The Grow Lights We Like

After more than 50 hours of research and interviews with five experts—including one who designed plant lighting for Antarctica and the moon—I can say that the Hydrofarm FLT24 2-Ft/4-Tube T5 Commercial System with Bulbs ($92) is economical, is low-maintenance, runs cool in small spaces, and provides bright light for stout plants. It’s the one we would buy if we were starting seeds indoors.

Hydrofarm FLT24
The $91 Hydrofarm FLT24 is economical and low-maintenance, runs cool in small spaces, and provides bright light for stout plants. It’s the grow light we’d spend our money on.

*At the time of publishing, the price was $91.

This was not an easy conclusion to reach—as part of our research, we found that a large number of grow light manufacturers and resellers either don’t actually know how to provide appropriate light output information or they simply don’t want to tell you. The typical lumen measurement is great for humans with eyeballs, but not directly relevant to what plants can use. Color temperature is also a common metric, which is only slightly relevant to gardeners and still doesn’t help tell you how powerful it will be to your plants. The photosynthetic photon flux density, or PPFD, is the best measure. (We discuss these things much later on in the guide). In fact, as we were performing the research for this guide, we had one manufacturer ask us for help on measuring PPFD, showing just how opaque this industry is. It’s no wonder gardeners are left feeling around in the dark to figure out which lights work best for what they’re growing.

Also Great
Sun Blaze T5-24
The Sun Blaze is a good runner-up if our main pick is unavailable for whatever reason. It’s essentially the same light, but there’s less information on its long-term performance.
If our main choice sells out, the Sun Blaze 2 ft. – 4 Lamp – F24T5-HO – Fluorescent Grow Light Fixture ($80) is a good bet and is only our second choice because there’s less information available about its long-term performance. That said, it’s essentially the same light as our top pick and gets excellent reviews.

Also Great
BML Light Bar Grow-Max Spectrum
Go with a setup like this if you plan on growing leafy greens like lettuce, kale, spinach, and herbs indoors.
Lettuce farmers and those who grow other greens are a special case. Growing a plant to maturity indoors requires far more photons than sprouting a seedling. If you’re thinking of attempting indoor lettuce, kale, spinach, herbs, or other leafy things, you need stronger stuff than a T5, which simply won’t give you enough photons in a day. Go directly to a setup with two BML Light Bar Grow-Max Spectrum lights, and don’t look back.

Table of contents

Who’s this guide for?

If you’re starting your seedlings on a sunny windowsill and you’re serious about your gardening, you should buy a grow light instead. You’ll see an immediate difference in the color and growth of your seedlings before you put them outdoors for the season. They’ll be greener and stockier, with stronger stems, and they’ll grow faster than they would with limited, filtered light through your window. If you already have a traditional fluorescent light setup with T8 or T12 bulbs (with a diameter of about 1-1.5 in./2.5-4 cm), you should consider upgrading. Over the course of five years, you’ll spend $30-60 more using T5 bulbs than you would with two T8 bulbs, depending on how much your electricity costs. But it’s worth it: Your seedlings will get greener and grow faster simply because the T5s will deliver more photons to their light-loving leaves than the T8s. You will see even more of a benefit with crops that require long indoor growing periods before you move them outdoors (tomatoes, peppers, eggplant, artichokes, cardoons, sluggish offspring…).

How we picked and tested

The best grow light is the light that grows what you want to grow. It produces enough photons in the right spectrum (400-700 nm) to give plants the energy they need to grow and has enough light at both “red” and “blue” wavelengths to make plants grow dark green leaves and thick stems. (You can read more about this in The light plants need.) The best lights give you just as much light as you need, but not more (because you don’t want to waste electricity), and don’t produce excess heat. Unfortunately for gardeners, it’s difficult to get good information about plant lights (see Why can’t you get useful measurements for grow lights?). Lighting companies don’t give out relevant measurements, and the typists stuffing up the internet only seem to want to grow either marijuana, salt-water corals, or orchids. (I’ve been searching for a blog about a garden that combines all three into a single seaside pleasure dome, but I have yet to find it.) Unfortunately, most hobbyist growers—from pot to petunias—do not publish rigorous scientific research into their growing methods. The aquarium lighting folks actually test their setups, but their criteria for a good light is a bit different from a typical tomato-grower’s (ditto for orchids). As such, I consulted five different experts to find out what makes a good grow light and what most gardeners need:

  • Gene Giacomelli, Ph.D., director of the Controlled Environment Agriculture Center at the University of Arizona, has designed plant lighting for the Antarctic research lab McMurdo Station and a “lunar greenhouse” for use on space missions.
  • Ed Harwood, Ph.D., is the chief technology officer and co-founder of AeroFarms, a “vertical farm” supplier that provides equipment for companies growing more than 250 varieties of greens and herbs under LED (light-emitting diode) lights inside warehouses.
  • Neil Mattson, Ph.D., is an associate professor and floriculture extension specialist at Cornell University’s Department of Horticulture.
  • Peter Wardenburg is the president of Hydrofarm hydroponic suppliers.
  • Bruce Bugbee is a professor at the Crop Physiology Laboratory, Department of Plant Soils and Climate, Utah State University. He co-authored an article about grow light efficiency with his graduate student Jacob Nelson, my sixth expert.

Basically, if you’re growing plants indoors, you can choose between fluorescent, compact fluorescent, metal halide, and high pressure sodium (HPS) lamps or LED bulbs (more on this in the “Types of lights” section). These bulbs vary in size, the amount of light they put out, and what wavelengths of light they emit. Using our experts’ plant lighting analysis, I looked for lights that provided the photon output required for regular seedlings, herbs, and tabletop lettuces. After estimating performance, I then calculated the five-year cost of using the fixture including purchase cost, bulb replacement costs, and the amount of electricity needed to give plants the required daily light integral (DLI). (See How we calculated output vs. cost.)

Our pick

Hydrofarm FLT24
The $91 Hydrofarm FLT24 is economical and low-maintenance, runs cool in small spaces, and provides bright light for stout plants. It’s the grow light we’d spend our money on.

*At the time of publishing, the price was $91.

