Sunday, March 10, 2013

Earthen Ducts Question


My husband and I are designing a modest home in northern Indiana to passive house standards. We have a beautiful building site on a south-facing ridge in the middle of forty acres. Picture a house earth sheltered on three sides and open to the south to incorporate passive solar heating/cooling. We will not need a furnace or AC. What we do need is a well-designed whole house ventilation and dehumidification system. I’ve been researching these topics for over a year.

My conclusion: I want to explore the concept of using an earth tube or tubes to precondition our fresh air intake. If I can build a tube system that will achieve a 50% reduction in moisture load on incoming air in the dog days of summer, then tubes are the way to go. My math is not up to the task of evaluating the design. I’ve been trying to connect with a mechanical engineer who could give advice on the psychrometrics. If I can’t find someone to help me geek out the system variables for our site (tube diameter/length, depth, rate of air flow, pressure) then we will have to go with a purely mechanical system to be on the safe side. And then I will always wonder if we missed a grand opportunity.


I love the idea of Passivhaus, though haven’t been able to work on any projects yet.  Good luck on the project.
Calculating the benefits of earth ducts is probably above my math too (I’m not an engineer).  Not one of the easier systems to analyze – I was expecting something easier J   We recently looked at it for a project in Durban, South Africa and besides calculating the potential energy savings, issues to consider include risk of condensation, fan power, and cost of construction.
The ground temperature below a couple of meters is usually around the average annual air temperature for your location.  The ground temperature tends to lag the air temperature by a couple of months and the deeper you go the less the ground temp changes.  Using data from South Bend – Michiana Regional Airport, we get the graph below, showing your ground temperature average is about 50 degrees F, but ranges from 40 to 62 at 13 feet and from 30 to 72 at 1.6 feet (this was all in meters originally – so the units for feet aren’t regular).
 With enough surface area for incoming air to duct contact, you can eventually get the outdoor air almost to the same point (dry bulb) as the ground temperature.  This means that in September, if your duct was 13 feet below ground, you could get the air to about 60 degrees even if it is 90 F outside.  Conversely, you could get the incoming air as high as 45 degrees (13 foot duct) in February, even if it is 0 degrees outside.  Unfortunately, to change the temperature that much, you need longer duct runs.  Assuming you need about 55 cfm of ventilation (2500 sf x 0.01 cfm/sf + 4 people x 7.5 cfm/person), the graph below shows what percentage of the temperature difference (between earth temperature and outside air temperature) you could make up  with different diameter ducts and different lengths (I assumed square / rectangle ducts).  This says that 20 meters of 0.1 x 0.1 duct would make up 90% of the difference between outdoor air and ground temperature.  For example, when it was 0 degrees outside and 50 degrees in the ground in January, you could make up 45 degrees just from going through the duct.
It doesn’t look like fan power will be that big of a deal for you (if you’re providing 55 cfm), regardless of the distance.  When I looked at the project in Durban we were trying to move 8,000 cfm and the fan power penalty was significant.  It looks like you’d really only need a 1 watt of fan power (though you’ll probably have to get one bigger just because that is what is available).  1 x 365 days / year x 24 hours / day / 1000 Wh / kWh = 8.76 kWh per year (about $0.88 per year in fan energy).  I’m only looking at 55 cfm as the ventilation requirement – this air isn’t meant to heat or cool. 
Finally, the last thing you’d want to consider is condensation.  The psychrometric chart below shows one dot for every hour of a typical year (8,760 dots) in South Bend.
The same chart below shows just the dots for a typical August.  During this month, your ground temperature could be around 60 degrees (assuming 13 feet deep).  During the hour that I’ve highlighted (point 1), the air temperature is about 87 degrees, 70% Rh.  When you cool this down towards 60 degrees, you start to get condensation at about 76 degrees (point 2 – the air is now 100% Rh).  By the time you get to point 3, you will have gone from 0.019 grains of water per pound of air to 0.011, with the water that is no longer in the air condensing out in your ducts.  I honestly don’t know how big of a concern this is, but in a typical air conditioner you’d collect the condensation at a specific point and be able to deal with it.  There may be concern about bacteria if you have standing water running the length of your ducts during the summer.  You wouldn’t have to worry about this in the winter.
That’s all I have for now.  If you have specific questions about your system, I might be able to do the calculations for you.  I made a few assumptions in the calcs above (6 bends in each duct work, turbulent flow, concrete ducts for example) and adapted calculations we already had for a larger commercial system, so if we had a real system and started from scratch the numbers could be tightened up  a bit, but I think this info is close.  Thanks to Alejandra Menchaca, PhD and Director of Operations for the Boston offfice of EA Buildings, who did most of the original calculations for our project in Durban.


Wednesday, March 6, 2013

Ground Source Heat Pump Question

Wed, 6 Mar 2013

Hi Nathan,  I am writing about our house in Lexington which my son and family are living in. He has contacted a firm who is planning to install heat pumps to heat/cool 2 floors of our split 3-level house. It is estimated to cost $20,000 with an interest free loan. I just am somewhat skeptical that heat can be supplied reasonably cheaply with electricity and heat pumps in this climate. I wonder if you are able to send me to appropriate literature or home owners who are using this equipment or perhaps give me your opinion. Sam



I usually think of GSHPs in terms of their coefficient of performance (COP). A good one can get around a COP of 4, meaning for every one unit of energy (electricity) you put in you get 4 units out as heat. According to the 2009 Residential Energy Consumption Survey ( the average house in Massachusetts uses 65,200,000 Btu of energy on heating per year (652 therms about $652 cost). Since most homes use a regular boiler or furnace of around 80%, I'd estimate they only need 52,192,000 Btu of heat (the other 20% is inefficiencies from the boiler). This is the same energy as 15,296 kWh. To get this from a GSHP with a COP of 4, you'd need 3,824 kWh of electricity (about $382 cost). Conversely, if you provided the same 52 MMBtu of heat with a 95% condensing boiler, you'd use 54,939,000 Btu of natural gas (549 therms and $549 cost). GSHPs will save some energy, but the payback is usually pretty long.

As to whether or not the GSHP will work in Boston, it definitely can. We used a lot at Harvard. I know a guy in Somerville who has one for his house (case study link below). Average ground temperature in Boston is about 50. At shallow depths it ranges between about 30 to 70 (see attached from Logan). The ground temperature usually lags behind the air temperature by a month or two (the coldest months for ground temp are a couple of months behind the coldest air months). Still more efficient than trying to use an air source heat pump where the delta T is greater and less in your favor.

The GSHP will also be more efficient than a typical air conditioner, but the annual air conditioning cost is less than the heating cost. Assume you could save 25% or so on your AC bill (say going from a COP of 3 to 4).

I hope this helps.