Many of the homemade
collectors shown on this site use aluminum fins to collect the incoming
solar heat and transfer it to a tube that carries the heat transfer
fluid. The fins typically have a groove of some type that fits
over the pipe to increase the heat transfer area from the fin to the
grooved aluminum fin wraps over the copper pipe
This page summaries the various ways that people have gone about making these fins, and also provides some places that you can buy commercial fins.
The page also provides some construction tips, and provides the calculated performance (fin efficiency) for several designs.
Directory for this page:
Homemade Fins -- several designs...
Bought fins -- several types and sources...
Construction Tips -- a few tips on making good fins...
Fin Efficiency -- the performance of homemade and commercial fins...
You can do a good job of making collector fins in your shop. Here are several methods.
This is the method I've used for the last couple collectors I've done. It seems crude, but it actually works well, and makes a nice fin.
Click on pictures for full size
The tool consists of a plywood form with the groove shape in it, and an ordinary 5/8th inch diameter solid steel rod.
The form is made from scrap plywood. The base is about 3 ft long and made from 3/4 inch plywood.
The slot that forms the groove is made from two pieces of 5/8ths plywood setup so that there is a 5/8 inch wide slot formed between them. These two pieces should be carefully glued and screwed to the base -- they take a beating during the forming process.
I've been using Rollex System 3 aluminum roof soffit material for the fins. I cut them to size on a table saw, but they cut easily with any kind of saw or shears. This brand comes with some shallow preformed grooves already in the right place, and I just put that groove into the slot in the tool, lay the 5/8 ths steel bar into the preformed groove, and bang on it with the sledge hammer. I hold the sledge as shown in the picture above, and move the handle straight up and then straight down. Basically, bang on the fin with the top the hammer, not the head of the hammer. Using the sledge in this way keeps the fin flatter.
You can use one foot to keep the steel rod in place for the first few swings.
Once you get the hang of this, its quite fast -- maybe 15 to 20 seconds per fin.
Be sure to check the first few fins for good fit -- make any adjustments that are required to get a snug fit.
If you were starting from flat stock aluminum, you would want to add some vertical pins (eg nails) to the form on both ends of the form to guide the steel bar -- one pair of pins at each end of the jig. This will keep the steel rod centered over the slot in the plywood form. It would also be good to add a stop along one side of the form to position the fin so that its centerline is over the grooving slot.
When installing these fins over the copper tubing, its very important to really bear down on the stapler so that the fin gets pulled down into tight contact with the copper all over. The fin to tube joint can be a thermal bottleneck if you don't pay careful attention to maintianing good tight contact over the full joint area. There is more info on the absorber assembly process at this link More detail on the fin making on this page...
Fins made with the sledge hammer method have been tested against copper fins continuously soldered to copper pipe and they do quite well... The key is to maximize fin to copper tube contact area, and to keep the gap between the fin and the copper tube zero to paper thin. Except as noted, I think that all of these fins designs will deliver good thermal performance.
I built this fin press on my bench when I was making a lot of fins for my radiant floor heating. Its a bit of work to get it setup and working right. In the end, it makes fins of about the same quality as the sledge hammer method. For one person, its really not that much faster than the sledge hammer method, so I don't see much benefit in going to the work of building this.
If you had tons of fins to do, and you could have one person pushing down on the press handles and a 2nd person feeding and removing the fin sheets into the press, you could do a whole lot of fins in an hour.
Click pictures for full size
More detail here...
Kevin worked out this method that uses two thinner aluminum layers to make the fin -- one under the riser tube, and one over the riser tube.
He used two layers of the 0.01 thick and 9 inch wide aluminum flashing material that is commonly available in hardware stores.
The first layer goes down flat on the plywood backing, and the 2nd layer (which is grooved using something like the sledge hammer method) goes over the top of the riser tube. So, you end up with 0.02 thick fins and with aluminum in contact with the copper tube for nearly the full circumference. Since the tubes are spaced about 6 inches apart, and the aluminum fins are 9 inches wide, they overlap each other on the outer edges -- this probably improves performance a little bit. I think the performance of this method should be quite good, and the materials might be more readily available to you.
Full details on Kevin's design here...
Henry worked out this way to make a fin that fully wraps the copper tube.
Details on the full wrap process here...
It seems to me that this fin design will perform well, but it does look like it adds a fair bit to the labor side.