The results of our research and expert input were unequivocal. If you want to give your seedlings a greenhouse-worthy start, go out and get a four-bulb T5 fixture with 24-watt T5 High Output (HO) bulbs (Plain T5 bulbs at 2-foot lengths are only 14 watts and won’t produce as much light.) There aren’t that many 2-foot four-bulb T5 fixtures on the market—and even fewer reviews of T5 systems—but the overwhelming favorite among Amazon reviewers is the Hydrofarm FLT24 2-Ft/4-Tube T5 Commercial System with Bulbs ($92). Users praise the light’s even illumination over an entire seedling flat, its brightness, and the fat, stocky seedlings they raise with this light. A few reviewers complain about customer service for replacing bulbs broken in shipping. Peter Warburg, Hydrofarm president, recommended T5s for gardeners over different kinds of lights. “For seed-starting, T5s are the most cost-effective. They’re practical and affordable … they’re the best practical choice.”

Neil Mattson, Cornell University horticulture professor, raises plants from tissue cultures in his laboratory with T5 lights as well. “It works well for me,” he said. “With four lights on a 2-foot [wide] shelf you get much better growth than two.” This light does put out some heat, however. An orchid grower wrote that this light fixture raised the temperature of his terrarium by 5-7ºF (3-5ºC). However, most seed-starters have their flats out in the open, not in a glass box. In that case, the temperature increase should be minimal. The Hydrofarm T5’s light temperature is rated as 6500 K, or “daylight,” which means the light will look slightly bluish-white. You can see the spectrum on this carnivorous plants fan page, or witness a nausea-inducing animation of the colors of different light temperatures on Wikipedia. When you look at your plants under the Hydrofarm T5s, their colors will look about the same as they would if you put them outside at noon. (Interior decorators for vampires will want to hang a dark curtain around the light to avoid evoking unpleasant memories.) If you prefer a woodsy, rustic brown hue for your plant light housing, instead of dazzling white, the Hydrofarm T5 Designer 2Ft 4 Tube Fixture w/ Bulbs -FLP24 ($108) is almost identical to the FLT24 above. Its one point of departure from the FLT24, apart from its distinguished mottled-bronze exterior, is that it features a three-prong outlet at one corner of the light. That outlet allowing growers to “daisy-chain” lights together and control multiple lights with a single timer, which is presumably why it costs $16 more than the white version. You could also just buy a two-outlet timer for $10 when the time comes to plug in a second light, but you’ll miss the thrill of viewing a block of gleaming mud-colored metal hovering above your tender seedlings. The choice is yours.

How much will it cost in electricity to run this thing?

Depending on the cost of your electricity, it will cost somewhere between $33 and $100 to run these lights for five years and your seedlings will get plenty of photons—a DLI of 6 over the course of a 16-hour day—to attain their maximum potential. (If you want them to attain a fairly average seedling potential, run your lights for just 11 hours a day to get a DLI of 4 and save a third of the electricity cost.) With an expected bulb life of 20,000 hours, you should be able to start seedlings for the next 22 years.

Flaws but not dealbreakers

As mentioned earlier, the Hydrofarm 4-Tube T5 fixture will not bring your tabletop greens to their full potential. If you keep this light on 24 hours a day, you’ll still only get a DLI of 9—considerably below the DLI of 12 that greenhouse growers consider necessary for producing lettuce, herbs, spinach, and other greens. If you’re hankering for leafy greens, read on to see our choice for lettuce farmers below.

For growing greens

Also Great
BML Light Bar Grow-Max Spectrum
Go with a setup like this if you plan on growing leafy greens like lettuce, kale, spinach, and herbs indoors.
If you really want to grow greens indoors, you’re going to have to spend a lot of lettuce … I mean money. As Giacomelli said, “The premium [for indoor plants] is two to three times the price of non-branded tomato or lettuce at the supermarket.” Depending on the cost of your electricity, you’ll spend anywhere from $600 to $1,200 growing lettuce for five years with two BML Light Bar Grow-Max Spectrum lights ($179 each), the cheapest alternative I found that would provide enough photons per day to grow greens. Mount two bars a foot over your seedling tray, and you’ll get a DLI of 12 if you keep these bars on for 17 hours a day and a DLI of 17 if you keep your greens lit for 24 hours a day like Aerofarm and other indoor growers. That’s a great DLI for green bibb lettuce, according to researchers at Cornell University’s Controlled Environment Agriculture Program, but more delicate greens may suffer “tip burn”—yucky-looking brown edges due to plant stress. Researchers like BML Horticulture for their extensive product information and quality build. Mattson called their Spyder light “the Cadillac of LEDs,” and BML’s lights tied for the cheapest electrical cost per photos delivered of the LEDs that Dr. Bruce Bugbee tested. Advanced Aquarist reviewer Dana Riddle wrote, “In a day where many LED fixtures are being offered with more and more frills, BML offers no-nonsense lighting at reasonable prices.” Be warned, though, that your interior decorator vampire friends will love the color of the Grow Max light bars, which tint everything a sort of bloody pink. If you’d like a little more green in your life, opt for BML Light Bar View-Max Spectrum lights instead. Choose a 90º beam angle for more even light distribution across your seedlings.

The budget options

If all you want to do is get your seedlings up to around ½-inch (1 cm) high, I can tell you from personal experience that a fixture with two T8s will do an adequate job. However, they’ll only give you a maximum DLI of 3 if you keep them on 16 hours a day; for a DLI of 4, you’ll need to run them at least 23 hours a day. Your seedlings will be a bit pale and leggy compared to plants grown under better lights, and they’ll benefit from getting outside as soon as possible. If you’re starting small with a single seedling flat, you don’t have that many options for shop lights with a reflecting hood that will direct slightly more light downwards and make viewing your plants slightly more pleasurable.

Also Great

*At the time of publishing, the price was $31.

