Tag Archives: thermal distribution

Insulation

10

Posted on April 7, 2014 by

The biggest difference between a cold damp cave and a warm modern earth shelter is insulation. No, the earth does not provide good insulation for your earth sheltered home.  While some estimate that a foot of moderately dry earth has an R value of ~5, it is not structurally practical to use earth for that purpose (above ground).  Instead, we see the earth as a large thermal capacitor and cover it with an umbrella of insulation.   Without the insulation, your home would be moderated to something a little less than the annual-average outside air temperature.  With the insulation umbrella, the walls and earth temperature within the envelope will eventually stabilize near your indoor air temperature, which is a lot more comfortable.

There are a number of types of insulation, as well as a number of different locations where it needs to be used.

Inside/outside

For passive solar design, we want thermal stability, which means we want as much mass inside the insulation envelope as possible.  Earth sheltered homes, particularly concrete ones, have a lot of thermal mass.  Insulating the inside and leaving most of the mass outside the thermal envelope would result in greater temperature swings and basically miss the point of earth sheltering.   PAHS or other earth sheltered “umbrella” concepts (such as my “by-passive annual heat storage”) relies on taking that concept further and including many tons of earth within the envelope.

If we assume that we are going to insulate outside of the home instead of inside, then we must rule out all the light and fluffy insulation options such as spun fiberglass (batts or blown in), wool, hemp, wood chips, etc.   Those types of air based insulation have no real structure of their own.  They can only be used in a framed wall situation where they can be kept dry and uncompressed.   Since the layer outside of the walls is often damp and very compressed, we are left with with rigid types of insulation as our only real option.

Most earth sheltered homes do have some conventional walls that are not earth sheltered.   The fluffy insulation may be applied in those locations, but spray foam insulation would do a much better job of keeping out infiltration.

 

Rigid Insulation

This brings us to the types of rigid insulation.  In traditional construction, this is used under the slab or around the basement walls.  A slightly less conventional application may be to cover shallow footings for frost protection.  Those environments are all somewhat similar to the earth sheltered umbrella concept (but the geometry is not).

rigid_insulation_annotate

There are several main types of rigid insulation.

Note that you can also purchase used rigid insulation for a fraction of the price of new…  It is worth looking for suppliers in your area.

Polyisocyanurate

Polyiso is considered to be the “best stuff” for above ground insulation.  It is the yellowish stuff with the metal foil backing.  It has the highest R value per inch (5.6 to 6.8 per inch, although many claim R7/inch), which means you can fit more R value into a given wall thickness.  It also has “green” attributes such as being mostly made of recycled materials and not including any “global warming” ingredients.

However, it is also the most expensive stuff.  For an earth sheltered application, we don’t care too much about thickness so it doesn’t make sense to pay for thinner insulation.  But more importantly, Polyiso absorbs water like a sponge and is not considered suitable for below grade use.  Just don’t even consider it.

XPS (Extruded Polystyrene)

Extruded Polystyrene is the pink stuff or blue stuff or green stuff.  This is the one you usually see on building sites, particularly for below ground applications.

To make XPS, crystals of solid polysyrene are combined with additives and a patened blowing agent.  The combination is melted under high pressure into a viscous (thick) liquid plastic state.  The goop is extruded through a die (a slot the exact thickness of the panels ) into air at standard pressure.  The blowing agent expands the foam immediately, and is trapped inside (increasing its Rvalue) as the panel is shaped by the die.  The extruder makes a continuous panel that comes out of the machine onto rollers where it cools and the sides are trimmed (sometimes cut with tongue and grove) and eventually, the boards are sliced to length.   Here is a video

The manufacturing process is what gives XPS its desirable properties; its precise size, its hogeneous structure, its compressive strength and tough skin.  The trapped blowing agent gives it a superior R value and premium price.

XPS is recommended by John Hait in his book about how to create a PAHS umbrella.   Proponents are quick to mention that it has a higher compressive strength and lower water permeability (compared to EPS).  It also has a decent R value of about ~5 per inch.  You can also buy even higher density XPS that has 25 PSI rating where you need it.

PinkPantherAndTheBuildingInspector

But the more I look into it, the more I suspect some of the difference is marketing to help sell products that are still patent protected.  Big companies like DOW (blue stuff) and Owens Corning (pink stuff) need to promote their products to compete with the much more widely available and lower cost EPS (Expanded Polystyrene).

