DIY Solar Collector: Optimizing the Thermal Conductivity of the Fin to Riser Tube Joint
I've received a some emails wondering if the silicone caulk used between the aluminum fins and copper or PEX tubing is really a thermal benefit, and suggesting that perhaps the silicone isolates or insulates the riser pipe from the aluminum fin, and may do more more harm than good. Others have have suggested that something other than silicone caulk with greater thermal conductance might work better to fill the gap between the aluminum fin and the riser tubes. I've done some thinking, testing, and calculating on this issue, and the raft of stuff below goes over all this. My thanks to all the people who made these suggestions!
From my knothole, the bottom line is that the silicone caulk remains a good choice because it shows good thermal performance and has several other very desirable characteristics. And, that any thermal improvement resulting from substituting a more conductive material would likely be small. But, please read the material below, and see if you agree -- or maybe have another approach to suggest?
If you are wondering what in the world this is all about ...
The reasons I've included the silicone caulk between the fin and tube on collectors built to date are:
The silicone caulk fills any
small gaps between the aluminum fin and the tube that would otherwise be filled with
air. The silicone has a
thermal conductance around 0.25 w/m-C and air has a thermal conductance around 0.025
w/m-C -- so the silicone is ten times more conductive than air, and I
believe that this results in better performance. The
small panel tests comparing air in the gap vs silicone caulk in the gap (see
to support this. The calculation of the thermal resistance of a very
small gap filled with air or filled with silicone caulk (below) also tend
to support the idea that 1) the gap filled with silicone will have better
thermal performance, and 2) that the thermal benefit of a more conductive
material than silicone caulk will not be large. The one caveat here is
that the amount of silicone must not be excessive, and must not result in a
larger gap because any excess silicone is not squeezed out -- This is
discussed below, but I don't thing that
its a problem in practice.
When you staple the grooved
aluminum fin over the tube very tightly, you should get a very small gap to
no gap between the fin and the tube. The silicone caulk is a
good adhesive, and will tend to keep this small to no gap condition for the
life of the collector. It basically glues everything into place and
keeps it there. If you don't put the silicone in, the thermal
expansion and contraction may work the staples over time and may allow a
larger gap to develop between the fin and tube over time -- mind you, I
don't know that this would happen, and it may not.
Silicone caulk is a very good
material to use in collectors -- it has a long track record of standing up
to the very difficult environment inside of collectors for a very long time
without degradation. Many materials don't do well inside of
Silicone caulk remains
flexible after curing, so it can handle the differential thermal expansion
between the fin and tubing.
Silicone caulk acts as a barrier
to prevent galvanic corrosion between the aluminum fin and the copper pipe
-- it prevents the entry of water from condensation into the area between
the fin and the tube and it physically separates and protects the aluminum
and the copper. It adds to the protection offered by the paint finish.
Silicone caulk is cheap, safe, and readily available.
This sounds a bit much like a commercial for silicone caulk, but (honest) I don't sell it make it. I do think we are lucky to have such a good material for use in collectors, and that it is cheap and easy to find.
The suggestions for improvement include:
Leave out the silicone caulk on
the idea that it does more harm than good. An offshoot of this is that
the silicone caulk may result in a larger gap between the fin and tube if it
is not well squeezed out.
Use a thermal grease or heat sink
compound that has a higher thermal conductivity than silicone caulk.
Us a silicone caulk that has been formulated for high thermal conductivity. There are some specialty silicone caulks that use fillers to achieve higher thermal conductivity.
On item 1, I don't see the merit of this. The test and the calculation below both indicate that a silicone caulk filled gap will have significantly better thermal performance than an air filled gap. The test panel and calculation indicate an about 2 to 6% improvement in heat output for collectors with a silicone filled gap over an air gap.
Based on some informal testing and sectioning of one of the test panels, I don't believe that the silicone results in a thicker gap as long as you take some care in the assembly. This is basically because the excess silicone can be squeezed into the area between the grooved fin and the flat aluminum base plate -- that is, the excess silicone does not have to travel to the ends of the fin groove.
On item 2, I am leery of using a non-curing thermal grease or heat sink compound. I don't think that these materials would glue the fin and tube together as the silicone caulk does, and I think that having the fin and tube glued together is important to prevent the gap between the fin and tube from increasing over time as the collector is worked by thermal expansion and contraction (of course, I may be totally wrong about this :) In any case, the calculation says that the gain in collector efficiency for using a more thermally conductive material would be small.
On Item 3, I do think there would be a small gain (see calculation below) if one could find and obtain a high thermal conductivity silicone caulk. I've asked for a sample of CHO-THERM 1641, which is such a caulk, and will test it if the sample comes through. Based on the calculation below, I would not expect to see a large improvement in performance, but I'd like to see how the test comes out. If anyone has a good source for a high thermal conductance silicone caulk, please let me know.
Update: May 2011 -- here is result of testing a high thermal conductivity silicone ...
The aim of this test is to compare the performance of a collector with silicone caulk in the gap between the fin and the riser tube to one with air in the gap.
Click on pictures for full size
Red line -- reservoir temperature for the panel with the silicone filled gap
Solid black line -- reservoir temperature for the panel with the air filled gap
Green line -- ambient temperature.
Both reservoirs filled with 29 lbs of water.
Absorber area for both collectors is 2 sqft.
Glazing is SunTuf corrugated polycarbonate.
Back insulation is 2 inches polyisocyanurate foam board.
Reservoir temperatures are logged with an Onset Computer U12 logger using 10K thermistor temperature sensors immersed in the reservoirs at the pump inlets.
Flow rate is about 0.4 gpm (relatively high to make sure flow does not limit heat transfer and to keep the reservoir well mixed).
The pumps are small (3 watt), submersible fountain pumps that are located in the reservoirs.