Hydrofarm FLV22
For getting your seedlings up to around ½-inch (1 cm) high on a budget, the FLV22 is your best option.
The 2’ Hydrofarm 2′ Fluorescent Grow Light Fixture FLV22 ($31) and the Lights of America Fluorescent Grow Light System with 2 T8 Grow Bulbs – 2 Foot ($44) appear to be exactly the same light and are the cheapest available options for a 2-foot shop light. The Lights of America fixture comes with two T8 bulbs; you’ll have to buy them separately for the Hydrofarm Fixture. Run them for five years for eight weeks a year, and your total cost including electricity will be around $60-83. Just don’t come running to me when your Radiator Charlie’s Mortgage Lifter tomatoes don’t fruit in time for the state fair.

How we calculated output vs. cost

Giacomelli was unequivocal: “Are you trying to grow seedlings? Go to Home Depot and buy fluorescent tubes.” Other experts had slightly more nuanced views. Harwood and Mattson observed that although LEDs have improved over the years and now put out as many μmoles/watt of photons as fluorescents, they’re still expensive. “For the average person using lights for a couple of months in the spring, it’s probably not worth it [to buy LEDs]” said Mattson. As Jacob A. Nelson and Bruce Bugbee wrote: “In large greenhouses with small aisles and uniformly spaced plants, the broad, even output pattern typically emitted from HPS fixtures provides uniform light distribution and good capture of photosynthetic photons. In smaller greenhouses with spaced benches, the more focused pattern typically found in LED fixtures can maximize radiation transfer to plant leaves.”

Although these opinions were useful, I wanted to confirm them. If you’re starting a commercial greenhouse, you can calculate the cost of lighting with this calculator spreadsheet created by Bruce Bugbee, Ph.D., director of the Utah State University Crop Physiology Lab. For home use, I put together a spreadsheet for different types of bulbs including

  • Initial purchase cost (bulb plus fixture)
  • Cost of electricity to operate
  • Replacement cost (bulb only)

I assumed that you, gentle reader, intend to grow thoroughly legal seedlings, herbs, and table-top lettuce, and that seedlings require approximately 4 moles/m²/day of photons, while lettuce needs least 12 moles/m²/day. I also assumed that you’re going to start small, with either one or two standard 21-by-11-inch black seedling flats that look like this. That’s 231 square inches, or .14903 m². Four feet is the standard length for grow lights used in the nursery trade, but companies that cater to home growers, aquarium hobbyists, and orchid growers commonly offer 2-foot bulbs and fixtures. Based on their guidance, including Jacob Nelson and Dr. Bugbee’s analysis of plant lighting, I made up a spreadsheet calculating the cost of using the lights for five years, using the daily light integral (DLI), the minimum total number of photons necessary each day that researchers recommend for growing seedlings (4 mol/m²/d) and lettuce and herbs (12 mol/m²/d).  I analyzed lights suitable for a single seedling flat (approximately 11 by 20 inches or 28 by 51 cm) in a small space—say, hung on a bookshelf. I also read reviews on Amazon, Home Depot, and various aquarium hobby sites to see what users thought of the lights.

I made two sets of calculations for seedlings and lettuce. For average usage, I assumed that growers would have the light on for 16 hours a day for eight weeks every spring for five years, or 3,480 hours. For tabletop lettuce, I assumed that, like AeroFarms’ clients, growers would leave the light on for 24 hours a day for five years straight, for 43,800 hours. To calculate the electricity cost of growing seedlings and lettuce, I made two sets of calculations assuming low-cost energy (10 cents per kilowatt hour) and high-cost electricity (30 cents per kilowatt hour). For more miserly users, I also calculated the cost of electricity for just getting plants the minimum daily light integral (DLI). I excluded lights that couldn’t provide the minimum DLI for seedlings in 24 hours. I did calculations for the top-rated T5 High Output fluorescent fixtures, standard shop-light T8 fluorescent bulbs, LED, and high pressure sodium (HPS) lights on Amazon and Home Depot, as well as the top three energy-efficient fixtures (in terms of cost per micromole of photos) in each category as evaluated by Jacob Nelson and Bruce Bugbee at the USU Crop Physiology Laboratory. I eliminated lights that were clearly overpowered for home seedlings (such as 1,000-watt HPS lights designed to light a 10’ by 10’ greenhouse, which would generate enough heat to cook seedlings if they were 1 foot away), but if there were lower-powered lights available from the same manufacturer in the same model line, I included them. I looked at photosynthetic photon flux density data (PPFD) for the lights, which refers to how many photons fall on a square meter of surface in a second, measured in μmoles/m²/sec. In cases where reliable PPFD data was not available, I made separate calculations for the highest and lowest values for photon production per joule for that type of light from the Nelson and Bugbee analysis, multiplied it by the wattage given for that light, and assumed that a narrow surface area underneath the light would be illuminated (an area the length of the fixture with a width equal to 1.3 times the fixture’s height above the seedling flat).

What about LEDs?

LEDs are a fascinating technology, but they’re still far more expensive than other types of plant lights, and they don’t produce enough photons per unit cost to justify their use by most home gardeners. One day, LEDs will produce enough photons per watt of wall-plug electricity to make them economically competitive with fluorescent lights. But as of today, you’ll pay a premium to get enough LEDs to grow plants indoors. One expert who chose to speak off the record said to me, “With LEDs, you’re paying for future research.” Horticulturists and lighting scientists expect that LED lights will become much more efficient over the next decade and start to become competitive with high-pressure sodium and metal halide lights for supplemental greenhouse lighting and indoor growers, but they’re not there yet. Still, there’s hope for a cooler, more efficient light. As Giacomelli said, “LEDs will improve in the years to come, but HPS [lights] won’t.” Mattson observed, “It’s eight times the [carbon] footprint to grow plants completely indoors vs. in a greenhouse.” If you use photovoltaic cells to power your indoor plant lights, you need to cover an area 25 times the size of your plants to power the lights. If you want to try to save the earth by growing your own lettuce, wait.

The competition

I looked high pressure sodium (HPS) lamps as well. Warburg said, “Most applications in commercial greenhouse use HPS [lights],” and according to Giacomelli, they are still the most practical lights for growing plants indoors. For small-scale home seedling and lettuce growers, though, HPS lamps are expensive and take up too much space. Thanks to their heat output, high-watt HPS need to be kept up to several feet away from young seedlings to prevent the lights from scorching their leaves—and all those watts make for expensive electric bills. I calculated the expected cost of starting seedlings and growing lettuce under a variety of High Pressure Sodium (HPS) lamps rated highly by Amazon users and cited as being most efficient in Dr. Bruce Bugbee’s paper. In cases where I did not have PPFD measurements available, I made separate estimates for both high-PPFD and low-PPFD scenarios, based on Bugbee’s HPS data.