Matt worked out what is clearly the coolest way to make the fins. He converted a bicycle to drive a forming roller to roll the groove into the fins.
|Both Matt and Woodsy mounted the fins on the bottom of
the copper tubes such that the part of the tube not covered by the fin
is exposed to the sun -- as shown in this picture. The metal clip
pulls the fins into contact with the copper pipe.
My feeling is that this method does not result in as good a thermal contact between the fin and tube as the methods discussed above, but I've not tested this, and it may be just fine.
Woodsy's collector is detailed here...
Matt's collector is detailed here...
Anthony worked out the fin making machine shown in the picture below. The top angle has a 5/8 rod fastened to the under side. The aluminum fin blank is placed under the top angle and on top of the two lower angles. The two nuts on the threaded rods are then turned with with an electric driver, pushing the rod down onto the fin, and forcing the fin material into the slot between the two lower angles. The two bottom angles can then be pinched together by tightening the other two nets so that the fin gets a better wrap on the tubing. Pretty neat.
Click on picture for full size
Anthony's fin making machine
Anthony reports that it takes less than 30 seconds to run the two nuts and make a fine.
John Canivan uses this method to make all the fins for a serpentine absorber from one sheet.
All the details here...
This fin maker uses a salvaged press ...
Kris does some amazing machine shop work on his solar systems...
The pictures below show a tool worked out by David that makes a double fin.
Tom Sullivan is a long time friend of Build-It-Solar, and has contributed a number of really well done projects to the site. When Tom was making his collectors, he set up a dedicated press to make fins for his collectors. This is a very nice fin press, and makes fins that fit really well, and wrap nearly all the way around the tube for good thermal contact.
More on the press here...
Tom ended up with a whole lot of extra fin stock, and offered to make fins for people to use up this excess stock. He has done this for a number of people, and may continue making the fins after the initial stock runs out. If you want to talk to Tom about making fins, the link above has contact information.
I got a batch of these fins for my new collector (which is in work), and they are very nice -- they certainly offer better thermal contact between the tube and fin than the commercial fins listed below.
Update on Tom's Absorber Plates
Tom has decided to make absorber plates from his press available as a regular product from his UP Truck Center business.
He offers plates to fit 1/2 inch and 5/8 inch copper or PEX pipe. Fin length can be customized from 24 to 34 inches. He offers a "standard" fit as well as a 90% wrap fit. The modified vice-grip that make it easier to clamp the fins tight to the tubes are also available.
The prices are very reasonable, and I can attest to the fact that the plate fit is very good. I think this is a very good option for anyone who does not want to bother with making fins.
All the details on ordering and using the fins on ordering fins from Tom...
These are commercially available fins that are made for use in radiant floor heating. They are aluminum sheet, usually 5 or 6 inches wide, typically made from 0.018 inch thick aluminum, and have a formed groove that fits over the PEX floor loops in radiant heating systems. While I've not actually tried these, it looks like they would do a descent job of transferring heat from fin to tube.
If using these, I think it would be worthwhile running the sheet of flat aluminum under each tube with the grooved fin above that as discussed below.
Here is one example ...
This example is made from 0.019 thick material, 5 inches wide and 24 inches long.
These plates are only 5 inches wide, so you would need to place the riser a on 5 inch centers. I would still recommend using the flat sheet of 3 inch wide aluminum directly under the tube centerline.
These are bare aluminum and should be cleaned and painted before applying over the copper tube to prevent corrosion -- see also the note on silicone caulk below.
There are also commercial fins that are heavier gage, and are extruded rather than formed from sheet metal. The extruded shape snaps on the tube with a satisfying snugness.
I am sure these would work well, but they are more expensive.
These particular ones are only 4 inches wide, so riser would have to be spaced 4 inches apart. At about $6.50 per 4 ft length plus the cost of the added risers, this makes for a pretty big price bump, but a collector built with these would still be a lot less expensive than a commercial collector.
Note that the makers of these tubes show 4 inch wide fins installed on tubes that are 8 or 9 inch centers. This results in a 4 inch gap between adjacent fins. While this may be OK for radiant floors, it is definitely NOT the right thing to do for solar collectors. The outer edges of the fins on adjacent riser tubes should touch each other.
These are bare aluminum and should be cleaned and painted before applying over the copper tube to prevent corrosion -- see also the note on silicone caulk below.
I like the idea of applying a thin bead of silicone in the groove of the fin before pressing it onto the copper pipe. No matter how well formed your fins are, and how carefully you fasten them in to position, there will be a very small gap between the fin and the copper in some areas -- the silicone in 10 times more conductive than air, so (I think) you are better off having silicone in the gap. The silicone also glues everything together so thermal expansion and contraction don't work the fin and pipe away from each other over time. In addition, the silicone reduces any chance of galvanic corrosion by sealing out any water and providing a protective coating. I known many people are skeptical of this given that silicone is not a highly conductive material, but be sure to read the link below before you leave out the silicone.