This is a small panel test in which the performance of two small collectors is compared. One collector uses aluminum fins and copper tubes with a layer of silicone caulk between the fin and the tube. The 2nd collector uses the same construction with no silicone between the fin and tube. Each collector heats a 29 lb charge of water in its own insulated reservoir. The reservoir water is pumped through the collector using a small submersible pump. The collector that raises the temperature of it's reservoir the most is the better performer. The idea is that since each collector is the same size, is heating the same amount of water, and is exposed to the same sun, the only difference in reservoir temperatures will is be due to the difference in the construction of the two collectors -- in this case, the fin to tube gap being filled with air or filled with silicone.
The performance plot shows the silicone filled gap collector starting about 1F below the air filled gap, and by the end of test, the silicone filled gap collector reservoir is about 1F warmer than the air gap reservoir.
|Collector Construction||Start temp||End temp||temp Rise||% of base|
|Test1: Silicone filled gap vs air filled gap|
|Copper tubes - Aluminum Fins -- Air in gap||76.07 F||127.60 F||+51.53 F||base|
|Copper tubes - Aluminum Fins -- Silicone in gap||75.20 F||128.00 F||+52.80 F||+2.5%|
So, the panel with the silicone caulk filling the tiny gap between the copper tube and the aluminum fin showed about 2.5% better performance. I did two similar tests with the collector that uses PEX tubing and aluminum fins. The results were similar with about 4% better performance for the silicone filled gap collector.
This is an analytical estimate of the performance improvement you might expect to see for the silicone filler in the gap between the tube and fin:
Thermal Conductivity of air = Kair = 0.025 w/m-C
Thermal Conductivity of silicone caulk = Ksil = 0.25 w/m-C
Consider a 1 meter long segment of aluminum fin and tube:
Solar collection area for this 1 m of fin and tube = (0.152m wide)(1 meter long) = 0.152 sqm
Solar Radiation on 1 m of fin and tube = (1000 watt/sqm)(0.152 m^2) = 152 watts
Assuming a typical 50% collector efficiency, the solar radiation going into 1 meter of tubing = (152 watts)(0.5 efic) = 76 watt
So, the 1m of fin must transfer 76 watts to the tube through the air or silicone filled gap. What is the temperature difference across the gap that is required to transfer this 76 watts?
Heat transfer area - fin to tube = (1.580 cm)(Pi)(100cm) = 496 sqcm = 0.0496 sqm = Aht
This is the heat transfer area from the fin to the tube that the 76 watts must be transferred through.
Assume that the average thickness of the gap is 0.004 inches = 0.0102 cm = 0.000102 m = Tgap
(equal to a thin sheet of paper).
Heat Transfer across the gap = Qgap = (Aht)(dTgap)(K/Tgap)
Where dTgap is the temperature difference across the gap -- rearranging for dTgap:
dTgap = (Qgap) (Tgap) / ((Aht) (K))
So, for the silicone filled gap, the temperature difference required across the gap to transfer 76 watts is:
dTgap = (76 watt) (0.000102 m) / ((0.0496 sqm)(0.25 w/m - C)) = 0.63 C = 1.1 F
If you use this collector efficiency calculator, and assume typical collection conditions (60F ambient, 120F storage temps), then the collector efficiency drops from 52.3% down to 52% -- a drop in efficiency of 0.3% -- this is the result of increasing the absorber (fin) temperature by 0.63 C (1.1F) to overcome the thermal resistance of the 0.004 inch thick silicone caulk layer. This is equivalent to a 0.6% drop in heat output from the collector compared to a a collector with a perfect fin to tube connection.
With air in the gap instead of silicone caulk, the temperature difference across the gap would have to be:
dTgap = (76 watt) (0.000102 m) / ((0.0496 sqm)(0.025 w/m - C)) = 6.25 C = 11.2 F
Using the collector efficiency calculator, the efficiency would drop from 52.3% down to 48.6% -- a drop of 3.7%. This is a bit more than the 2.5% result from the test above, but generally in agreement.
So, both the test panel and the calculation suggest a gain in collector output in the area of 2 to 6% for filling the gap between the fin the tube with silicone caulk. In addition, the caulk bonds the fin and tube together tightly so that there is no tendency for them to separate as thermal expansion and contraction work the collector over time.
If one were able to buy and use a thermally conductive silicone caulk like this CHO-THERM 1641, the temperature drop across the gap would be reduced by a factor 3 to about 0.2 C (0.4F), and the efficiency drop to overcome the thermal resistance of the gap would be reduced to only 0.1%. This would be a gain in efficiency over the plain silicone of 0.2%, and a heat output gain of about 0.5% -- this is in the nice to get category, but probably not worth paying a great deal more.
I've asked for a sample of the CHO-THERM 1641, and if they come through, I'll do a small panel test with it.
The calculation assumes that the presence of the silicone in the gap does not make the gap any thicker. It has been suggested that the gap might be thicker with the silicone because of its resistant to being squeezed out. After some playing around with some short sections of tube and fin to see how well the silicone squeezes out, I don't think that this is a problem as long as firm pressure is applied when stapling. Any excess silicone only has to work its way down to the bottom of the tube where it is easy for it to be squeezed out between the bottom sheet and the fin. I also sectioned one of the early test panels with a saw at 4 inch intervals, and did not see any evidence of the silicone resulting in a larger gap. You do need to use moderation in applying the silicone to the fin groove, and to use lots of pressure when stapling so that any excess silicone is squeezed out. It is a good idea to just make a test section a couple feet long, then let it cure for a day, and then cut across it in a couple locations to see if you were successful in keeping the gap small. Spraying a mist of water into the air around and in the fin gap before applying the silicone will allow the silicone to cure more quickly.
Gary April 7, 2009