The lamps higher than 200 watts were obviously overkill. They’re designed to light a 3’ by 3’ area, produce a DLI of 4 in three hours, and need to be suspended at least 24 inches over seedlings according to this Sunlight Supply Information Sheet. I eliminated the G8LED 240 Watt LED Grow Light with Optimal 8-Band plus Infrared (IR) and Ultraviolet (UV) – 3 Watt Chips – All in One for Veg and Flower, the Horizon Grow Light Kit (250 W), and the ViaVolt 250 Watt HPS Grow Light Fixture. That left three low-wattage HPS fixtures: Horizon Grow Light Kit (150 W),Hydrofarm SBM150S 150-Watt Mini Sunburst with HPS Lamp, and the Sun System 150 Watt HPS Reflector, Ballast and Bulb.

They need to be suspended about 12 to 24 inches above plants, depending on your setup’s ventilation. All three of these HPS systems are a bit overpowered for seed-starting as well: They’ll produce a DLI of 4 in four hours for seedlings, and a DLI of 12 for lettuce in just 16 hours. If you run them for just the amount of time needed to produce the minimum DLI, they’re economical. The total cost for five years of seedlings will run from $90-200 for four hours per day of light, depending on your electricity cost, while lettuce will run you somewhere between $450-1000 for 16 hours per day. Heck, as long as you’re pumping out that many photons, you might as well keep it on for 20 hours and try out some tomatoes (DLI 15, in the “good” range for lycopersicon).

That said, I am not recommending HPS lights per se because their heat output means that they require more attention. If your plants grow too close to these bulbs, they will be damaged—and if you spill water on the bulb while it’s on, the bulb could explode. If you feel like you’d like to experiment with HPS lights, both the Hydrofarm SBM150S 150-Watt Mini Sunburst with HPS Lamp ($78) and the Sun System 150 Watt HPS Reflector, Ballast and Bulb ($85) received enthusiastic user reviews like, “I LOVE THIS LIGHT!”

Some T8 fluorescent fixtures are a little different from our main picks, but they aren’t any better. The 2-foot Flora Hydroponics Fluorescent Grow Light System ($50) has the same features as the Lights of America T8 fixture, but comes with bulbs that have a light temperature of 5000 K, not 41000 K like Lights of Ameriabout That means that light will look slightly bluer. If that matters to you, go ahead and spend $6 more on this light, but don’t expect your plants to notice the difference. If you still haven’t spent enough money on T8 bulbs, you could choose spend $20 apiece on “full spectrum” T12 grow bulbs, which will fit into the same fixture, such as the Agrosun T12 or an Agrobrite 24-inch 20-watt T12 bulb. I’m sure they’re charming, but they’ll only get you to a maximum DLI of 2 if you leave a two-bulb fixture on for 24 hours a day less than a foot away from little leaves. That’s enough light to raise ferns, African violets, and orchids, but seedlings of other species won’t live up to their full potential. If you have $40 to spend on bulbs and another $40 for a fixture, you should just give up lattes for the week, save up, and buy the $92 Hydrofarm FLT24 2-Ft/4-Tube T5 Commercial System with Bulbs instead.

Types of lights

Grow light catalogs confront you with five different types of lights: fluorescent, compact fluorescent, metal halide, high-pressure sodium, and LEDs. Unfortunately, it can be hard for consumers to directly compare their performance. For the complete story, read The light plants need below.

Incandescent bulbs, the type that Thomas Edison invented and that appear over cartoon characters’ heads when they have a brilliant idea, aren’t generally used for growing plants because they give out so much heat for the amount of photons they produce. I excluded incandescent bulbs from the list.

Fluorescent bulbs are those tubes that you remember hissing and flickering from school or the library. They’re glass tubes filled with mercury glass and coated inside with material that fluoresces—that is, lights up when it’s excited by ultraviolet light. When you turn a fluorescent light on, the electrical current makes the mercury atoms emit ultraviolet light, which makes the coating fluoresce. (Canon has a good animation of how this process works on their Science Lab pages.) Fluorescent bulbs can have different colors depending on what substance exactly you use for that coating. Older fluorescent fixtures flicker because they’re not putting enough electrical current through the bulb to get the mercury atoms going.

The reason people use fluorescent bulbs for lighting is that they convert a larger proportion of their electricity to light than incandescent bulbs, so they don’t require as much power to illuminate a spot. Since they’re using less power, they also generate less heat than incandescents. Lighting companies equivocate about exactly how hot their bulbs get, saying that it all depends on the environment around them, the fixture they’re in, etc. I took a laser thermometer to fluorescent bulbs in my house, measuring their temperature after running for two hours at 65ºF (we’re cheap with heating oil around here). The surface temperature of a T5 grow-light bulb was around 100ºF; the ancient T12 bulb in the laundry room was around 140ºF. Neither of them will burn the house down, but your tender seedlings wouldn’t appreciate it having the T12 bulb too close; water at 140ºF is hot enough to cause a third-degree burn in five seconds, according to the Burn Foundation.

Fluorescent tubes also last longer than incandescent bulbs, running for up to 20,000 hours, about ten times as long as incandescents. You’ll see references to T12, T8, and T5 fluorescent bulbs; the T just tells you that you’re measuring the diameter of the bulb in eighths of an inch. T12 bulbs are 1.5 inches in diameter, T8 bulbs are 1 inch, and T5 bulbs are ⅝ of an inch and are built a little differently. You can generally plug T8 and T12 tubes into the same fixtures, but not T5 tubes. The smaller-diameter fluorescent tubes put out more light per watt than the larger tubes.

