Light Sources


Photosynthesis evolved under sunlight.  We learned previously in our Other Lighting Metrics section that the sun has the approximate characteristics of a black-body radiation source with T = 5900K.  For a reference point, direct sunlight at midday has a PPFD of approximately 2000 μmol/s/m².  However, the spectra can vary with a large number of factors including geographic location, time of day, season, local landscape and local weather.   Above is the spectra of the D65 illuminant which the best approximation of the solar spectra at midday in Western/Northern Europe.  
Fluorescent lights have found ubiquitous applications since their commercialization.  The linear form factor of the most popular models have enabled them to be fit into grow racks, terrariums and other horticultural setups. There exist many types; however, the general principles are the same.  Bulbs coated internally with phosphors are filled with mercury vapor and argon gas.  A spike voltage is applied to the gas which causes ionization and subsequent emission of UV light.  The UV excites the phosphors which then fluorescence visible light as well as emit heat.  The phosphor type dictates the spectrum of the bulb.  The spectra are usually very locally peaked corresponding to atomic transitions characteristic of the phosphors.
Over the years, fluorescent bulbs and ballasts have made continuous improvements.  The adoption of T8 bulb technology over the last 2 decades has greatly improved the range of applications in horticulture.  Generally, 4 foot fluorescent bulbs have an intake power of 32 watts and have an initial luminous flux output of around 3000 lumens (approximately 93 lumens/watt).  PPE values are around .84 μmol/J.  However, bulb performance drops significantly over time.  Published bulbs life estimates are around 20,000 hours (3.5 years with a 16 hour daily photoperiod).  However, I've found that their output drops to 60% or less of its initial value in 10,000 hours of use ( looking at PPE ~.5 as the bulb ages).  Therefore, bulbs should be replaced on a yearly basis.  Costs for bulbs ranges from $3-$12 depending on the type.  Luckily, the widespread use of T8 fixtures for many applications has driven costs down to as low as $25 for a 4 foot, 2 light ballast and fixture. 

T8 fluorescent fixtures are still a staple in tissue culture labs worldwide.  Many growers still use them for growing a wide range of plants and I've found that a combination of 1 warm white for every 1 cool white produces the best results.
T5 high output (T5HO) fixtures have revolutionized horticultural growing in compact spaces.  Typically bulbs are 2-4 feet long.  The 4 foot bulbs are 54 watts and output around 5000 lumens.  Their PPE has been recorded a bit higher at around 1.23 μmol/J when new.  Just like other fluorescents, the light output of T5HO bulbs drops to about 60% of initial output at 10,000 hours so bulbs should be replaced every year.  These units are a bit more expensive than T8 and cost around $70 for 2 bulb, 4 foot fixtures.  Replacement bulbs are around $10 each.

The high light output of T5HO fixtures enabled many genera which were previously only grown in greenhouses or in much larger grow chambers to be grown under artificial lights.  This includes light loving species such as venus flytraps (Dionaea), Drosera, Heliamphora and Sarracenia and the market driver, cannabis.

Although fluorescents are not technically waterproof, I have noticed that most reflector are designed so that splashes from above would be unlikely to do significant immediate damage.  However, I have noticed that the unsealed electronics and wiring can degrade over time in very humid environments.

The major downsides for fluorescents are their inferior PPE compared to LEDs and some HID.  The linear form factor provides benefits for many applications but for supplemental lighting in a greenhouse it's a downside since multi-bulb units suspended above plants can block a good deal of sunlight during daylight hours.  

Variations of metal halide lamps have long been used in horticulture.  They operate by using ballasts to produce an electrical arc through a gaseous mixture of vaporized mercury and metal halides all contained inside an arc tube. The latest renditions to the market are those which incorporate a ceramic arc tube instead of the quartz.  These are 10-20% more efficient than the quartz tube types and seem to have a more balanced spectra. In terms of efficiency, ceramic metal halide units are generally as good or better than fluorescent fixtures.  This depends a lot on the quality of ballast and bulb used but PPE values are generally between 1 - 1.5 μmol/J.

