Kevin built a version of the $1K Solar Water Heating system, and provides a lot of good information on 1) how he designed and built the system, 2) well thought out changes he made to adapt the system to his situation, and 3) the performance of the system.
Kevin faced some unique tank space limits and collector placement limits -- the techniques he worked out to adapt to these limits should be helpful to people facing similar situations.
Kevin's system is located in far nothern British Columbia at 55 degrees latitude, so it is a good challenge for the system.
Thanks very much to Kevin for taking the time to detail the project!
Kevin's system definitely shows the benefits of taking your time and working carefully.
Kevin provides an excellent description of the way in which he designed and built the system in this pdf:
Full design and construction details on Kevin's system... (pdf)
I really like the way Kevin did the aluminum fins, and the heat exchanger is a very nice solution to the limited inside dimensions of the tank.
Note also the pitching of the copper pipe grid towards the center of the collector for good drain back. This is essential for freeze protection.
Kevin provides detailed monthly performance plots for the periods from May 2009 to through October 2009, along with comments on factors effecting performance, and the effect of adding an additional collector in October.
Kevin has lots of experience in data logging, and has set up a system to monitor and record the performance of his system.
All the details here...
Kevin decided to try a hybrid copper and PEX heat exchanger. This arrangement fits his limited tank space better than the large PEX coil that I use. The construction details are shown in the pdf above in the Construction section.
Kevin's test of the heat exchanger performance:
As promised, here are the two heat exchanger tests. The pattern is very similar to yours - hot water temps are the same as, or a degree lower than, the tank temps during the initial emptying of the exchanger. Then the hot water temp drops in tandem with the tank temp, staying ~2 degrees lower. At higher flows, the temp falls off, but makes a total recovery after the rest period.
Click on images for full size
It appears to me that hybrid heat
exchanger gives very good performance -- about equivalent to the 300 ft coil of
PEX exchanger that I used -- so, there is more than one way to get the job done.
Kevin describes the datalogging system in the construction pdf. Here is a little more detail on the hot water line sensor that is pictured just below:
Not sure if these will stand the test of time, but at $5 for a DS18B20 sensor vs. $47 for the only waterproof commercial DS18B20 probe I could find online, the price is right. The datalogging tank sensor is located at the same depth as the Steca one; about 1/3 up from the tank bottom. I am logging at a 5 minute interval.
I installed a sensor on the hot water line out from the exchanger. This is mainly to do an exchanger test, but it also tells me when we are using hot water. I drilled a hole in a 3/4" end cap and embedded a homemade probe in epoxy, then installed this into a tee in the line (see photos). Don't worry about the hose clamp - just temporary until I got a PEX clamp. So far so good.
Click on pictures for full size
This is how Kevin sums up the size of the job -- I think that it is a very accurate assessment:
As far as building the system, now that it's over I don't think it was that bad :) Seriously, it's not a trivial project, but well within the average DIYer's capabilities if a person takes it slowly and puts some care into it. It's not one of those "hey, build this simple system and you can be saving thousands right away" type projects that you see being hyped these days. Maybe commercial tanks could be used, like plastic tanks that could be insulated. This might simplify the construction.
The bottom line is that I am harnessing the power of the sun in a cost effective and fun way!
If you have questions about the system, Kevin provides and email address in the Construction pdf above.
Thanks again to Kevin for doing such a great job of documenting and testing this system!!
Gary June 26, 2009, Added performance update November 2009