Print This Page

DayCreek Journal

May 6 , 2009

Solar Heating System Modifications

solar panels 
Here's a view of the solar panels in back of the house. The foreground panels are solar electric. The background panels are solar collectors used to provide heat to the house.

I used to think that when I retired from the rat race that life would slow down to a more leisurely pace. Such is not the case! I've had my hands in so many different projects, it's been difficult to get things done. One project that did get done this past winter and has worked magnificently is the modification of the solar heating system.

The solar heating system up until this winter dumped the solar heat load directly into a insulated sand bed located underneath the floor of the house. Although this system has worked, I found it more often than not only capable of supplementing the heat load rather than replacing it. After three sunny days in a row, the floor would give off enough heat to provide 80 to 90% of the heat load, but usually I would have to light a fire in the stove once per day. This is certainly nothing to sneeze about, but I wondered if there might be a way to go one step better.

When I first designed the sand bed heating system, solar heating guru Steve Krug gave me his opinion on the sand bed vs. a large water tank concept. It was Steve's opinion that the sand bed system would dry out after time and would not store nearly enough heat to heat the house.

Thanks to Steve's doubts, I made sure I had a backup system in case the sand bed ever dried out. As of 2009, the system has not dried out but with multiple year's of experience now under my belt, I was curious as to if I could squeeze more Btu's out of the system—especially in the dead of winter.

It is rare that during the midwestern winter there are more than a couple of days of sunny weather. This can be a detriment to a sand bed system—just as the sand bed becomes charged, the sun is replaced by clouds and within a day or two the sand bed is not giving off much heat at all. Too much heat energy is used to get the floor temperature up to a value that is substantial.

Backup Plan
When I installed the sand bed heating system, I also laid 4 sections of 300', 1/2" pex tubing directly into the concrete floor slab. This 1,200' of pex tubing was there for two reasons:

(1) In the event I am away from the house for an extended period, a boiler can be turned on to heat the house. Setting the thermostat at 55°F, the house will never get too cold.

(2) In the event that the sand bed heating system did not live up to expectations, I could still heat the floor using the sun's energy by pumping solar heat directly into the slab tubing.

So when it came time to replace the non-toxic glycol and replace the DC circulators (pumps), I decided to marry both the sand bed loop with the slab loop. If played my cards right, I could valve the loops to give me multiple options and that's what I did.

Here are the options:

Option 1. Summertime. Solar heat only goes to the 80 gallon domestic hot water tank. (All but two panels are tarped to keep the system from overheating.)

Option 2. Late Spring and Early Autumn. Heat from the solar panels heats the 80 gallon DHW tank and the sand bed.

Option 3. Early Spring and Late Autumn. Heat from the solar panels heats the 80 gallon DHW tank, sand bed and slab.

Option 4. Early and Late Winter. Heat from the solar panels heats the 80 gallon DHW tank and slab, bypassing the sand bed.

Option 5. Mid Winter. Heat from the solar panels goes directly into the slab and nothing else.

Having the system valved as such provides many options to route the heat from the solar collectors and the results have been quite favorable.

For example, this past winter during the first week in February it was clear and cold for seven days in a row. Nightly temperatures were below zero (F) and daytime temperatures were in the teens (F). Although it was rather chilly, the sun shown brightly for most of those days.

I found that by pumping the solar heat directly into the floor slab and nothing else, the house stayed comfortably warm for those seven days without any other heat source whatsoever. Daytime temperatures in the house usually approached 75°F and morning temperatures dropped to about 68°F.

I will be the first to admit that having seven days in a row of sunny weather in the middle of winter is unusual, but it proved the viability of the system. Seven days in cold weather without using any other heat other than the sun had me grinning ear to ear.

Don't forget that this house is a high thermal mass house which helps regulate the temperature. Doing this without lots of thermal mass would cause temperatures to fluctuate too much causing higher indoor temperatures during the day and lower temperatures at night.

Pumps Changed Again
When I re-engineered the system, I wanted to stay with DC pumps powered by the sun via a 75W PV panel. The two bus pumps had met their demise, so I replaced them with three Hartell MD-10-HEH. These are brushless DC pumps and do a pretty good job. With that said, they didn't have the horsepower that I had hoped. After two months of experimenting, I decided to go with an AC pump instead of the DC pumps I had used in the past.

This broke my heart. I always liked the simplicity of the DC pump. If they are properly sized with a DC PV panel, the system does not require any kind of controller to monitor collector and floor temperatures. When the sun is shining brightly, the pumps are running full steam but when the sun is at a low angle or partially covered by clouds, the pumps run slowly. This variable speed concept worked like a charm, but finding 10+ GPM pumps that are DC powered are few and far between and expensive.

The Hartell MD-10-HEH is a great pump, but it just wasn't powerful enough for a whole-house heating system. After two months of fiddling around, I replaced the DC pumps with one, Grundfos AC pump rated at 10 GPM and a Siemens differential conrtoller. The Siemens controller was one that I had here, but never used since it was designed for a different type of solar heating system. It needed a bit of modification, but it and the Grundfos pump are working just great.

solar heating system
Here's a view of the solar heating system. On the left is an 80 gallon tank used to heat the domestic hot water supply. To the right of the tank are the two sets of pex tubing used to heat the floor. The darker red tubing (3/4") goes into the sand bed, while the brighter red tubing (1/2") goes directly into the concrete slab. On the far right is the electric boiler used for off-peak, backup heat.


Unusually Sunny Winter/Early Spring
Tying into the subject of solar heating, I thought I would add a few tidbits of good solar generation news. The solar PV electric system has generated the following power so far this year:

January: 346 kW
February: 417 kW
March: 445 kW
April: 496 kW

January, February and April have all been record setting months over five years of collecting data and March was close to a record as well.

Using Thermal Mass in Northern Climates -- MREA Fair
As a final note, I'll be doing a presentation at the Midwest Renewable Energy Association's Energy Fair (June 19-21) near Steven's Point, WI. This is the 20th anniversary of the fair and it's always a great place to learn just about anything related to green living.

I'll be giving my presentation on Friday at 10am in Pink Tent and again on Saturday, 12 noon in Green Tent. Hope to see you there!

Sandhill Crane with Colt
Spring has sprung in Southeast Minnesota. Here is a photograph of a 2-day old colt and parent Sandhill Crane taken near Hokah, Minnesota.



(3) Hartell MD10HEH Circulator / Pumps
These are high efficiency, DC brushless motorized pumps that are rated at 30,000 hours of life. They were used for 2 months in my solar heating system and were only used on days when the sun was shining.

Price is $275 each or make me an offer on all three. Use the comments page to send me an email about the circulators -- Alan