Some T5 bulbs are T5HO, where HO stands for “high output.” The T5HO tubes do put out more light than the T5 bulbs, but they also use more energy to do so. You’ll get less light per watt from these bulbs, according to this chart put together by the Rensselaer Polytechnic Institute Lighting Research Center. Even more confusing, fluorescent bulbs’ light output varies with the room temperature. T8s give out more light at 77ºF, T5s at 95ºF. T5 performance falls off rapidly between 86-68º F (30-20ºC)  from 95 percent to 75 percent of their maximum light output. At 59ºF (15ºC), you’ll only get 60 percent of the light out of a T5 as you would at 95ºF, while T8s are still putting out 90 percent of their maximum light.

The moral: Keep your seedlings warm. Fluorescents do eventually wear out. As the glass on the tube gradually absorbs mercury, there’s less mercury vapor in the tube to create ultraviolet light. Both T8 and T5 bulbs continue to put out about 90 percent of their maximum lumens over the course of their 20,000-hour lives, while T12 output falls to 75 percent by 5,000 hours. For that reason, and the sheer bulk of T12 lights, I excluded those bulbs from my sample.

Compact fluorescent bulbs are also available as grow lights, but they use more energy to create light than the tubes (more watts per lumen). They last about 10,000 hours, or half as long as the fluorescent tubes. The only real advantages to CFLs are that they can be plugged into conventional light sockets and that they contain less mercury than the tubes—but they do contain mercury and need to be disposed of carefully. Because of their inefficiency and relatively short useful life, I excluded them from the grow light sample.

Metal halide (MH) and high pressure sodium (HPS) lamps are both types of high intensity discharge (HID) lamps. HID lamps contain tungsten electrodes inside a tube filled with gas and metal salts. When electricity flows through the lamp, electricity arcs between the electrodes through the gas, melting and heating the metal salts until they form a glowing plasma. They put out a lot of light, but they use of energy (250 to 1000 watts per bulb) and consequently they give off a lot of heat. You can’t put a 600ºC HID bulb on top of your seedlings, or your laptop, or pretty much anything except another HID bulb. You need plenty of space and air circulation.

Metal halide lamps are filled with vaporized mercury and metal halides (compounds of bromine and/or iodide). HPS lights contain—surprise!—sodium instead. MH lights put out somewhat more light in the blue part of the spectrum than the HPS lights, which produce more yellow-to-red light. But, as Giacomelli said, most commercial applications use HPS because plants can do reasonably well without the additional blue light.

If you want to see what an HPS light looks like, go to a parking lot at night and look up. HPS lights are frequently used for outdoor lighting, giving everything that special orange glow. Many people find this particular hue aesthetically unpleasing. If you’re going to be looking at your plants under lights, consider what color you’d like to see. That said, HPS bulbs are much more popular for horticulture, in part because the MH bulbs last just 10,000 hours on average, compared to about 18,000 hours for the HPS, and their light output decreases fairly rapidly with use. According to this chart, MH bulbs’ light output will deteriorate by about 25-51 percent before they’re halfway through their expected use. HPS lights will still put out 70 percent of their initial lumens at the end of their useful life. I eliminated MH bulbs from the sample.

LEDs shoot out photons at single wavelengths, not across a spectrum.
LEDs, or light-emitting diodes, are the hot new light. Well, really, they’re the cool new light, because they give off very little heat. As How Stuff Works writes, “[LEDs] are illuminated solely by the movement of electrons in a semiconductor material, and they last just as long as a standard transistor.”

That means that LEDs have very long useful lives: about 50,000 to 100,000 hours, in theory. As Bugbee said, “LEDs are too new a technology to be sure they’ll last 20 years.”

Unfortunately, there are two important things to know about LED lighting:

1) LEDs shoot out photons at single wavelengths, not across a spectrum.

2) LEDs can’t deliver as many photons per watt as other lights … yet.

The first point is important because plants need some light in both the “blue” spectrum (400-500 nm) and the “red” spectrum (600-700 nm) to survive and thrive (see The light plants need, below). But that may not be all they need. There’s research suggesting that plants growing under LEDs do better when they get some far-red light (700-740 nm) every day; lettuces grown with a half-hour of this far-red light had 28 percent more mass at the end than controls. Most LEDs don’t provide far-red light. Greenhouse growers like this, according to Harwood. “It helps with integrated pest management [IPM]” Harwood said. “LEDs with no infrared cut insect visibility.”

Bugs can’t see the leaves as well if you leave out near-red and green spectrum lights, Harwood said, and some diseases and funguses are inhibited if you leave out certain spectra as well. “White” LEDs are actually groups of three or more single-wavelength LEDs put together to make light that looks white to humans—and to make your plants look like they belong on earth instead of on Death Zombie Planet.

The second point—that LEDs can’t deliver as many photons per watt for a given area—is partly due to the state of technology.

Bugs can’t see the leaves as well if you leave out near-red and green spectrum lights…
For incandescent, fluorescent, HPS, and MH lights, if you make a fixture that can withstand more current, you get a brighter light when you put more energy into the system. In LEDs, though, there seems to be a limit to how much light you can get out of a system. After a certain point, the excited electrons start running into each other and producing more heat instead of light, as Popular Science explains.

There’s plenty of research into making better LEDs, but for now, the options for the consumer aren’t very powerful. Instead of making LEDs higher watt, you have to add more of them to get more light to your plants, which costs more and takes up more space. Giacomelli tries to cluster LEDs together to look like an HPS light.

LED manufacturers also talk about the beam angle, or how far the light spreads from the center of the LED. The larger the beam angle, the more ground the light covers, but the less light you’re getting per unit area. For the purposes of this review, I’m assuming that you’ll be hanging your lights within a foot of your plants, so you’ll need a wide beam angle (90º) to get light to the edges of your seedling tray. If you’re hanging your light higher, you’ll want to choose a narrower beam angle to cover the same surface area. light_source_graphj

The light plants need

Grow lights exist to get photons to plants. Unfortunately, it’s extremely hard to tell if a grow light will do that from the information retailers list on their web sites. Listings commonly say something like “Lumen output rated up to 8000” or “6400 K bulb” or “280 Watts Super Bright Output.” None of these statements really tell you what to know: How many photons will get to your plants’ leaves.