The major application for metal halide lamps has traditionally been use as supplemental lighting in greenhouses or a primary source in grow rooms.  For many years, common practice in the cannabis industry was to use the more blue shifted spectrum of metal halide bulbs for vegetative growth.  However, this practice has become less popular in recent years as improved lighting solutions have become available.

High Pressure Sodium (HPS) lights have long been used for horticultural applications.  They operate under a similar principle to metal halide lamps but use an amalgam of mercury and metallic sodium inside the arc tube.  HPS lights were the most efficient lights on the market for decades with PPE between 1 and 1.72 μmol/J. 

The major application for metal halide lamps has traditionally been use as supplemental lighting in greenhouses or a primary source in grow rooms.  For many years, common practice in the cannabis industry was to use the more red shifted spectrum of HPS for flowering.  However, this practice has become less popular in recent years as improved lighting solutions have become available.
The HPS and metal halide lighting discussed here fall into the general category of high intensity discharge (HID) lamps. Most fixtures range from input wattages of 250W-1000W with the most efficient units usually on the higher end.  Their relatively high efficiency and abundant light output made them industry standards for decades.  The lights are very powerful and are often intended to be used 4', 6' or further from their target crop below.  If the HID is used in a greenhouse for supplemental lighting, this means there's a relatively small angle of sunlight blocked by the light itself.  There are definitely some downsides for HID with the main ones being:

  • They not compatible with shelf or rack configurations.  Generally, HID lights are intended to be suspended 4' or higher from their target.  For taller plants, this is good.  However, this presents a sub-optimal space arrangement for seedlings and plants of smaller stature. 
  • HID produce a great deal of heat.  Plants too close to the lamps can be burned from the excess heat.  The lights heat up grow tents.  Most have vents built around the built which work with blower fans to attempt to remove heat from the grow tent.
  • HID efficiency doesn't stack up against many modern LED units
  • Light output drops over time.  Most lamps are specified to drop to 70% of less of their initial levels at 10,000 hours of use.  This means bulb replacements required on a yearly basis
  • HID explosions are rare but when they occur, it can be dangerous (high pressure hazardous gas inside a glass enclosure).  I've been in a research lab where an HID microscope source blew and I am glad I wasn't in the room.
  • Not waterproof.  For horticultural units, I've noticed an effort to make the electronics enclosed but no IP ratings are given.  When the unit is on, the heat of the HID bulb likely keeps the electronics dry.
Light Emitting Diodes (LEDs)

LEDs are the most promising technology for the future of lighting.  Over the last decade, LED performance has rapidly improved while cost has been driven down.  The modular form factor has given engineers and manufacturers more flexibility in design than any other lighting technology.  Thus, we are currently seeing a revolution in applications and products. 

LEDs function through the principle of electroluminescence.  An applied voltage causes photons to be emitted as electrons "jump" across the p-n junction formed between semi-conductor materials.  The electrons (and their partner holes) have a range of energies across the junction, thus the photons emitted as they cross also have a range of energies/wavelengths.  In the visible, most LED spectra  have a full width half max (FWHM) of around 30nm-50nm.  This isn't quite as monochromatic as a laser; however, to the human eye it looks like a very well defined color.  It should be noted that there's a limited number of semi-conductor material combinations and not all colors can be efficiently created directly.  Phosphors can be used to "spread" photons to longer wavelengths.  Many white light LEDs rely on phosphors over the top of blue LEDs.  Therefore, a signature blue peak is often superimposed over a broader white spectra.  Note that most units have a mixture of LED modules and the type and ratio will determine the spectra of the unit.  

Here's 2 different white grow light spectra above.  Below see a spectra with LEDs only targeting the chlorophyll peak absorption ranges.  This lighting is affectionately referred to as "blurple".

LED form factor and light output can be tailored for the application.  The most common LED bulbs on the market are those intended to replace fluorescent and incandescent bulbs in residential lighting.  These bulbs are of limited value as grow lights except for just a few plants placed very close to them.  Many of these bulbs aren't specified in terms of watts or lumens but as the "perceptible equivalent" to other light sources.  The "perceptible equivalent" metric is muddled because the human eye's perception to brightness is highly dependent on spectra.  Even when spectra are the same, the eye responds to intensity differences more logarithmic than linear.  Thus, a 50% difference in lumens may not be apparent to the untrained eye.  There's a lot of play for marketing purposes here.

