To investigate the LED panel characteristics beyond the manufacturers specifications, tests were carried out on selected lights in a laboratory environment. The tests undertaken on other candidate lights are laid out for reference in Appendix B. In the first instance, the spectral characteristics are determined using a light spectrometer (Uprtek AI-MK350D). The spectral output of a typical panel is visualised in Fig. 3.6,
where it can be seen that most energy packets arrive from the blue wavelength band, with a peak at 448 nm. However, the phosphor coating that has been applied to the LEDs by the manufacturer, in order to yield a white colour output, results in some additional light arriving from the green and orange–red wavelengths.
Secondly, a broad wavelength photosynthetically active radiation (PAR) meter is utilised to measure the PPFD magnitude (µmols m−2 s−1) over the operating
range. The latter test was subsequently extended in order to assess the distribution of PPFD in the area directly below and adjacent to the light source, as well as to quantify the accumulation of PPFD when two light sources are placed next to each other. Note that the PAR meter utilised for this research only counts the moles of photons within the 400-700 nm band, hence any relatively little light arriving
400 450 500 550 600 650 700 750 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Relative intensity Wavelength (nm)
Figure 3.6: Spectral output of lights installed in the grow–cell.
from the near infra-red region is not added to the cumulative PPFD. Fig. 3.7 shows one light panel fixed at 0.2 m above the centre of the measuring board, with the holes representing 104 measuring points below and adjacent to the sides of the light panel. These measurements were taken in a dark room without any other light sources present and with a dark coloured measuring board in order to minimise its reflectivity. Fig. 3.8 shows the spatial distribution of PPFD at a distance of 0.2 m below one panel and under different power supply levels. The area projected in each subplot has the same 0.5 m ×0.3 m dimensions as the light panel. The 40 measuring points pertinent to the panel were interpolated to yield the PPFD distributions. Fig. 3.8 shows that most of the energy is delivered at the centre of the illuminated area, as would be expected. Table 3.1 states the observed PPFD levels at the centre (maximum PPFD) and corner (minimum PPFD) of the board for each power input.
Related to these results, Fig.3.9 shows that a power supply set to 73% of the standard (maximum) setting, yields a light output only slightly lowered (96%) from the maximum PPFD, indicating considerable scope for energy savings by suitable tuning of the system. In fact, it is clear these light panels operate most efficiently within the 50% to 73% power supply band, which can deliver a PPFD level between 120 to 210 µmols m−2 s−1. Here, the 50% limit is based on discussions with growers
Light Panel
Measuring Board
Figure 3.7: Schematic diagram of the 0.9 m× 0.5 m board for measuring PPFD magnitude, with one light panel and 104 measurement points.
Voltage (%) min (µmols m−2 s−1) max (µmols m−2 s−1)
40.0 33.4 113.0 46.5 38.0 128.0 53.0 42.7 144.4 60.0 48.5 163.8 67.0 56.0 189.0 100.0 64.2 217.0
Table 3.1: Minimum and maximum light intensity for Fig. 3.8.
divine) whilst the 73% is the inflexion point in Fig. 3.9. All twenty transformer units require 15 kW to provide maximum power supply to the lights but at 67% power supply, for example, this drops to 11 kW. For plants that do not require high PPFD levels, it is possible to reduce the power requirements even further.
Although most of the light energy is delivered directly below the panel, some light is naturally dispersed towards adjacent sides. In general, the magnitude of this dispersion depends on the distance of the light from the illuminated area and the angle at which the individual LEDs are manufactured to emit. In this case, the PPFD ramps down to approximately 2 µmols m−2 s−1 at 0.3 m adjacent to each
i 40 60 80 100 120 140 160 180 200 ii iii iv v vi
Figure 3.8: Spatial distribution of light intensity (µmols m−2 s−1) at a distance of 20 cm. Subplots i through to vi are for voltage levels of 40%, 46.5%, 53%, 60%, 67% and 100% intensity of the variable power supply. Each subplot indicates the light intensity over the 50 cm (vertical axis) by 30 cm (horizontal) light panel.
Figure 3.9: Single panel light intensity plotted against supply voltage expressed as a percentage of the maximum, highlighting the most energy efficient intensities (shaded).
i 50 100 150 ii 50 100 150 200 iii PPFD 50 100 150 200
Figure 3.10: Spatial distribution of light intensity (µmols m−2 s−1) for 20 light panels in one layer, with i) 60%, ii) 67% and iii) 100% of the maximum power supply. Each subplot indicates the light intensity over 50 cm (vertical axis) by 6 m (horizontal).
side of the panel. Extrapolating from these results, Fig. 3.10 i-iii displays the light intensity distribution along the length of one side layer within the grow–cell (i.e. 20 panels over 6 meters length), for three different power supply levels. On account of the cumulative effect of the dispersed light when light panels are arranged next to each other, it is observed that for all three power supply levels the overall light intensity is increased by 11%.
3.6.1
Preliminary LED Growth Test
A laboratory based test using these lights was conducted for non-stop Tuberous begonias (Begonia tuberhybrida), a type of begonia grown for propagation by the third party company that also ran the full scale growth trial. Production takes 13 weeks in their greenhouse environment, whilst it took 50% less time to grow them to the same stage under the LED lights. Fig. 3.11 shows the difference between 4 week begonia plantlets under the LEDs (middle tray) and those grown in the greenhouse. This type of experiment suggests that the energy consumption of LED lights can potentially be compensated for by increased production rates. However, this is an
Figure 3.11: Four weeks old Tuberous begonia plants grown under white LEDs (middle tray) and in a greenhouse (the other trays).
illustrative result only, used to test the LEDs before proceeding to the growth trials.