As you may remember from high school physics, light is a type of electromagnetic radiation that acts as both a wave and a particle. Light comes in different wavelengths, and the wavelengths that humans can see (called the visible spectrum) roughly measure between 400 and 700 nanometers. Make the wavelengths shorter—say, 350 nanometers—and you’ll get ultraviolet radiation, which humans can’t see, but is visible to bees, birds, fish, some amphibians, reptiles, and cats. Make the wavelengths slightly longer (700 nm-1 mm) and you’ll get infrared radiation. Some snakes, insects, and bats seem to be able to perceive infrared radiation, but not with their eyes—they have special heat-sensing organs.

Plants don’t have eyes, though, or heat-sensing pit organs for catching prey … yet.
Plants don’t have eyes, though, or heat-sensing pit organs for catching prey … yet. They absorb photons via chlorophyll in their leaves and stems. Plants look green because they reflect light in that part of the spectrum, although they also absorb photons at “green wavelengths” (more on that in a moment).

But wait! There’s more! Plants don’t just use light to make energy with their chlorophyll. They also have a color pigment called phytochrome that can absorb red light or infrared radiation. If the plant gets too little red light, the phytochrome signals the plant to elongate and grow up toward the sun to get more light. This sequence is how plants “get leggy,” as gardeners put it. “Leggy” plants and seedlings are tall, thin, pale, and spindly, and their stems tend to snap. They do not actually grow legs … yet.

In general, sprouting seedlings, greens like lettuce and kale, and low-growing herbs need less light than sun-lovers like tomato plants. Horticultural researchers talk about the daily light integral (DLI), or the total number of photons a plant needs to grow and stay healthy. To find out if your lighting set-up is giving your plants a high enough DLI, multiply the manufacturer’s PPFD (μmoles/m²/sec) by the number of seconds you run your lights per day and voila! Light! Thyme needs a  DLI of at least 12 moles/m²/day. Cornell University’s Controlled Environment Agriculture Center recommends a DLI of 17 moles/m²/day for spinach, while researchers at the University of Arizona recommend a DLI of at least 13 moles/m²/day for leafy greens and seedlings, and 30-35 moles/m²/day for tomatoes and cucumbers. For a more thorough explanation of DLI and a list of minimum requirements for flowers and some herbs and crops, see this Purdue Extension paper on measuring DLI.

Light also controls the blooming in some plants. As BML puts it, “There are long-day plants (which require short nights to flower), short-day plants (requiring long nights) and day-neutral plants which have no specific requirement for the photoperiod.”

Now, let’s talk about how to think about grow lights. Lots of grow light ads talk about “lumen output.” They are not helping. As Gavita Professional Lighting puts it, “Lumens are for humans.” Lumens are a measure of  the amount of visible light a bulb produces, which is an interesting thing to know, but it won’t help you figure out if your grow light works. First of all, what matters isn’t how much light a bulb puts out, but how much is actually reaching the surface of your plants. That’s measured in luminous flux (also known as lux), or lumens per square meter. Secondly, lumens are measured in terms of light wavelengths that are most visible to humans―an utterly useless measurement for plants. Human eyes are most sensitive to light with a wavelength around 555 nm. We’re good at detecting green, orange, and yellow, not so hot at seeing red, blue, or purple. That’s one reason why many firefighters wear neon green instead of red hats nowadays; it makes them easier to see in low light.

Plants are different. Here’s a graph showing how sensitive human eyes and cucumber leaves are to light in the visible spectrum (wavelengths about 400-700 nm). The cucumber is relatively less sensitive to wavelengths around 500-550 nm than humans, which means that it might not notice when traffic lights turn green. That is, if it had eyes, and a car.sensitivity But, as you may have deduced from the lack of philodendron mascara options, plants don’t have eyes. Instead, they use photons to power their photosynthesis. Photons at some wavelengths are slightly better than others for powering your plants; they reflect back some of the photons in the “green” wavelengths right around 555 nm, but they still absorb most of them for energy. With a few key exceptions, plants aren’t that fussy about the exact wavelength you shine on them as long as they get enough photons. (Remember how the Plants in Plants Vs. Zombies have to capture little falling suns to get energy? That’s how it works, without the zombies … I hope.)

If you want to think about light the way a plant would, if plants had thoughts, you need to measure light in terms of photosynthetically active radiation (PAR). PAR is adjusted for the wavelengths that plants use for photosynthesis. PAR is expressed in units of photosynthetic photon flux density (PPFD), a measure of how many photons fall on a square meter of surface in a second: micromoles/meter squared/second, or μmol/m­²/s­, also known as “einsteins” with a small e. (Remember from high school physics how light is both a wave and a particle? Did you ever really understand that? We’re talking about the particle part now.) PPFD for land in full sunlight in the middle of the summer is about 2000-2500 μmol/m­²/s . Many sites like to give their lights’ spectrum in terms of “color temperature” or Kelvin, usually expressed as something like “bulb 6400 K.” That’s slightly helpful for gardeners, but only slightly. It tells you what color the light will look and what proportion of the light will be at what wavelength, but it doesn’t tell you how many photons will be arriving at your plants’ leaves.

Why can’t you get useful measurements for grow lights?

So why do plant light sellers insist on telling customers their lights’ lumens, and not PPFD? It’s not that difficult to take light measurements. For about $500, you can buy the same Licor 190 PAR meter that Utah State University Crop Physiology Laboratory faculty used to evaluate HID and LED plant lights, and measure how many μmoles/m²/s at standard intervals to get a PAR map, like the excellent examples at BML. It can be a little tedious, but it wouldn’t take more than an hour or two. Yet very few manufacturers seem to do this.

There are at least two reasons why this might be the case:

1) Some retailers don’t understand how light affects plants very well. Many garden suppliers talk about how many lumens their lights supply and nothing else (here are two examples). As I’ve discussed, this number doesn’t tell you anything about how well plants grow. Some lighting companies don’t even know how to measure light—or at least they don’t bother telling their staff that it’s possible. One marketing rep wrote to me, “We’re very interested in getting the PPFD measurements for our grow lights so is there any way you could help us get them? Do you know someone who can do the measurements for us or do you have a possibility to do the measurements yourself? Please let us know.” I’m sure the marketing rep isn’t at fault, which is why I’m not disclosing the company’s name. But if an auto salesman said to me, “No, I don’t know how many miles this car could go on a gallon of gas. We don’t have an MPG meter. Could you test this car out and let me know?” I would leave the premises very, very quickly.