More detail and performance testing on the silicone in the groove story here...
I use a 3 inch wide flat strip of aluminum under the center of each copper tube as shown in this picture.
The brown aluminum under tube is a separate 3 inch wide strip of 0.018 thick aluminum. This extra aluminum strip results in more contact area between the tube and fin, and also increases the aluminum total thickness for heat transfer to 0.036 in the area where heat flow is the greatest (the heat flow is zero out at the outer edge of the fin, and greatest at the fin tube joint).
Some aluminum alloys and tempers are very springy, and just don't take a groove very well at all. So, you need to find out if the aluminum you want to buy is easy to form. I think its best to buy a small sample and try it if you can.
Some of the solar references provide a formula for calculating the efficiency of a fin. A fin that conducted all heat to the tube without any losses (ie a made from a material with infinite thermal conductivity) has a fin efficiency of 1.0. Fins made from real materials will require that the outer edge of the fin run warmer than the tube area of the fin in order to produce some temperature differential to conduct heat into the center of the fin. This temperature differential from fin edge to fin root increases the average temperature of the absorber, which increases heat loss out the glazing, which in turn reduces the collector efficiency.
Fins on commercial collectors often have fin efficiencies in the mid 0.9's. The table just below gives calculated efficiencies for several of the common homemade fins and a typical commercial fin.
The fin efficiency depends not only on the fin material, fin thickness, and fin width. It also depends on the collector heat loss. The table gives efficiencies for both single and double glazed collectors. The equations I used for fin efficiency come from "Solar Thermal Engineering" by Peter Lunde, page 183 (a good solar reference).
NOTE: that this calculation for fin efficiency assumes that the thermal connection of the fin to the tube is perfect -- that is, it only considers the fin up to the fin to tube joint, not the joint itself. So, a fin has a high calculated fin efficiency can still blow it all on an poor thermal connection to the copper tube -- looking around at some of the Internet DIY designs, I think that this happens a lot.
|Fin Material||Fin Thickness||Total fin width
|Fin Efficiency (single glazed)||Fin Efficiency (double glazed)||Delta T Across Fin
The first entry is for the most common of the homemade fins -- a 0.018 fin that is 6 inches wide. The fin efficiency is 0.91for single glazed and 0.94 for double glazed which (I think) is fine.
The 2nd entry is for an aluminum fin that is twice as thick. The configuration I like to use where the center part is two thicknesses of 0.018 for 0036 total but the outer part is 0.018 thick would be somewhere between the first two entries -- maybe 0.96. So, by adding the 2nd piece in the center only, you pick up a bit on fin efficiency and also (probably more importantly) get more heat transfer area from fin to tube. Where the 2nd advantage does not apply to the fin designs that wrap or nearly wrap the full tube diameter (like Tom's and Henry's).
The third entry for the 4.5 inch wide copper fin is typical of commercial collectors -- at 0.96 its about the same as the homemade aluminum ones -- the advantages of the more conductive copper and narrower fin are offset by the very thin material typically used in the commercial fins.
The 4th entry is what you would get if you used the extruded aluminum fins mentioned above. The thick and narrow fin brings the efficiency all the way up to 0.99, but at a pretty high dollar price.
The 5th is for a fin like Steve Baer's (Zomeworks) "Big Fin" -- it shows that you can get good fin efficiency with very wide fins if you go with a thick fin. The thick material will make it harder to form the groove in the fin. I think the Big Fins use more of a clamping arrangement between the fin and tube.
I get emails asking about using steel for fins. The last entry changes the material of the fin in row 1 from aluminum to steel. Steel has only about 1/6th the conductivity of aluminum, and this causes the fin performance to really drop. If you want to use steel fins, they would have to be narrow and thick to work well.
The last column gives the delta temperature from the outer edge of the fin (the hottest part) to the root of the fin at the tube (the coolest part). If you assume that the average collector temperature goes up by half the amount of this delta temperature you can get a rough idea what this costs in collector efficiency using this calculator.
For example, if picking a typical flat plate collector, 40F ambient temp, 110F tank temp, 300 BTU/sf-hr sun, and a value of 8F for the fin delta T (from above table), the collector efficiency drops by about 1.2% compared to the same collector with a perfect fin.
Gary Feb 21, 2010