Solar Shed -- Heat Transfer Trench

Heat is transferred from the storage tank in the Solar Shed to the house via two 3/4 inch CPVC pipes.   The pipes are buried in a 4.5 ft deep trench that runs from the shed to the house crawl space.  The pipes are insulated to around R15 by enclosing them in rigid polystyrene foam board (see pictures).   The pipes are installed in the trench by 1) laying down a 3 inch bed of screened gravel, 2) laying the insulated pipe assembly on the bed of gravel, 3) filling around the pipe with more screened gravel, 4) adding a covering 6 mil poly on top of the gravel, and 5) backfilling dirt to fill the trench. 


I used extruded polystyrene ("pink" foam) because it is used routinely for foundation insulation, and (I believe) will hold up to the outside environment without deterioration or excessive moisture absorption.  The thinking behind the gravel is that it provides a dry environment for the pipe/insulation assembly to live in, and may provide some additional insulation value.


The pipes are buried 4 ft deep to ensure that they are below the frost line, and to take advantage of the somewhat warmer ground temperature as you go down.  I may have gone a bit overboard on this -- one concern was that the shed is about 120 feet from the house, and I wanted to make sure that there was not excessive heat loss on the trip (see calcs below).



Digging the trench with a rented mini-excavator, and gaining a whole new respect for the skills of a good backhoe operator.


  Note: More methods for insulating underground pipe runs here...



Placing the CPVC pipe into the polystyrene foam boards.  This is the bottom board.  It has two grooves dadoed into it to accept the two pipes.   Polyurethane foam ("Great Stuff" cans) was then applied to the top of this foam board, and a 2nd foam board with the grooves facing down was place over it.  The  foam glues the two foam boards together, and seals any gaps.  The strings are to tie the two halves together while the foam is curing.  The foam expands as it cures, so the boards need to be held together while it is curing.  

An alternative to the dadoed grooves would be to use 1 inch thick foam to space the top and bottom foam boards far enough apart to run the two pipes between.


The foam boards are extruded polystyrene (the pink or blue stuff), and are 8 inches wide.  The cost for insulating the two pipes is about $1.30 per foot.  I think it does at least as well as some of the $7 per foot commercial schemes.


Scraps of one inch thick foam board are foam glued over the top of the two two inch thick foam boards just to provide some more insulation.   This method of placing boards over the foam boards, and weighting them with rocks while the liquid foam is curing is easier than the tie strings in the previous picture.



Insulated pipes placed in trench with gravel under and over.  A layer of 6 mil poly was placed over the gravel before backfilling the trench.  The wires are for some sensors that will monitor the performance of the system.  You will probably want to include wires for the sensors that your differential controller will need, so that the controller can be mounted in the house.  You may also want to include an AC  power line to run the pump -- check your codes on this.  Be sure to do a pressure and flow test at this stage to make sure the pipes don't leak, and have no blockages -- its a lot of work to dig the trench -- you don't want to have to do it twice.


I used CPVC pipes because I do not expect the storage tank temperature to go much above 160F, and this is well within the capability of CPVC, especially given that the pressure in the system will be low.  The CPVC will be living in a fairly friendly environment with the foam board and gravel surrounding it.  PEX would be another candidate, and (I think) would also work well -- I picked the CPVC over the PEX mostly because I thought it would be a hassle to install a big coil of PEX in the foam board grooves.



To get the pipe line through the concrete foundation wall, I rented a rotating hammer drill with a 3 inch core bit. 

I drilled three holes:  one for the two Solar Shed pipes, a 2nd for the some wires that go from the house to the Solar Shed (these are for sensors, and maybe pump power), and a 3rd for the eventual horizontal pond solar water heater project.  Dusty, but fast.









Picture from inside the crawl space. 


Picture from outside where the trench comes into the house --->












Heat transfer capability:

I wanted a sufficiently large pipe to be able to transfer all the the heat that can be collected in a day and/or stored in the 500 gallon storage tank to be transferred to the house in a reasonable time:


Day heat collection:

Qday = (1900 BTU/day-ft^2)(240 ft^2)(0.5 efic) = 230K BTU/sunny day

Where 1900 BTU is the solar radiation incident on 1 sqft of a 70 deg sloped surface in Nov on a sunny day at 46 deg N latittude.

Storage Tank Capacity:

Qstor  = (500 gal)(8.2 lb/gal)(150F - 90F)(1 BTU/lb-F) = 250K BTU


For a 3/4 inch pipe:

There is about 250 ft of pipe (round trip)


At 3 gpm, the head loss for 250 ft of 3/4 inch pipe is (2.5)(2.19 ft/100ft) = 5.5 ft of head loss

This seems easily in the range of the common Grundfos and Taco ciruclation pumps.


The heat transfer rate at 3 gpm with a 10F temperature drop in the house radiators is:


Qhouse = (3 gal/min)(10F) (60 min/hr)(8.2lb/gal) (1 BTU/F-lb) = 15K BTU/hr


Or, about 230K/15K =  15 hrs to transfer a full sunny day of collection to the house. 

Seems about right?



The heat loss rate of the house is about 400 BTU/hr-F, so at 30F outside:


QhouseLoss = (400BTU/hr-F)(65F-30F) = 14K BTU/hr

Or, about the same as the pipe can easily transfer.  I hope to get the house loss late down to more like 300 BTU/hr-F with some improvements in insulation and further weatherizing.


Heat loss rate from the pipe:

The heat loss while fluid is flowing through the pipe is relatively small with the R15 pipe insulation:




When the pump turns off, the fluid in the line starts cooling off.  If you assume (at worst) that it cools down to around ground temp, then you lose the heat that was stored in the pipe fluid:


Qpipe = (250ft) (0.02gal/ft) (120F-50F)(8.2lb/gal)(1BTU/lb-F) = 2900 BTU

Or, about 1% of a sunny days collection.

Not too bad(?), but, It might be better than this due to the two pipes being enclosed together, the insulation offered by the gravel and ground, and the time constant of the cooling fluid.




1/15/07 update

The temperature sensor that I buried in the trench near the pipe has settled down to about 38F to 40F -- a bit colder than I expected at 4 ft down.  When the pipe from the storage tank to the house is used to transfer water for several hours, the trench temperature goes up a couple degrees, and then drifts back down to about 40F in 10 hours or so.



Gary Updated 10/25/06, 1/15/07