301 Moved Permanently

---

nginx

A family foundation in Seattle, working globally to ensure that people can live healthy and productive lives, was interested in adding a solar component to its new office building. Its team reviewed proposals for potentially adding a photovoltaic or solar thermal system for the project. Neither system was large enough on its own to qualify for a Leadership in Energy and Environmental Design point, but solar hot water had a much larger contribution to lowering the building’s carbon footprint and provided a better return on investment.

There were several challenges to meeting the client’s requirements. First was the load profile. You cannot get an actual load or time-of-day usage number in an empty, new building. We knew the main load would be the cafeteria serving 1,500 meals a day. At first, we had no idea how many people took showers at work. Think bicycle-friendly Seattle and an unusually warm August. We later discovered that hundreds of very sweaty employees showered at work in the summer. This was to be a green building after all.

 

Analysis and system design

Entering design phase, it was not known what the real-world load would be. Was it 1,000 or 3,000 gallons per day of hot water? What about the weekend load?

The most important balancing act at this point was determining how to set up the temperature ceilings for the collectors to ensure that a supply of hot water was available for the building’s occupants first thing on a Monday morning. Once the tank’s max was reached on a given day, the collectors would quickly go above the 250°F limit. This would shut down the collector to drain back tank loop pumps until the collectors cooled the next day. The idea of having the system limited out in the morning did not sit well with the team.

We decided to split the system into two parts. The A system - with a smaller pump and less piping - would have a higher temperature limit. The B system was set with a lower limit and would not run as long.

The system was load matched for a light-load weekend condition, thus not getting as easily to the limit - which shut down the whole system. Part of the load was the building recirculation loop. The design engineer gave us 16,000 btu/hour of heat loss at all times.

Real-world testing showed it to be 20,000-30,000 btu/hour, which included standby heat loss of the two 1,000-gallon hot water tanks. The data log showed that managing heat loss was a much bigger part of the client’s solar savings than we had anticipated during the design stage. The water heater tank would be set to 140°F at 6 a.m., then set back to 120°F at midnight and on weekends. The second tank was for a heat recovery system, which was also used as solar storage.

The design also incorporated a three-way valve that put the water heater tank in series with the solar storage once the solar tank was heated to 125°F or more. At that time, the domestic hot water (DHW) would be 100% solar-heated, and the boiler would be cut off.

This was an important factor in keeping the boiler off at night, as well as on the weekends. Our approach covered much of the standby heat tank’s losses and recirculation loss. At first, the recirculation was returned to the solar tank. That was changed to return to the DHW tank for a number of reasons - the most important being the effect on system efficiency.

The drain-back system had its own temperature control and acted as a buffer tank. There was a nightly hot water load of 300-500 gallons, which would take about 40 minutes to produce. On the data logger, a big spike in btu input to the storage tanks between 10:30-11 p.m. was seen. The btu meter was on the potable side of the drain-back tank and plate heat exchanger. This setup has no heat exchanger loss between the collectors and the drain-back tank, resulting in more efficiency and tighter temperature control from the collectors to storage.

 

Flat plate or vacuum tube?

The next challenge was the type of system that would be used: flat plate or vacuum tube? We had done jobs with both.

Living so far north, we have long summer days and short winter days. Surprisingly, we don’t get much cloudy, rainy weather in the summer through September, and we have low humidity as well. Then come December and January; the skies in the Northwest are very dark gray most the time. They have been called “the throwaway months” when it comes to solar hot water.

In the summer months, a tube system we had installed for a brewery worked from 7 a.m. to 7 p.m., due in part to the 360° absorber. The sun would hit the back of the tubes just after sunrise. They were mounted solar south at a 35° angle, which is a good angle for the Pacific Northwest. By 9 a.m., they were at 70% of the max output for the day. At 5 p.m. - the heat of the day in the Northwest - the system also was about 70%. Output then dropped fast afterward.

The flat plate systems we had installed 80 miles away were starting about 9:15 to 9:30 a.m., depending on collector sensor placement. They were off by 4:45 to 5 p.m in the summer months.

Our team chose vacuum tube collectors. The solar hot water design for the foundation’s commercial solar system consists of 47 tube collectors with a footprint of 2,100 square feet in three collector banks and 16 arrays. The collectors are mounted at 35° angles and face 7° east of solar south. The collector angle is suited for the long summer days and the most sun annually overall. It worked out that no winter shade is given to the collectors from 9 a.m. to 3 p.m. The 47 collectors also fit on a single wing of the roof area.

Six-hundred feet of well-insulated pipe were used to do both arrays in “reverse return” for the closed-loop drain-back system. Because we were using reverse return, we implemented flow balancing. With correct pipe sizing, there was no need for any balancing valves on the arrays. This offset the added pipe cost and the complexity of more valves.

The system pressure was set to 13 psi, with a cold 60°F tank. Normally, this was left at 0 psi. This was done for two reasons: The pumps on the drain-back tank needed more suction head to prevent cavitations, and it also allowed the system to run at a higher safe temperature. An oversized 375-gallon drain-back tank was used as storage and as a buffer without a heat exchanger between the collectors and the tank.

The controller has an additional collector high limit set at 250°F. The system would not pump water to the collectors at that temperature under any conditions.

The foundation building uses 1,300-7,500 gallons of hot water daily, and the drain-back design is appropriate for preventing overheating in systems with large fluctuations of daily or seasonal hot water loads.

The solar thermal system has a peak sun output of 73 kW thermal and 230,000 btu/hour from a gross collector area of 2,103 square feet - which accounts for an estimated 50% of the building’s daily design load for producing domestic hot water.

The system has top-of-the line metering and is remotely monitored 24/7. The custom controls were integrated into the building’s automated control system.

Those at the foundation are impressed with the system’s performance and thrilled to be able to use solar energy cost-effectively in the cloudy Pacific Northwest. S

 

Tim Connolly is senior technical advisor at Pioneer Renewables LLC, a solar thermal consulting firm based in Vancouver, Wash. He can be reached by email at tim@pioneerrenewables.com.