"Perceptible equivalent" LED marketing can be even more misleading for grow lights.  For example, a "1500 watt" LED grow light currently popping up as "Amazon's Choice" only consumes 280 watts of electricity and it's suggested that it can replace a 10,000 watt HPS or Metal Halide bulb.  Even if the generous assumption was made that these LEDs were state of the art (2.6umol/J) and the HPS used for comparison was on the low end (1 umol/J), the 280 watt model would at most have a PPF equivalent to a 728 watt HPS fixture.  It's probably MUCH lower given the low price point.  The overarching message here is beware of misleading specifications and run independent tests if possible.  Look for brands that publish specifications in terms of PPF, PPFD and PPE.

LED shop lights intended for illuminating garages have begun to replace fluorescent lighting.  Generally, these products specify light output of around ~5000 lumens and power consumption of around ~50 watts.  For the brands that uphold these specs, that's ~100 lumens/watt efficiency which is on par to slightly higher with the fluorescent lights discussed previously.  I've evaluated the spectra on a few of these and the spectra tend to be very blue heavy.  They could perhaps replace a 2 bulb T8 or single bulb T5HO fixture.  However, I can't give any specific recommendations since the spectra varies and it seems manufacturers disappear as quickly as they pop up.  Many don't actually meet their published lumen specification.  And also I can't comment on the reliability of these units in wet environments.

LED tubes intended to replacement T8 fluorescent bulbs have become more common on the market as of the last few years.  Some bulbs are intended to be "plug and play" and rely on the electricity provided by the fluorescent ballast while others are direct-wire meaning that the input power (usually 120 or 277 volts) is directly wired into the tombstone connections inside the fixture which attach to the tube.  Extreme caution needs to be used to avoid electrocution when re-wiring any fixture or when installing an LED direct wire tube.

The best choice for any lighting application weighs initial cost, operational cost, lighting output, reliability, performance, thermal management as well as the many other factors involved in the application.  While shop lights and replacement bulbs prioritize fixture economy, they aren't intended to be an every day workhouse so less emphasis is given to the other factors.  Dedicated grow lights are engineered and optimized for horticultural applications.  Their heavy use demands some of the highest efficiencies of any LEDs on the market.  LED grow lights enable some possibilities not attainable with prior technologies.  Some are waterproofed in order to withstand the warm and wet environments of greenhouses and grow areas.    Other LEDs may be dimmable or have changeable spectra changes with the turn of a knob.

Greenhouse supplementation LEDs are meant to replace HPS and metal halide lighting.  They are often high powered (300 watts+) and have high efficiency.  These units are generally engineered to be compact so they aren't blocking sunlight in the greenhouse.  These units are also suitable for other high throw applications such as grow rooms and grow tents.

Large LED arrays are often used in grow rooms.  The shared power source can increase efficiency while the wide layout increases uniformity. 

Some grow lights have a sleek form factor.  These are ideal for tissue culture laboratories, seedling shelves or under bench lighting.  Hobbyists using grow racks may also find these very appealing.

Grow Light Efficiency Summary

Source PPE (umol/J)
Strip Light LED ~0.2
T8 Fluorescent 0.84
Ceramic Metal Halide (CHM) 1-1.5
High Pressure Sodium (HPS) 1-1.72
T5HO Fluorescent 1.23
Shoplight LEDs 1-1.3
LED Retrofit Tubes 1-1.8
Horticultural LED* 1.5-2.7
Florawave FS LED 2.3-2.6
Theoretical LED bounds 5.1
*Reference to major brands that publish specifications.  Many off-brands do not test or publish specs.

Above is an efficiency table documenting average efficiencies of a range of technologies.  Specific brands and product lines may differ.  

It's important to remember the efficiency of the system is a product of the efficiency of each individual component.  For example, during my M.S. program, I designed and constructed some of my own own LED units using off the shelf chips.  The LED chip efficiency along with the losses from the power supply, cooling fans, and other components was required to determine the overall system efficiency. (See below about thermal concerns).