Here is an article questioning the supremacy of XPS, and here is a website Owens Corning put together to protect their brand…  Personally, the Owens Corning arguments didn’t sway me.  Yes, they show water passing thru EPS, but it passes right out again (that could be a good feature).  The water injected into the XPS stays inside and permanently degrades it (by forcing out the blowing agent), just as the EPS people say it will.

Traits like dimensional stability, R-value per inch, and even the higher compressive strength and water resistance are not as critical as you may thing when constructing an insulating umbrella.  I explain why in the EPS section below.

If you actually expect that your rigid insulation will be submerged in water much of the time, we recommend you reconsider building an earth shelter on that site.

XPS was originally invented by DOW in 1941 (from Polystyrene which was invented over 100 years before that).  Its first application was to help float life rafts for the US Navy.  They have been promoting and defending their products aggressively ever since.  Polystyrene is less aggressively defended because it is not owned by any one company.  Competition between polystyrene producers reduces costs but leaves less money for marketing and little time for cooperation.

EPS (Expanded Polystyrene)

This is the stuff white coffee cups or packing peanuts are made of.  It can be made (steam compressed) into any shape, which is why it is also the stuff that ICF (Insulated Concrete Forms) companies use.  Its manufacture is not so limited to a handful of competitors, so the prices are much more competitive.  It is often sold without a brand name, or you can find branded versions that add features like a polylaminate waterproof facing.  It does have a lower R-Value per inch, but a higher R-Value per dollar.

To make EPS, the factory starts with tiny 1 mm polystyrene pellets.  These are expaned with steam  to 40 times their original size (pre-expanded as the blowing agent, pentene gas, escapes and is replaced by air).  The now larger pellets are cooled, dried and stored until needed.  To make anything from the pellets, they pack them into the right shaped mold.  In this video, they use a block mold because they are just making rigid panels.  The amount they pack into the mold affects the density, compressive strength, impermeability and the price.  The pellets are steam fused together.  In the case of rigid foam boards, the large block is sliced precisely with hot wires strung across the assembly line.  This second video shows how cups are made in much smaller molds.

Waterlogged?

The primary criticism against using EPS below grade is that it absorbs water more easily than XPS.  However, studies show that while it does absorb water more easily, it also dries out more easily.  Studies done on samples that were buried for over a decade show that the EPS actually retains its R-value much better than XPS.

Manified cross section of EPS foam showing the microscopic gaps between the beads where water can pass through

Manified cross section of EPS foam showing the microscopic gaps between the beads where water can pass through

EPS is basically bubbles of air surrounded by rigid polystyrene.  If the air is displaced out by water, other air can replace it when the water leaves.  On the other hand, The R-value for XPS depends on the special blowing agent gases trapped in its cells.  When that gas leaks out over time, the R-Value is permanently reduced.  In an effort to keep the gas from leaking out in the first place, XPS has more polystyrene around its bubbles.  This gives it extra density and compressive strength, but also means that once water gets in, it is very difficult to dry the XPS out again.

But the ASTM C272 test shows that EPS absorbs water easily…?   In that test, a 3 inch square of a half inch thick sample is completely submerged for 24 hours and then immediately weighed without giving the water any time to run out.  This is hopefully not what will happen to your panels.  EPS industry excavations of real EPS and XPS boards that were buried for 15 years tell a different story. www.epsindustry.org, Or here, or the technical bulletin here…

 

Crushed? Lets do some math.

Some builders are concerned that EPS has a lower compressive strength than XPS.  This is true, but the real question is how much compressive strength do you need?  Lets do some math.

What compressive strength do you need to support 6 inches of concrete (assume 150lbs/cuft)?  Lets take that cubic ft and cut it in half to get down to a 6 inch thickness weighing 75 lbs/sqft.  Then lets divide by 144 sqin/sqft to get 0.52 psi…  A lot less than the 10 to 15 psi offered by low end EPS.

Lets park a 10,000 lb forklift on the concrete, that weight gets spread wide by the concrete and rebar.  Even if we assume that the weight is all applied in a small 6ft x 6ft area (it would really spread much further), the pressure on the underlying insulation never exceeds 2.5 psi.

