C dew point air will mean you are removing some of the moisture from the air inside. Then when the A/C later cools the space it will not need to do as much dehumidification.

    One of the complications in understanding how fast you can cool and dehumidify a space is that there is moisture storage in the interior finish materials in your house, and in the furniture, books, etc., and on the surfaces of them. So at first your A/C is just dehumidifying the air. But then moisture starts to come out of and off of the the stuff in the house. It's the moisture equivalent of thermal mass--your A/C is also cooling all those materials and objects, not just cooling the air. But that's probably part of why it appears that your relative humidity inside gets down to 60% and sticks there even while you continue to remove substantial moisture each hour.

    Another factor is infiltration of outside air, even with the windows closed, which will tend to raise humidity in the house whenever the outside dew point is higher than the inside dew point.

  12. GBA Editor
    Martin Holladay | | #12

    Thanks for your excellent comments.

  13. user-4053553 | | #13

    I never considered how much dew point enters into this, and i was not sure if the 60% was from "humidity loading" or from the basement even though its dehumidifier is set to 60% and its a lower temp.

    Two thoughts i'd like to mention
    On the south side of the house i have two rooms, one is about 60% of the south facing side (also 60% of the east) and the other is about 40% (and 30% of the west), the 40% one stays a few degrees cooler, i'm trying to decide because of lesser roof area (south roof crests N-S so both rooms they both get south sun all day, west side, slight extra thermal mass outside the foundation, poltergeists?

    Also i've noticed the stack effect is not helping in summer, often at night house is close to 30C, and the outside is 20C or sometimes 15, yet house is only cooled 1-2C at the most without many windows open all night, window fan or central air being used. The ceiling is not well airsealed and the attic has maybe R5 insulation between the rafters. Is the low stack effect cooling because the the delta T is too low?

  14. GBA Editor
    Martin Holladay | | #14

    At night, your internal loads are still present -- your refrigerator, your television, your water heater, and the sleeping humans.

    Houses can't cool off quickly at night unless you operate a whole-house fan or an air conditioner.

  15. charlie_sullivan | | #15

    In addition to the internal loads, there's the thermal mass of the structure and the contents of the house. Cooling the air alone is much easier than cooling all the thermal mass.

  16. Expert Member
    Dana Dorsett | | #16

    With (really?, only) R5 between the rafters the peak temperatures of the thermal mass of the house & it's contents is quite a bit higher than in an insulated house, which adds up to a long cool-down ramp to get to 22-25C from stack effect convection alone, even when it's 15C outside, even in a "low mass" house. (5x the temperature swing at 1/5 thermal mass is the same number of BTUs.) With only R5 between the rafters I'd bet your peak ceiling temps are in excess of 35C on sunny days (maybe even 40C) even on days when outdoor air temps don't hit 30C. With better insulation the peak ceiling temps will be much better moderated.

    With better insulation it wouldn't necessarily hit 30C indoors if starting out at 23C in the AM, at which point convection cooling overnight might work just fine at sub 15C outdoor temps. But a whole house fan moves a LOT more air, and can still work in an nearly uninsulated house.

    Only getting down to 20C outside overnight is usually an indication of a fairly high dew point (well north of 15C) and which is substantial latent load if the goal is 50% RH.

  17. user-4053553 | | #17

    I do attempt to reduce night time loads, phantom loads in the house are near zero, hot water tank is in the basement which is close to 15C or less anyways, the fridge uses 0.5kW for the whole night or less and i have reduced the number of humans to the bare minimum (not for heat gain purposes but all roads lead to Rome) :P

    @ Dana
    The energy auditor rated the attic as R8 whole assembly, but i never thought to check the temp of the ceiling. It has a steel roof so that probably reduced the attic temps a bit, but with an IR thermometer this past evening it read 29C, floor reading 26C, HVAC thermostat reading 27C, outdoor 28C. The IR thermometer is not calibrated so these numbers are probably not quite accurate, its the best i've got.

    Last night i also used the central air, humidity was at 90%, 3 hours later the house went down to 68% humidity and 2C less. So it really does appear that dehumidifying is sucking down the juice. I wonder how efficient a 10SEER unit is at dehumidifying (for comparison my 30 pint dehumidifier is rated at 1.85L/kWh though in practice one does not achieve this)

    If you lived in my area and i had the means i would love to get a Dana Dorsett audit on this house :)

  18. user-4053553 | | #18

    Its been 32C outside for about 4 hours, feels like 39, 49% humidity, 5:15pm
    No central air running in last two days 26C, 72% humidity

    Two rooms above basement
    In the south-east room the floor reads 26C, ceiling 28-28.5C, south wall 29C, east wall 28C
    The south-west room reads floor 24.5C, ceiling 28C south wall 27C, west wall

    Northeast kitchen above crawlspace
    Floor 25.5ish, east wall 27C, north wall 27, ceiling 28C
    Fridge operating but no cooking today

    The attached image is the attic, the rest of it looks about the same

    I don't know how to calculate the efficiency or L/kWh of water removal by the AC vs the dehumidifier, the AC is 10SEER and the dehumidifier is energy star rated at 1.85L/kW. During the experiment i described above i was extracting about 1.5L/hour with the central air, but i don't think that would be a useful for calculating efficiency because ratio of cooling to dehumidification is unavailable. I suspect the only way would be to convert 1.85L/kWh to COP and 10 SEER to COP, does anyone know how to do this calculation?

    Before i forget i wanted to thank everyone who posted thoughts on this thread, i almost decided against starting it because i expected i would get replies something along the lines of your house is poorly insulated, nothing to see here.

  19. GBA Editor
    Martin Holladay | | #19

    You seem to enjoy making measurements and recording temperatures, and that's good.

    The next step is to begin fixing up your old house. As you are apparently aware, your house has a lousy thermal envelope and an air conditioner that is at least a decade old. You're managing to limp along, which is fine.

    As you can afford it, you need to improve the thermal envelope of your old home, following the usual advice: Seal air leaks. Add more insulation to your attic. Think about improving some of your windows. Identify your oldest appliances and begin replacing them with more efficient equipment.

    You may already know enough to create a list of priorities. If you don't, hire an energy rater to assess your home and to give you a list of prioritized energy retrofit measures.

    Good luck.

  20. user-4053553 | | #20

    Thanks Martin

    As a final thought does this make sense
    Dehumidifier rated at 1.85L/kWh
    1.85L x heat of vapourization (2257kJ/Kg) = 4175.45 kJ to convert from vapour to liquid
    4175.45kJ = 1.1599kW
    COP = 1.1599kW/1kW
    Energy star rated but actually pretty poor

    COP of 10SEER central air ~ 2.9

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |