Marc-Antoine Meilleur holds a degree in Mechanical Engineering from the Cégep de Sherbrooke. He is pursuing a Bachelor’s degree in Mechanical Engineering at ÉTS.
In winter, greenhouses require a large number of air changes to reduce moisture levels. According to analyses conducted by the student club SerreÉTS, SerreÉTS, a well-insulated passive type greenhouse, like that in Figure 1, would need about 1350 kWh/m2 to heat itself, with about 80% of the load used to heat the air changes. Moreover, in the hot season, a ventilation rate of 5 m3/s may be necessary to passively cool a 50 m2 greenhouse. Solar walls can help solve these two problems. However, this technology is not yet embedded in either greenhouse or residential sector. Nevertheless, a solar wall is easy to manufacture and many of its main elements can be made from recycled materials. Among others, the main element of a solar collector—the aluminium absorber—can be made of aluminium cans instead of perforated aluminium foil. The aim of this research project is to estimate how long it will take to recoup the cost of a solar wall depending on whether the materials used are new or recycled.
In order to achieve this, a test bench comprising two totally identical solar collectors—except for the absorber—was manufactured and installed on a support, as shown in Figure 2. The first sensor, “ECO” for ecological, is composed of an absorber made from aluminium cans. The second, “STD” for standard, is composed of a more conventional absorber, made from a sheet of perforated aluminium foil.
Heat transfer from the STD sensor occurs mainly when the forced air passes through the hot aluminium plate, which has a porosity of 0.003. In the ECO sensor, the air is directed into 15 pipes made from aluminium cans as shown schematically in Figure 3. The transfer of heat thus occurs mainly when the air is in contact with the wall of the cans. The two sensors are insulated, glazed and oriented vertically due south.
The ease of fabrication associated with the use of recovered materials was one of the main constraints in the design of the walls. Indeed, the sensors constructed are reproducible with basic tools and common recovered materials. Manufacturing time is only a few hours and the lifetime of the sensors should exceed ten years. For this test bench, all materials were purchased new, which made it possible to determine the maximum cost of manufacture. Materials were purchased with the support of the ÉTS Student Association Sustainable Development Fund (FDDAÉTS) and the ÉTS Development Fund.
As shown by Table 1, the more recycled elements used in the construction of the sensors (green boxes), the lower their manufacturing cost. Different manufacturing cost scenarios are proposed to analyze the payback period of each. At most, the ECO sensor costs $410 by recycling only the absorber (aluminum cans), and about $71 if the other elements are also recycled: wood (2” x 6” and plywood 4′ x 8′), insulation (3 sheets of extruded polystyrene 4′ x 8′ by 1″), glazing (glass or acrylic 4′ x 8′) and the electronic component (fan). At most, the STD sensor will cost $508 ($100 more to buy the aluminium absorber); at a minimum, about $175 if most materials are recycled.
Table 1 Manufacturing Cost Scenarios for Recycled Items
The sensors were installed at the CANMET Energy Center in Varennes to have ideal sun exposure and access to locally measured climate data such as solar intensity in W/m2.
Sensor yields were established during the month of February 2017 by comparing the energy produced with the potential energy on their surface. With a continuous measurement of air temperatures at the input and output of the sensors with an accuracy of ± 0.5 ˚C  and a manual air flow measurement with an accuracy of ± 3% , we used the following equation to determine the amount of useful energy generated during the month.
Table 2 presents the rate of consumption and sensor yields as a function of the measurement uncertainty.
Table 2 February 2017 Production and Uncertainty Performance
An average hourly yield of 30% was determined for the ECO sensor and 39% for the STD. The RETScreen Expert  software was then used to estimate the energy generated in a typical year in Montreal. Annually, the ECO sensor would have an energy production of 588 kWh and the STD, of 617 kWh. For an electricity cost of $0.08/kWh, annual savings are $47 for the ECO sensor and $49 for the STD. This means that, depending on the number of recycled materials, it takes between 2 and 9 years to recoup the cost of the ECO and 4 to 10 years for the STD.
By comparing these results with those of a commercial sensor such as the “Grammer SLK“  under the same conditions, the efficiency of the Grammer SLK is higher but it takes about 28 years to recoup the costs because of its high purchase price (approximately $950/m2). Even considering labour costs of $20/hour for the ECO sensor, the pay-back period is only 5 to 13 years, depending on the number of recycled materials.
The hypothesis proved to be correct. Even if the average efficiency of a sensor made with a canister absorber is lower, its pay-back is more advantageous. It is therefore quite profitable to take advantage of renewable energy in Quebec. In some cases it is enough to demonstrate creativity and to highlight recycled materials in a design.
For the rest of the research project, the walls will be installed on real greenhouses where the fresh air flow will be increased to about 120 CFM. A higher flow rate will only increase the efficiency of the sensors, as shown in the diagram demonstrating the efficiency of a solar wall as a function of its flow rate, Figure 4. In addition, in these new facilities, we will try to take advantage of the natural ventilation that the sensors can offer in the summer, because overheating is another important greenhouse problem.
The implementation of this first phase was possible thanks to the involvement of many people in the project, including members of SerreÉTS and Professor Stéphane Hallé. A special mention goes to the CanmetÉnergie research centre, which offered us its support and a place to experiment during these first tests. Finally, this research would not have been possible without the financial support of ÉTS.