Now lets imagine we are talking about an earth sheltered umbrella without a concrete layer to spread the load out.  Lets imagine that we have 3 ft of earth over the umbrella (assume 120lbs/cuft).  That is 360 lbs/sqft divided by 144 sqin/sqft equals  2.5 PSI.

What if we drove a 35,000 lb bulldozer over it?  If we assume that the bulldozers treads spread its weight across a 10 ft x 10 ft section of the surface and assume that the load comes down at about 45 degrees (angle of repose), then we would be spreading that load over 256 sqft of EPS.  That comes to less than 1 psi of additional pressure.  Interestingly, the pressure would be higher if the earth layer were thinner.

 

Falling apart?

A third concern is that the EPS beads can crumble apart.  This is less of a concern with the higher density EPS, but you could certainly have thicker boards crack or break apart when trying to build your earth sheltered umbrella.  This problem only gets worse with thicker boards (they are less flexible).  I don’t have a good solution for the cracking…  But perhaps an umbrella design, featuring sheets of waterproof vinyl or plastic between the layers of rigid EPS insulation, will mitigate the problem or at least trap the cracks.

 

My Rigid Recommendation:

I would recommend not buying in to the marketing about only using XPS for underground applications.  Instead, look at the research that shows that buried EPS outperforms XPS over time, and for a lower price.  Shop around and go for the best r-value price.

I ended up with a truck load of new XPS foamular 250 sheets that I got a great discount on. However, I also plan to buy a load of used EPS insulation from another source.

Since my umbrella design is layered, I may try a hybrid design with XPS on the top layer to help distribute point loads out a little further and reduce compression of the less rigid EPS…

 

Sourcing?

One of the most pleasant surprises that I found was that I could get good recycled EPS for a fraction of the price of new.  One company offered me a truckload of more than 20,000 sqft of type 1 EPS for under $1000 (this deal evaporated later when I tried to order).    Another company offered me a 7x higher price on a truckload of EPS that was 2nd hand, but never actually used (and therefore in better shape). Google “reclaimed EPS rigid insulation” to find companies near you.

XPS is also available, but tends to fly out of the yard as quickly as it comes in.

Reading various discussion boards on the subject, such as this one, you should be careful to make sure you know what condition to expect.  The reused EPS may have some small holes or other minor imperfections, but that doesn’t matter much when you are laying it under your slab or making an earth sheltered umbrella.  From time to time, you may also be able to get “new” EPS from the recycling place.

While you are looking for recycled EPS, you could also find used billboard vinyls to build your umbrella with.

Thermal Distribution

Posted on September 17, 2012 by

Obviously I just started here… I have a bunch of rough notes and sketches and will be back 😉

 

Overview

Every heating system needs some way to distribute the heat, solar heating is no exception.  The heat can be distributed thru conduction, convection, or radiation   The heat storage medium can also be physically transported.  Various materials have various rates of conduction and capacities for heat storage.  Surface finishes can reflect or re-radiate heat.  Various configurations can lead to passive convection currents, or the head distribution could be augmented with an active system.  Here, I will try to sort it all out.

Note: since I am not very interested in forced air heating, I have separated heat distribution from ventilation

 

Direct Solar Gain

Photons are tiny packets of energy emitted from the Sun.  Some make it thru the atmosphere.  Some of these are headed toward homes.  Some of those may actually strike windows.  Of the photons of sunlight that do interact with the window glass, some are reflected and lost.  Other photos are filtered out by the low-E coatings (1/SHGC) and, if the glass is well oriented, the remainder defract (are bent slightly) and enter the home.  If the floor is not reflective, the photons strike the atoms of the floor and energize them.  The additional energy causes the atoms to vibrate faster.   If the floor is conductive, these atoms vibrate neighboring atoms and the energy is distributed.  If the floor is not very conductive, the vibration remains more concentrated at the surface.   Some of that vibration energy is always wasted (atomic scale friction) and becomes heat.  As the energy is distributed thru the floor, waste energy slowly raises the temperature of the floor.    If the energy is trapped at the surface, the resulting heat is radiated, conducted and convected into the room very quickly.

Low-E coatings have many tiny holes that allow the majority of sunlight to enter because the higher frequency waves literally “fit” thru the holes.  After heating up internal surfaces, some of this energy is re-emitted as lower frequency radiation.  These have larger wavelengths that do not “fit” thru the holes in the low-E window coatings, so they are trapped in the home.

 

In passive solar design terms, the mass of a cement floor directly in the path of the sunlight is called “primary storage”.   The solar energy to heat conversion takes place in the mass its self.  Ideally, the mass can absorb the energy at the same rate it is arriving, but this is not always the case.  “Energy backups” at the surface of the primary storage mass increase the rate of energy conversion to heat and can lead to overheating the room.  Keeping the primary storage area clear of furniture (shadows) or carpets (insulation) helps ensure the energy reaches the storage medium.  Choosing a dark surface for the cement reduces reflection and is generally considered a good idea with concrete floors, but a little reflection to secondary storage is not all bad.

One Btu (British thermal unit of energy) is defined as the amount of energy required to raise a lb of water 1F°.  Concrete has about 1/5th the thermal capacy of water, so 1Btu absorbed into 1 lb of concrete will raise the temperature of the concrete by 5F°, or conversely, 1 lb of concrete can store 1 Btu for every 5F° of temperature rise.  I personally prefer SI units.  BTU’s are energy, 1 Btu = 1.055 J (Joules), but most SI solar gain tables give power in Watts, the rate at which energy is transfered (energy over time, J/s).  To convert to SI, you must use 1 Btu/hr = 0.29307 Watts.

 Solar intensity tables published by organizations like ASHRAE often provide “isolation” in Btu/(hr·ft^2) for each latitude.

 

Since primary storage is only the mass that the sun strikes directly, the rest of the mass (typically the majority) is known as secondary storage.  Primary mass is ideal because it has fewer energy conversions between the incoming solar energy and the resulting thermal storage.  As mentioned earlier, energy that is not absorbed by the primary storage is reflected, radiated, and/or convected to the secondary storage mass.  For instance, losses on the primary surface may heat air which then drifts over to another wall slowly warming it.  This process is much slower than the direct radiant energy to conduction that happens on the primary storage surfaces.  If energy is entering the home faster that than it can be stored, there is an energy backup and the room will overheat…  Passive solar design books list secondary storage mass as having 1/10th the storage capacity of primary storage.   But of course that is not quite right.  The storage capacity is actually the same, lb for lb or kg per kg…  It is the energy backup that is the problem.  Primary storage, with its closer coupling, is 10 times faster at storing the incoming energy than secondary storage.   If your solar home is running on a very short 24 hour charge-to-charge cycle, then this difference is very important.  However, if you are running on a much longer cycle, such as a seasonal cycle, then that secondary mass becomes much more useful.  It may not absorb the energy as quickly, but it can still store it and release it as necessary.  You may even want to use lighter colors on the primary mass in order to reflect more energy to the secondary mass and distribute  that solar charge around the room (radiant reflected to secondary mass is faster/more efficient than radiant to primary to conduction and convection to secondary).

Convection, Etc.

I don’t plan on having a forced air system in my home, but I can’t ignore my ventilation needs or the fact that, like gravity,  “convection happens”.  This is certainly something I will want to work with, rather than against, and I have several designs in this direction.

Radiant Floor

Traditional radiant floor heating is the modern equivalent of the ancient roman hypocaust.  Modern radiant air floors are still available (popular with early passive solar builders, but not recommended by me), as are electric radiant floors (ideal for renovations), but I am most interested in “hydronic radiant floors”, which is the most efficient and effective method.  Water is heated and pumped thru PEX tubing embedded in the cement slab floors (wet installation).  Instead of traveling from the source by the three main methods of thermal distribution (radiation, conduction or convection), the thermal storage medium its self is transported.   Along the way, heat from the water is conducted thru the PEX and cement and comes up thru the floor.  It leaves the floor by radiant and (mostly) convective means in a way that produces a very comfortable environment (warm feet, cool head).  Other advantages include improved efficiency over other heating methods (particularly forced-air heating because it eliminates duct loses), no distribution of dust an alergens, peace and quiet, and a very flexible range of heating sources.  The heat can come from conventional burners, wood stoves, solar water heaters, heat pumps, etc.

I am very interested in the ability of a hydronic system to carry heat from one area to another.  Instead of just carrying heated water from the hotwater tank out to the floor, what if we could use it to  redistribute heat from the primary storage areas to secondary storage areas?

More to come…