The Workshop Solar Installation


FIGURE 1 Click to see a large view of the floor slab just before concrete was poured

When I built my workshop in 1992 I had planned on someday installing solar hydronic heating. The slab was designed with in-slab heating right from the start. Embedded in the slab (Figure 1) is 500' of 3/4" plastic tube in 2 loops which is routed through to the attic where a 5 gallon water heater is located. That tank has a single 2300 watt electric element to provide all the heat for the 22x26' building. It works wonderfully. Working in a shop with the floor slab sitting at 17 C when it's -30 C outside is something else. The beauty of this system is that it takes no floor space, is silent and there is no open flame. You don't even know it's there.

With our Premier Ralph Clown's stupid idea of "Piratizing" the electric utilities, the cost of heating the workshop began to climb quickly, so in 2002 I finally got my act together and started looking for solar panels.  Guess what? Gas and electricity are CHEAP, so the solar industry is way behind non-renewable energy sources. There are no suppliers in Edmonton and the only one in Calgary is primarily photovoltaic, although they were dealers for Canadian made Thermo Dynamics products. They use old flat-panel technology, but their product would do the job. In June I ordered two 4x8'  panels and waited. Three months later, when the panels were supposed to be delivered, I contacted the dealer again. By the sound of things, my order was lost during a company merger or something, so I asked them to reorder. Two months later the panels were lost in shipping and the salesman was not answering the phone, nor were head office staff. When I finally got hold of them, I gave them another 2 weeks to deliver the order but cancelled it when the excuses continued. It was obvious they were not in a position to either supply or support solar hydronic systems so I took my business elsewhere.

I gave the order to Thermomax in Victoria BC primarily because the rep was very knowledgeable and helpful. Unfortunately, the product is made in Ireland and it took a while for the material  to arrive in Edmonton. Luckily winter held off until late December, so the panels were commissioned on December 9th (Figure 2). 

The way heat is moved from the roof to the floor slab is pretty simple. Basically the heat from the glycol solution in the solar circuit is transferred to the floor slab circuit through a small heat exchanger. It is positioned upstream of the heater tank (Figure 3) so if the circulating temperature is high enough, the electric backup kicks out and heating is 100% solar. Fluid sent to the floor slab is heated to only 30 C by electricity and boosted to 40 C by solar heat. The shop temperature is regulated based on slab temperature with a base temperature of 15 C, but it is allowed to float higher if solar heat is available for harvest.

One of the main problems with solar heating is access to the sky. Alberta  is one of the sunniest places in Canada, particularly in winter, so it is ideally suited to solar heating. Mapping the sky is relatively easy, if you don't mind your neighbours looking at you as if you are a loony for sitting on the roof with a transit! A map of the solar sky (Figure 4) can be generated with some neat software called Sundi. 

Performance curves for solar panes are just plain baffling. There was absolutely no way to determine how the panels would perform before they were installed. After they were installed, it was obvious I needed data. To gather information on performance through the day I installed two HOBO data loggers, one 4 channel temperature logger on the heat exchanger and a 3 channel (plus humidity) logger to record slab temperature, inside and outside air temperature. Knowing the delta temperature across the primary and secondary side of the heat exchanger I was able to determine the amount of solar energy gained. Basically,  the solar panels are delivering as much energy as the electric backup whenever the system is running >60 C (Figure 5) and boosting the slab circuit significantly whenever the solar side is functioning (greater than 40 C). This cut out 2 months of electric dependency and 25% throughout the worst parts of winter. On good days the floor slab is warmed sufficiently during the day that it coasts through the night without needing more energy added. 

Sadly, my neighbour 2 doors down has a HUGE willow tree in her back yard that puts the boots to solar gain for about an hour a day for 2 months (Figure 5), the best hour of the worst months! Overall the panels are collecting for 70% of the solar day (Figure 6). I will be changing the angle of incidence to improve efficiency during the winter months since the shop doesn't need too much heat during the summer. For most of the summer the floor slab was sitting at 30 C! (Figure 7). Increasing the angle will help the panels shed snow which is an annoying maintenance problem.

Throughout the year I have been tweeking the system and learning more about how it operates. One change was the installation of an automotive expansion tank to accommodate the wide temperature swings (-40 to +100 C) , scoop entrained air and protect against header boiling should the power go out. Another change was a big surprise. As spring approached I was concerned that the system was running too hot and was worried it wouldn't  be able to cope with summer sun without boiling. As an experiment I added Redline Water Wetter, a product I use in the engines I run on the dynamometer. The data logger showed an immediate 15 C decrease in system temperature and a corresponding increase in delta temperature across the heat exchanger. The product, which I think is just dish soap, improved the efficiency of the heat transfer at both the solar panel header, and the heat exchanger. After that, the panels operated in full summer sun without exceeding 90 C.

Bottom line, is it worth it? Well, gas and electricity are never going to come down in price (thanks Ralph, you asshole) so solar heating makes sense. It makes more sense the earlier the system is installed so it can start paying for itself. My total investment for panels, pumps, insulation, loggers etc is $8000 CDN (NOTE there is no government rebate for solar installations!!) so payback at current energy prices is about 10 years. Now if only I had installed the panels when I built the shop in '92...

FIGURE 2 This is what the installation looks like on the roof

FIGURE 3 Click to see a cheesy diagram of how the solar array is plumbed into the electric primary heating system

FIGURE 4 Here's a map of my solar sky. The dominant features are my neighbour's house in the SE, her #@&^*!ing tree in the south, and a 6 story senior's centre 500 m away in the SW


FIGURE 5 This is a plot of the solar array outlet temperature at the heat exchanger through the day. These are monthly averages for 2003. Not surprisingly, this plot has the same form as Figure 4, notably the ghoddamn tree!!

FIGURE 6 The logs show that on a good day 70% of the available sunlight is converted to heat

FIGURE 7 The temperature probe measuring ambient outside air temperature shows that 2003 was a pretty average year. The impact of solar assist can be seen on the slab temperature trace

Last updated: 12/27/2006 .