Heating Downstairs Versus Upstairs

I’m seeing strong evidence that over 80% of the heat needed for two-story buildings, like many homes, is used downstairs.  And in some cases, it’s 90%-100%.   I now have data for several buildings, and it all says the same thing:  The majority of heating that happens in two-story buildings happens downstairs. I’m specifically referring to buildings with an open stairway between the first and second floors.

If this is correct, there are implications for energy use in both existing buildings and in new buildings.  For new buildings, we could install high-efficiency systems (like heat pumps) ONLY downstairs, and install something cheap and inefficient like electric resistance heat upstairs.  The installation cost will be much lower than heat pumps throughout the house, and the energy savings and carbon emissions will be almost as high.  Similarly, for existing buildings, we could just install efficient heating systems downstairs, at a more affordable cost than a whole building, and reap substantial energy benefits or reduction in fossil fuels and carbon emissions.  It also helps to explain why the upstairs of buildings often overheat, resulting in discomfort, open windows, and more.

I first noticed this a few years ago, on a research project at a house in Buffalo, NY.  In that house, a full 89% of the heat was used downstairs.  I thought it was an anomaly.  Over subsequent years, two work colleagues at Taitem Engineering mentioned seeing this phenomenon in two different buildings.  The pattern caught my attention, but I did not have a chance to look into it. It’s only recently that the issue has come up again, as we try to affordably convert buildings heated with fossil fuels to heat pumps, and I began to wonder if we could just convert the downstairs heating system, and achieve most of our goals, without the high cost of converting a whole house.  The issue has also come up for new buildings, as I have seen examples of people just putting heat pumps in living rooms, and putting electric resistance heat in bedrooms.

Why would two-story buildings use almost all their heat downstairs?  Well, we know that the stack effect carries warm air upstairs:  warm air rises.  The stack effect also draws in most of a building’s infiltration downstairs.  But could this explain how as much as 90% of the heating in a building would be downstairs? I did a quick back-of-the-envelope calculation.  Over a wide range of typical infiltration rates, even if all the infiltration happens downstairs (and serves as a load on the downstairs heating system), and even if all this air rises at a significant warm temperature, up to the second floor (to deliver warm air upstairs), I can only explain a 60%-40% split between heat downstairs and upstairs.  I believe that something else is going on.  It does not explain an 80%-20% split, or a 90%-10% split.

How about conduction up through the first floor ceiling?  I found some good upstairs/downstairs heating data for a two-story duplex, without a connecting stairway.  The split here is 53% of the heat downstairs, 47% upstairs.  So, yes, there is some conduction.  But it’s not much.  I believe that something else is going on.

Here’s my guess:  In addition to stack effect warm air rising, and in addition to conduction, I believe that there is a third phenomenon, and in fact this one may be the dominant one.  I believe that what’s going on is cold air falling, from the second floor, down the stairs, to the first floor.  This falling cold air is accompanied by an equal amount of additional warm air rising, but the additional rising warm air is not what we think of as traditional stack effect airflow – it does not leave the building.  It rises, enters the second floor rooms to heat them, and then recirculates back down the stairway to the first floor.  It is in a way similar to the thermosiphon effect that is used for passive solar hot water systems used in warm climates, or to single-pipe steam systems (without the two-phase flow), or to heat pipes. The fluid (air) rises and falls in the same single “pipe” (the stairway).

If that’s the case, then a smoke test should show cold air spilling down the stairs, at floor level, and warm air rising, at ceiling level.  So, I brought a smoke generator home from work, and looked at airflow at the top of my stairs at home.  Bingo.  See the photo.  I took the smoke generator to some other two-story buildings:  Same effect everywhere.  Cold air tumbles down the stairs at the level of the stair treads.

By observation of the smoke velocity, and by analysis to match the heating split of 80%-20% or 90%-10%, I believe this recirculating airflow amounts to hundreds of cubic feet per minute, in addition to the rising-only stack effect that is associated with infiltration.

If this effect relates primarily to recirculation upstairs/downstairs (warm air rising and cold air falling), then we should see it EVEN in a building that has very little infiltration, and so very little upward-only stack effect.  Right? So, I ran the test on a very tight building that is designed to be a net-zero energy user.  The results?  Well over 80% of the heating is happening downstairs.  This confirms that the dominant downstairs heating is not just the result of infiltration-related stack effect.

I’m running some more tests and will publish the findings later in the winter or spring.  A key question is “What is the effect of open or closed doors in the upstairs rooms?”, so I’m looking at that.  I’m also looking at the impact of heating thermostat temperature settings, upstairs and downstairs.  A first test showed that if the upstairs heating is set just a few degrees lower than downstairs, the upstairs heaters never run – ALL the heat is provided by the downstairs systems.

More to come.  We need a little more evidence.  But, again, if this is correct, we can begin to think of all the implications.  It means we absolutely MUST zone two-story buildings so that upstairs and downstairs heat are temperature-controlled separately, otherwise we are doomed to overheating our upstairs. We must also re-think the sizing of distribution systems.  And we’ll have interesting discussions about whether to compartmentalize the two floors of new high-performance two-story buildings, yes or no?  Maybe we keep those nice open stairways after all, and take advantage of this phenomenon, with heat pumps downstairs, and nice small strips of electric heaters upstairs.  Maybe we direct added attention to reducing heat loss upstairs (better windows, more insulation), so that an efficient downstairs heating system can indeed reliably handle the full upstairs load.  The potential for reducing the cost of eliminating carbon emissions in homes is great.  So much to think about!

For now, if you’re not convinced, do the following ten-second experiment.  Wet a hand, and reach down toward the top step of the stairs in your home. What side of your hand feels cold?  Now raise it toward the ceiling.  What side feels cold now?  And I’m happy to lend you my smoke generator if you want to not only feel it, but see it for yourself.



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s