Performance of Kevin's $1K Solar Water Heating System

This page gives the details on the performance of Kevin's $1K solar water heating system.  Plots and comments on performance are provided for each month for nearly a full year.  Kevin lives in a very difficult climate and gets some shading in mid winter, but the system still (I think) does quite well. 

 

Thanks very much to Kevin for providing this long track record of performance!!

 

Monthly Performance Plots

May 2009...
 

June 2009...
 

July 2009...
 

August 2009 ...
 

September 2009...
 

October 2009...

 

November 2009...  (see the note here on Nov, Dec, Jan, Feb performance)

 

December 2009...

 

January 2010...

 

February 2010...

 

March 2010...

 

May and June 2009

Performance for May 21 through June 19 of 2009.

 

 

The system is achieving a somewhat lower solar fraction than my system -- we think that this is due to the following:

The collector area per gallon of hot water demand is lower than my system.  Kevin's family hot water use is about 57 gallons per day and his collector area is 45 sqft of collector area.  This compares to our roughly 24 gallons per day and an effective 40 sqft of collector.

 

The ground water temperature in Kevin's area is very low -- about 43F, so there is more heating to do.

In order to get better summer performance, Kevin has changed the tilt to a more horizontal angle.  The performance graph shows the before and after, and the tilt change definitely improves performance.

 

Some additional observations from Kevin on performance:

    * tank temp gain on sunny days is 8 to 13 C

    * Cloudy days don't gain much if anything, especially with hot
      water demand at the same time (May 31 is a good example).

    * Tank temp during periods of no hot water use is stable,
      indicating minimal heat losses from the tank.  See night of May
      22/23, where the only drop was in the morning after a shower. I
      added my insulation scraps to the sides and top of the tank
      today for even more insulation, so there is a minimum of 3" of foam.

    * Our family uses a lot of hot water!  Tank temp drops fast with
      hot water use.  I was disappointed with this at first, but this
      is exactly what I want - ready exchange of heat from the tank
      into the water flowing through.  Incoming water is at 6C, so the
      exchanger is transferring a lot of heat during those 12 minute
      showers that I timed.

    * The stagnation test yesterday at 12:30 PM on a sunny day had a
      peak temp of 76C (I saw 81C on the Steca between the 5 minute
      datalogging readings).
 

Overall, I am happy with my system. I realized beforehand that my hot water demand would be higher than yours, but hoped the system would do better. But still, I am getting a solar fraction of:

Incoming water temp 6C
Hot water tank set at 40C
Temp rise needed: 34C
Ave tank temp these days 35C
Solar temp rise = 35 - 6 = 29C

Solar fraction = 29/34 = 85%

July 2009 Performance Update:

Kevin sent in the performance update for the month of July

 

 

From Kevin:

As promised, here is July data.  The family was away for a few weeks early in the month, so I thought this would get a good test of the performance with low demand.  Unfortunately, it coincided with a period of cool, cloudy weather, so my hopes of breaking the 50 C barrier weren't realized.  Top tank temp was 48C.  The last 2 weeks of the month were mostly hot and sunny.  The tank temps average about 40C, so this is hot enough that my hot water heater hasn't come on at all since ~July 14th.
 

This looks better to me, so it looks like the increased tilt helped some.

 

 

August 2009 Performance Update:

Kevin sent in the performance update for the month of August.

Here is Kevin's note on the August performance:

The notable part of this month was a 7 day period when we were on vacation.  This is a good illustration of what may be the maximum I can expect from the current system.  The tank temp got to 53C while we were gone, in a period of sunny weather.  The tank temps declined in the last few cloudy days of our absence, showing heat loss from the tank.  While we were home, the tank temps were mostly between 30 and 40C, but dipped as low as 20C. 
 
This month's performance led me to start work on an additional collector panel, which will increase the collector area by ~50%.  That, and some tree removal now that the sun is lower and the collectors are getting partial shade, should improve things.
 
The data gaps were caused by low battery voltage, running the datalogger using AA cells.  I have since switched to a wallwart power supply.
 
I decided to replace our showerhead.  The old one wasn't bad, at 2 USGPM, but we bought a new Waterpik multi-setting one, and I especially like the pause feature.  Flows from the new one are 1.4 GPM in normal mode, 1.6 GPM in massage mode, and 0.6 GPM in pause mode.  Everyone is happy with the new one.  Now it's a matter of training the household to use the pause setting when we can.
 
http://www.waterpik.com/shower-head-products/fixed-mount/ecoflow/VBE-423/
 

September 2009 Performance Update:

Performance plot for September 2009.

 

Some comments on September:

 

Kevin

 

 

 

October 2009 Performance Update:

Performance plot for October of 2009.  Includes effect of an added 3 by 8 ft collector on the 11th.

