Some Notes on Earth Tubes

Update -- REHAU Earth Tube systems -  May 22, 2008 ...

The full Earth Tube section is here ...


First, please read this.


These are just some thoughts on earth tube design based on nothing more that reading the articles listed in this section and a few others.  If you are going to use earth tubes to cool or heat your residence, you are in somewhat uncharted territory, so do your homework carefully.


If you think I have gone astray on any of the notes below, or have some something to add, please contact me.



There seems to be a consensus in the articles that earth tubes do not work well in hot humid climates without some form of dehumidification prevent water from condensing in the earth tubes.


How Long?

The longer you make the earth tube, the closer the air flowing through the tube will get to the earth temperature.

But, there is a diminishing returns effect as the tube gets longer and longer. 

Somewhere around 75 to 100 ft might be pretty close to the sweet spot. 

The temperature change in the tube air after 150 ft appears to be small.

As the tubes get longer, the cost to install them goes up, and the fan energy to drive air through them also goes up.


How big?

The various installations use tubes anywhere from 4 inches on up to ones you can walk through.

For residential applications, somewhere in the 4 to 12 inch category is probably sensible.

Obviously, multiple 4 inch pipes would need to be used to get the same cooling as one 12 inch pipe.


What Material?

The articles use everything from thin wall plastic sewer pipe up through large concrete pipes.

Metal tubes are also used. 

I don't think that the lower conductivity of the plastic pipe is likely to effect performance much given the fairly low conductivity of the inside air film and the relatively low conductivity of the dirt.


How Deep?

The Hait book states that as a general rule, the ground temperature at 20 ft down does not vary over the year, and that it is equal to the average yearly air temperature at that location.  This probably varies with the soil type, moisture content, and ground cover, but may be generally helpful ground rule.

The Sharan paper did a careful survey of ground temperatures in their location and found that at 3 m (10 ft) the yearly variation in ground temperature was about 2.5C (4.5F) -- they felt that 3m depth was a good compromise between cost and yearly temperature variation.


I have heard people say things like "the ground temperature 4 ft down is a constant 50F all over the world"  -- wrong on all counts :)


Free energy?

It seems to me that if you want to extract a useful amount of coolth from the ground you will likely need a fan driven system.  Fans use power, so the cooling you get is not really free.  The Sharan paper reports COPs that range from about 2 to around 5.  This means that for each KWH of electricity you use to drive the fan you get 2 to 5 times that many KWH in cooling (or heating).  This is respectable, but not free  -- a good AC might do about as well?   It may be that these COPs can be improved with careful attention to fan selection and duct velocities.


One exception to the above is the paper that discusses earth tubes that are driven by a solar chimney.  This appears to work, but is not capable of producing as much flow as a fan system, and incurs the extra costs of the solar chimney.


I guess that another solution would be to use a PV powered fan.


Cooling Output:

Using the hourly data reported in the Sharan paper as an example:


Summer cooling test (May -- India)

Time = 2pm

Tube length = 50m = 164 ft

Tamb = 40.8C = 105.4F

Tground = 26.6C = 80F (10 ft under surface)

Toutlet = 27.2C = 81F

Velocity = 11 m/sec = 2145 ft/min

Tube diameter = 10 cm = 3.9 inches

Tube area = 0.083 ft^2


This gives a volume flow rate of 178 cfm, and transit time through the tube of about 5 seconds.

This fairly aggressive velocity is achieved using a blower that consumes 300 watts.


Cooling = (weight flow)(specific heat of air)(dTemp)

= (2145 ft/min)(0.083 ft^2)(0.065 lb/ft^3) (105.4F- 81F) (0.22 BTU/lb-F) = 62.1 BTU/min or 3427 BTU/hr or 0.31 ton of AC


So, that's a fair bit of cooling for one 4 inch tube, but bear in mind that the velocity used in this test was on the high side, which makes the weight flow and cooling high.

For many parts of the world, the ground temperature at 10 ft deep would be substantially less than the 80F where this test was done, so the output air temperature from the earth tube would be correspondingly less as well.



Ground Temperature change?

None of the papers mentions a change in ground temperature as the earth tube extracts or adds heat to the ground.  It appears that getting reduced performance over time from an earth tube because the tube is heating or cooling the ground is not a problem?


Update -- May 22, 2008:


The REHAU company makes a whole line of components for building earth tube systems that can provide ventilation air that is cooler in the summer and warmer in the winter.

REHAU Air - Ground Heat Exchangers ...



This company makes components for earth tube heating and cooling systems.

The downloadable pdf provides some useful information for designing earth tube systems.

(thanks to Charles for finding this)


Here are some of the design guidelines that the REHAU Air-Ground Heat Exchanger System guide provides:

Notes on Laying

- Pipes must be laid at a depth of at least 1.5 m.

- For best possible heat transfer, pipe should be laid in solid ground (not laid in sand). Make sure the ground is well compressed around the pipes.

- The pipes are to be laid at least 1 m from the building and from each other.

- The fall of the pipes is to be approx. 2 %.

- The pipes should be cut using a fine toothed saw or pipe cutter. Pipes are to be cut square and the ends are to be de-burred and slightly chamfered.

- Seals are to be cleaned and checked for damage before being inserted. The chamfered pipe end is to be coated with REHAU lubricant and the pushfit socket is then inserted.

- The pipes should be installed similar to BS EN 1610.

Much more helpful information in the guide. 


If you have one of these systems, I would love to hear from you.


My thanks to Charles for finding this.





7/24/07, updated May 22, 2008