Appendix A: Agricultural Lighting

Efficacy in lumens per watt, the metric for lighting for humans, measures the photopic spectrum, which is the light used for color vision under normal conditions. But lumens per watt is not a relevant metric for agricultural applications because plants respond to different parts of the spectrum: mostly red and blue, and different plants have different spectral needs. Additionally, the same plants have different needs at different times of day and in different parts of their life cycle. Figure 1 illustrates some of the differences.

FIGURE 1: Plants respond to a broader light spectrum than humans

Plants respond strongly to light in the wavelengths between 400 to 700 nanometers (nm), whereas humans respond strongly to a much narrower spectrum—between roughly 500 and 600 nm.
Chart showing spectral differences in plant and human sensitivity

Light emitted in the wavelengths between 400 and 700 nanometers (nm) is used by plants for photosynthesis, and this range is collectively known as photosynthetic active radiation (PAR). The rate of light produced in the PAR range at the light source is referred to as photosynthetic photon flux (PPF), which is the number of photons produced per second, reported as micromoles per second (μmol/s). PPF is used to describe the photon output of a lamp, but it does not specify the light reaching the plant. The rate of light in the PAR range that is actually reaching the plant is referred to as photosynthetic photon flux density (PPFD), and it is reported in units of micromoles per square-meter-second (μmol/m2·s). The metric most relevant to actual plant growth is called daily light integral (DLI), which is the cumulative sum of PPFD incident on the plant over the course of a day, and is reported in units of micromoles per square-meter-day (μmol/m2·d).

The most relevant efficiency metric is the number of PAR photons produced (PPF) per watt of electrical power, known as the photon efficiency, and it is given in units of PPF/We (or μmol/s-We). If expressed in terms of energy, the metric would be PPF per joule (μmol/J). For conventional grow lights, such as high-pressure sodium and metal halide lamps, there are conversion factors between conventional metrics and PAR. For conventional light sources, a single conversion factor for a particular type of lamp is more appropriate because the lamps have similar spectral power distributions (that is, they all put out relatively the same fractions of light energy at various wavelengths). For LEDs, different products can have very different spectral power distributions; LED luminaires can be designed to optimize the spectral power distribution; and some LED products are tunable, meaning that the spectrum can be tailored to needs as they change over time.

Some plants reach a saturated state after receiving their daily dose of DLI or PPFD, and extra light beyond this point doesn’t yield any more growth. In the case of cannabis, however, artificial light is so much weaker than direct sunlight that indoor cultivation facilities can’t provide nearly enough light to reach this threshold.1 What this means to the cultivator is that cannabis will absorb as much light as can be thrown at it as long as the lamps aren’t too close to overheat the plant. But the lamps must also be turned off for a period of time each day to maintain the plants’ circadian rhythms.

Appendix B: Economizer Controls

Economizer controls range from simple solutions that measure only the outside air, called fixed control (that is, they let in outside air if it’s below a certain threshold, such as 65° Fahrenheit) to more-complicated controls that measure both the outside and return air, called differential control (for example, they let in outside air if it’s colder than the return air). In addition to temperature, economizer controls can also measure humidity, called enthalpy control (for example, allowing in outside air only if it’s cooler and drier than the return air). Each algorithm used by an economizer has some error associated with it. Theoretical errors occur when the control logic chooses the more energy-intensive choice, either economizing when it shouldn’t or not economizing when it should. Every control scheme has some theoretical error associated with it, under certain conditions. Practical errors can occur when the sensor measurement is askew. Humidity sensors are notoriously inaccurate,2 so HVAC contractors should buy the best humidity sensors available.

Our analysis of recent research in the field of economizer control logic indicates that the combination of fixed enthalpy and fixed temperature control (when calibrated to the return-air conditions) is likely the best control algorithm for cultivation facilities. The reasons for this are twofold: First, this control algorithm has very little theoretical or practical error. Second, no energy modeling is required to determine optimal setpoint control. The return-air conditions should be measured and the control algorithm set based on those conditions. Because the climate inside a cultivation facility is only marginally affected by outside weather, the return-air conditions should stay relatively constant. And if greater accuracy is desired, the return-air conditions can be measured seasonally and the control setpoints updated.

Appendix C: Indirect Evaporative Cooling

When adding indirect evaporative cooling in parallel with other HVAC equipment, it’s important to configure the controls correctly so that the most energy-efficient mode is selected at any given time. The first stage of cooling should be from the rooftop unit’s (RTU’s) economizer; if the economizer doesn’t provide enough cooling or if the weather is too warm for it to engage, the second stage should be the indirect evaporative cooler. If further cooling is necessary, the final stage should be direct-expansion (DX) cooling of recirculated indoor air. If a facility doesn’t use an economizer, they can simply skip that stage. This is a much more complicated control setup than is normally used for RTUs, so care should be taken to ensure that the contractor is capable of the configuration.

The greatest energy savings are likely achieved by bringing in enough air from the indirect evaporative cooler to eliminate any need for dehumidification, but in practical terms this may not be feasible. This arrangement may require more equipment and extremely high airflow rates, which likely wouldn’t be cost-effective. The most cost-effective ratio of evaporative-cooled outside air to DX cooling would have to be determined by an HVAC contractor for each individual facility.

Care must also be taken to design supply and exhaust airflows in such a way that cool air gets to where it’s needed and hot air is exhausted. When drawing in outside air with an indirect evaporative cooler, indoor air must also be exhausted. Supply and exhaust vents should be positioned in such a way that supply air is drawn across the room and the lighting fixtures; this design will help remove as much heat as possible. Also, supply air from separate DX cooling equipment should be arranged so that it is not being drawn away by the exhaust airstream.


1   Joseph DiMasi (June 11, 2015), CEO, BrightSpace Technologies.

2   Steven T. Taylor and C. Hwakong Cheng, Economizer High Limit Controls and Why Enthalpy Economizers Dont Work, ASHRAE Journal (November 2010), p. 12.