 

 

I commissioned my additional panel on October 11th.  It's a 3' X 8' one that is very similar to my existing ones.  The main changes are:

Luckily I had a blue sky day for the first day of operation, although it was unseasonably cold in early October.  This was the first day the system has started, run, and stopped in below freezing conditions, and I am happy to report that there were no problems.  Starting temp was -7C, getting up to -1C.  I gained 11C in the tank in the 4 hours of operation until the trees in my yard shaded the collectors, which bodes well for performance.  A rise of 11C would be good performance for running all day with the original collectors.   I continue to be amazed at the power of the sun, even in the cool half of the year - when it was -5C out, the collector temp was +45C, giving a 50C difference!

 

October was not a very sunny month at all.  A pre- and post-expansion comparison can be seen for October 5th performance (8 C tank temp rise) and after Oct. 11th (10 to 12C rise).  There were also some stormy days that were completely flat, with no collector temp rise over the outside temp.

 

I am seeing that the new panel runs 2 to 4C higher in temp than the old ones, and I am thinking that this is because of the insulation I put on the sides, as well as maybe the tighter construction with the foam wiggle strips.  If the new panel runs hotter before the pump turns on, this is a good thing, indicating better heat retention, but if it runs hotter when the pump is running, it may mean that the panel is less efficient at transferring heat to the tank.  So I decided to do a stagnation test to check this out.

 

The new collector was running 3C hotter when I turned the pump off for the test, and the new one climbed to 7C hotter and stayed there.  The outside air temp was +4C during the test, and the collector temps went from 44C to over 80C, which covers the temperature range that they run at when the pumps are running.  The test results give me confidence that the extra insulation is effective and should result in more heat available for the circulating water to collect.

 

Kevin

 

 

 

November 2009 Performance Update:

 

Here are monthly performance graphs for my system for November through February. The short story is that from mid November to the end of January, the system didn’t provide much useable heat, but by February I was back in the game. It ran during those months when there was sun, and I’m relieved to report that there were no freeze-up problems whatsoever. Incoming water temps ranged from 8C at the start of November to 6C now. With the low sun angles in winter, there is a lot more shading from trees in my yard, which compounds the performance reduction from the fewer hours of sun.

November shows a good first half of the month, with sunny days giving a maximum collector temp of around 50C, then a period of cloudy weather that dropped the tank temperature quite low.

The dark period around the solstice was interesting. I really didn’t expect much if any useable heat harvest, and I was right. The best daily tank temp rise was 5C. The most impressive day was Dec. 13th. The overnight low was –41C (ouch) but the next day was clear once the ice fog burned off. The high outside temp for the day was -30C, but the collector ran! Inside the collector it got to +20C, which amazed me. On warmer days the collector temps got to +40C, which shows less heat loss to the outside of the collector in warmer outside temperatures. No problems with the drainback, either – I was quite happy with myself for keeping steep (3% or more) angles on the collector tubes and the lines to the basement.

January was pretty quiet too. Reasonable performance at the start of the month with a succession of sunny days, but dropping off to nothing in the second half. A sunny day showing a 10C tank temp rise on Jan. 27th shows that the sun was getting stronger a month after the solstice.

I also replaced my domestic hot water tank in January. My old tank was running fine, but it was the original one, 25 years old, and there was a federal Home Renovation Tax Credit program which gave a 15% tax credit for items like this. It sounded like a good time for a planned replacement, as opposed to waiting until it packed it in at some really inconvenient time. I debated about instantaneous gas ones, but the midwinter performance of my solar system told me that I would have to size an instantaneous one to cover all of our hot water needs, and also account for very cold incoming water. For most of the year, this would be oversized due to help from the solar system. So, the price scared me away. The next option to consider was a higher efficiency storage type tank (Energy Star rated). These are not too easy to find in our area; the big box stores like Home Depot don’t carry them and I would have had to special order one. So, I bought a regular tank locally and put an insulating jacket on it.

February was much better, going from a tank temp below 10C at the start of the month to over 40C. The second half had record warm and sunny weather; the same as what the Vancouver Olympics was getting. Due to the higher sun angle, the maximum collector temps were over 60C.

So for the first winter, I am getting about what I predicted, maybe more. I am looking forward to the summer months to see what performance I will get, especially with the extra panel I installed in October.

 

Kevin
 

 

 

 

 

December 2009 Performance Update:

 

 

 

January 2010 Performance Update:

 

 

 

February 2010 Performance Update:

 

 

 

March 2010 Performance Update:

Kevin's solar water heating system performance for March of 2010.

 

Hi Gary,
 
Here is a graph of March performance data to add to the Nov-Feb ones I sent you a while ago.
 
March saw good performance.  We were away for over a week, and during that time the tank temp got up to 49C.  Late in that period, I lost the data logging sensor in the tank, so there is a gap in tank temp data.  It was one of the homemade ones I made by encasing a DS18B20 sensor in epoxy, so I wasn't too upset.
 
Regards,
 
Kevin

 

 

Gary November 14, 2009, April 17, 2010