2) Some retailers seem to be afraid of what you would find if you did do the measurements. There are some sad excuses out there.1

Care and maintenance

Assuming you don’t back into your lights with an anvil, your grow lights shouldn’t need any care for years. If your plants start looking a little pale, the problem is probably a bulb that is reaching the end of its life, not the fixture itself. You may not be able to see the difference in the quality of the light with your human eyes, but your plants will notice if there’s less light reaching them. Replace a bulb and see if things improve. For a gross measure of the energy reaching your plants, use a light meter app on your phone (there are plenty available for Android and iPhone that use your phone’s built-in light sensors). It won’t be accurate, and it won’t measure the photons reaching your plants, but it will give you an idea of whether replacing a bulb is making a difference.

Most indoor growers hang their lights on adjustable cables or chains over their plants so that they can raise the lights as their plants grow. Hang fluorescent lights and LEDs just a few inches above seedlings; HPS lights produce more heat and need to be placed farther from plants—at least a foot for 150-watt HPS lights. When in doubt, put your hand under the light at plant level. If it feels hot, hang the light farther away.

All fluorescent bulbs contain mercury and will release mercury into the air when you break them. They need to be recycled at special facilities for toxic waste, although many hardware stores and municipal public waste departments will accept them. To find out how to recycle fluorescent bulbs—or anything else—visit

Wrapping it up

The Hydrofarm FLT24 2-Ft/4-Tube T5 Commercial System with Bulbs will get your seedlings started fast, keep them cool and looking natural under its white light, and make them grow until you’re ready to share them with the sun—all at less than 100 watts. Green in your garden, green in your pocket. What’s not to like?


1. Example A

We agree that measurement of PAR (Photosynthetically Active Radiation) using a quantum meter is a better method of testing light output for growing plants than using a standard light meter, but it is not the best method. The BEST method is to try the product and determine that it actually works. If a marketing rep said, “We’re not going to give you evidence that our product works, so you’ll just have to try it,” would you buy that product? If so, I recommend you go back to No-MPG-Autos and apply for a sales job. They need people like you.

Example B LED Plant Light Bar FAQ: I would like to compare LED-PlantBars to high-pressure sodium lamps. What is the micromols per square meter per second (μmol·m-2·s-1) of photons, or the PAR watts/square meter of each bar? Direct comparisons are not possible since LED-PlantBars only emit those light wavelengths needed for plant growth and health. This answer is nonsense. Plants use photons from all wavelengths between 400-700 nm for photosynthesis, and PAR meters are designed to correct for plants’ ability to use different wavelengths. Your plants need to have a certain number of photons from that spectrum to hit their leaves every day. If these LED-PlantBars don’t give out enough photons, it doesn’t matter how pretty they wavelengths look, the plants won’t grow. If you don’t measure μmol·m2·s(otherwise known as PPFD), you don’t know how many photons are reaching a surface below the light. And metaphorically, you’re right back at No-MPG-Autos, negotiating an auto loan who says you can’t compare their auto loans to other auto loans because NMA calculates its interest in troy ounces of gold, not dollars, when the real issue is the percentage rate.

Example C LED Light FAQ’s Only a small percentage of the light produced by an HID light is absorbed by plants which means that 80 percent of the Lumens are wasted.” This sentence doesn’t make any sense—and if it did, it wouldn’t be true anyway. Plants don’t grow on lumens, they grown on photons.  If you rewrote this sentence “80 percent of photons are wasted,” you’d probably be misinterpreting the graph at the top of this Wikipedia page on PAR, which makes it look like the middle of the spectrum is useless to plants. Unfortunately, the Wikipedia graph is based on graph from unpublished data from a 1999 textbook, so it’s hard to tell exactly what was being measured. Here’s a typical gaudy color version of the graph from an LED sales site:

Wow! That graph really makes it look like plants just can’t use most visible light. And it’s wrong. That graph is probably based on how chlorophyll in a solution responds to light, not how plants respond to light. Yeah, sure, chlorophyll A and B don’t absorb photons with a wavelength of about 550 nm very well—but leaves have a lot of chlorophyll. As the Australian Society of Plant Scientists wrote, “Leaf absorbance has a hyperbolic dependence on chlorophyll content. For most leaves, 80–85 percent of 400–700 nm light is absorbed and it is only in leaves produced under severe nitrogen deficiency where there is less than 0.25 mmol Chl m–2 that absorbance falls below 75 percent.” There’s also evidence that plants change the balance of different types of chlorophyll and pigments in their leaves in response to the light that hits them. Here’s a graph that compares leaves vs. chlorophyll.
leaf_solution In short, plants use light from all across the visible spectrum pretty well, and light at 550 nm isn’t “wasted.” Jump back.

*At the time of publishing, the price was $91.

To send this guide via email, fill out the fields below:
Message Sent!
Oops! Please try again


  1. Peter Wardenburg, President, Hydrofarm hydroponic suppliers, Interview
  2. Munns, R., Schmidt, S., & Beveridge, C., Leaf anatomy, light interception, and gas exchange, Plants in Action, 2nd edition, 2014
  3. Hernandez, R., Plant Lighting Basics and Applications, University of Arizona
  4. Torres, A., & Lopez, R.G., Measuring Daily Light Integral in the Greenhouse, Purdue Extension, 2010
  5. Ricardo Hernández, Plant Lighting Basics and Applications, The University of Arizona
  6. Sander W. Hogewoning, Emilie Wientjes, Peter Douwstraa, Govert Trouwborst, Wim van Ieperen, Roberta Croce, and Jeremy Harbinson, Photosynthetic Quantum Yield Dynamics: From Photosystems to Leaves, American Society of Plant Biologists, 2012

Originally published: March 4, 2015

  • ericdfields

    I’ve got a living wall setup with decorative and edible plants and a couple of TaoTronics LED Grow Lights on a timer:

    For the low price/performance ratio, these are the best grow lights I found while researching and couldn’t be happier. I used basic clamp fixtures and manufactured an armature for them with some misc. parts I had around the house.