It should also be mentioned that LEDs have the longest life any lighting technology.  Generally this is specified as the LT-70 of a unit and is often measured in the tens of thousands of hours (years).

There are a number of important Thermal Considerations that must be taken into account for grow light applications.  These include HVAC costs, environmental considerations, equipment failures and loss of efficiency due to heat sinking issues.

Up next: Thermal Considerations

Navigation List


Blankenship, Robert E. Molecular Mechanisms of Photosynthesis. Oxford: Blackwell Science, 2002. Print.

Blankenship RE, Tiede DM, Barber J, et al. Comparing photosynthetic and photovoltaic efficiencies and recognizing the potential for improvement. Science. 2011;332(6031):805-809.

Both, A. J. Measuring LED Lighting Systems and Developing Guidelines for Evaluation, Comparison and Use. Rep. SCRI‐LED, 11 June 2013. Web. 12 Feb. 2014.

Bugbee, Bruce. 2020. Apogee Instruments. (01-01-2020)

Ellison AM, Farnsworth EJ. 2008. Prey availability directly affects physiology, growth, nutrient allocation and scaling relationships among leaf traits in 10 carnivorous plant species. Journal of Ecology 96: 213–221.

Green, M. A., Emery, K., Hishikawa, Y., Warta, W. and Dunlop, E. D. (2014), Solar cell efficiency tables (version 43). Prog. Photovolt: Res. Appl., 22: 1–9.

Khan, M. Nisa. Understanding LED Illumination. Boca Raton: CRC, 2014. Print.

Koning, Ross E. 1994. Light. Plant Physiology Information Website. (8-13-2014)

Koshel, R. John. Illumination Engineering: Design with Nonimaging Optics. Piscataway, NJ: IEEE, 2013. Print.

Kyte, Lydiane, John Kleyn, Holly Scroggins, and Mark Bridgen. Plants From Test Tubes. Portland:Timber Press, 2013. Print.

McCree, K.J. (1972a) Action Spectrum, Absorptance and Quantum Yield of Photosynthesis in Crop Plants. Agricultural Meteorology, 9, 191-216.

McCree, K.J. (1972b) Test of Current Definitions of Photosynthetically Active Radiation against Leaf Photosynthesis Data. Agricultural Meteorology, 10, 443-453.

Mitchell, Cary A., A. J. Both, C. M. Bourget, John F. Burr, Chieri Kubota, Roberto G. Lopez, Robert C. Morrow, and Erik S. Runkle. "LEDs: The Future of Greenhouse Lighting!" Chronica HORTICULTURAE 52.1 (2012): 6-12. Print.

Mitchell, Cary A. Developing LED Lighting Technologies and Practices for Sustainable Specialty-Crop Production. Rep. NIFA SCRI, 15 July 2012. Web. 12 Feb. 2014.

Narukawa, Yukio, et al. “White light emitting diodes with super-high luminous efficacy.” J. Phys. D: Appl. Phys. 43 (2010) 354002 (6pp).

Ross, J. and M. Sulev. 2000. Sources of errors in measurements of PAR. Agricultural and Forest Meteorology 100, 103-125.

Shenzhen Runlite Technology Co., Ltd., “Runlite Epistar,” SMD 5050 Series Data Sheet, Feb. 2014

Singhal, G. S., G. Renger, S.K. Sopory, K.D. Irrgang, and Govindjee. Concepts in Photobiology: Photosynthesis and Photomorphogenesis. Boston: Kluwer Academic, 1999. Print.

Torres, Ariana P., Christopher J. Currey, and Roberto G. Lopez. "Getting The Most Out Of Light Measurements." Greenhouse Grower (2010): 46-54. Issue. 27 Aug. 2010. Web. 14 Feb. 2014.

Wu, Nancy. "High Brightness Led Tube Light Fixtures Bulbs Replacement T8 8ft 2400mm 35W." - Quality LED Tube Light Fixtures for Sale. Shenzhen Greelife Technology Co., Ltd., 2012. Web. 15 July 2014.

Žukauskas, Artūras, Michael Shur, and Remis Gaska. Introduction to Solid-state Lighting. New York: J. Wiley, 2002. Print. 

Get Connected