  • WhatChrisLikes

    What size is the growing area that these lights cover? From the Amazon page it looks like the enclosure is 23×13 inches, but I’m hoping that for $100 the light footprint would be a little bigger.

    • David Hughes

      You would be able to fit two standard seed trays, but not much more. So that’s an effective growing area of 20×40. Too far to the sides and the light drops off rapidly.

      • WhatChrisLikes

        Thanks David.

  • David Hughes

    I’m surprised you focused specifically on 2-foot fixtures. 4-foot fixtures double the size of the growing area for about $20 more, and I’d think gardeners serious enough to start seedlings indoors will want more than 2 flats of seedlings.

    This year I did okay with a 4-foot 2-bulb T8 fixture getting plants up to the 4th true leaf. For next year I was going to go with a T5HO fixture from a commercial high-bay company (more expensive retail, but I get them at cost through my work) which adds a better quality ballast and a specular reflector for even greater efficiency, but I had no idea about the drop off in light output in T5s.

    I think instead I will go the HPS route, because my basement is COLD (low 60s, high 50s F) when I’m starting seedlings in the early spring.

    Love all the detail.

  • Carol M

    what would you recommend for growing cool weather crops in a hobby greenhouse, 10 x 12, in Texas?

    • tony kaye

      Via our researcher:

      It has good maps of how much light different regions of the US receive, on average, during different months of the year, and how much light different plants need, so growers can figure out for themselves what they need.

      The grower should be aware that both HPS and MH lights give off a lot of heat. That can be a good thing in a greenhouse in winter; come April in Texas, it might not be so good.

  • Jeffrey Mastin

    From my experience, growers of greens (microgreens, lettuce, herbs, etc., also tissue cultures) that rely on lights year-round quickly get fed up with replacing their T5 fluorescent tubes and ballasts. It might be the environment that grows the plants best, but the lights probably average a year of life when used 12-24 hrs/day every day.

    The most effective and affordable light I know of for this application is the 11W TotalGrow Broad Grow Spectrum Light bulb. MSRP is $42, each light covers about 1 square foot with 13.5 µmol/s output, and it’s a 40,000+ hr lifetime, 3 year warranty product. That’s at least the growing power of the BML 7″ bar with its MSRP of $125. I know a microgreen/herb grower using a few hundred to grow for local restaurants, a hydroponic lettuce grower using them for herbs, microgreens and lettuce, and a lot of people starting to use them for small home gardens. For heavy use applications, the payback over T5s can be a year or less. It’s also very convenient to use in a standard screw-in socket mounted wherever you want it to get light only and exactly where you need it.

    I’d also argue that the Cornell suggestion of 17 mol/day DLI for lettuce is more theoretical than practical. That was the most light that a bibb lettuce could use without tip burn with a perfect environment (air movement is critical). Other lettuce can only take 15, and without the air movement that drops to 12. Most importantly, the lettuce growers I know do not target that because it is not economical or practical for them to purchase that many lights and optimize their environment to that level.

    • Tyler Barlow

      A quick Google search exposed that you work for TotalGrow. Not sure about the moral and ethical implications of writing comments/reviews against your competitors, but I do have a problem with the similarity between the chart above referencing photosynthetic response of chlorophyll in a solution and the chart you guys use on your website. It’s as if you are marketing in the exact same way that this article denounces. I currently live in MI and I see you guys are a MI business. Usually I would be stoked about finding a local business to patronize, but your science and questionable ethics has cast serious doubt IMO. With that said I think your light does look interesting and has promise. I personally wish it had a little more balance to the spectrum offered though. I’m not the type of guy who thinks he can outsmart nature.

      • tony kaye

        Wow he snuck his comment in there on the DL. Not OK! Thank you!

  • sophiaperry

    I guess the various alternatives regarding the Grow Lights is quite a benevolent one and the added guide along is again something one can adhere to ascertain all the aspects about it. Which light is going to be good for you, how to use the same, and many other aspects are there that are supposed to be known only because of the above post. One must go through the same.

    To know more about the grow lights visit:

  • Davilyn Eversz

    Probably need to re-think your view on LED’s. A good LED light like Hydrogrowled will grow anything. You can even grow watermelon to fruiting stage. There is a YouTube video of some high school kids that grew 181 cucumbers on one plant. Research is nice but if you don’t actually have experience growing with T5s or LEDs, articles should be written by those who do have experience.

    • tony kaye

      Via our expert:

      I go into detail about why I think LED’s aren’t ready for prime time. They don’t give enough light for the money you spend, and most home gardeners don’t need them to start seedlings. Yeah, sure, you can raise cucumbers under lights, but you’ll have to spend several dollars per cucumber, and you’ll use a lot of energy in the process.

      • Davilyn Eversz

        Ah no. LED’s cost next to nothing to run. I have two huge ones and my bill went up $10 a month. T5’s, although rated for a lot of hours don’t actually run that long as stated in specs – most people have to replace them every year. LED’s will run about 50,000 hours before having to be replaced. So yes, I might spend $550 for one fixture but it is going to keep going for about a decade before I have to replace anything. I also like that they are not hot at all so aren’t going to heat up the room to an uncomfortable temperature for new seedlings. Mine has three settings – from seedlings to fruiting. With a grow tent you can easily grow 3×3 space per light – which is a lot more space than a standard T5 fixture. It just depends on what you are looking for in a light; everyone has different needs. Mine are predicated on long-term cost, whether they are easy to maintain but mostly end result. LED’s don’t produce leggy seedlings and that was an important factor to me. Plus now I don’t have to be concerned if the weather doesn’t get warm enough to put the seedlings out – I can just switch the setting and let them continue growing until outside conditions are optimal for transplanting.

        • tony kaye

          Duly noted thank you!

  • yashrg

    Why does this article now say originally published: March 4, 2015?

    • tony kaye

      Must be an error. Checking right now

  • gmwebster

    I hate to gripe, but in the age of marijuana legalization, dancing around the issue of the best grow lights for MJ is just silly.

    • tony kaye

      What do you mean by dancing around the issue?