Fluorescent lighting

In another thread, Larry Davis mentioned some things about fluorescent lights. I thought it might be better to give fluorescent lighting its own thread so it would not be hidden under the First Germination title.


About lights. Many years ago there was an article in some journal of the Am Soc Hort Sci where all kinds of fluorescent, and some other lights were compared. Plain, cheap, cool white fluorescent gave the most efficient photosynthesis. Other colors I think were less efficient in converting the mercury discharge to visible light so they couldn’t be quite as bright- check the lumens on the package. But how old the lights are may be a bigger factor. They lose about half their output before they finally die. So new lights will do a lot better than ones that have run for several seasons, if light is what is limiting growth in your setting. For me its usually that I don’t use enough fertilizer. The various purple-looking grow lights may make plants look better, for sure that works with African violets. But they don’t need or want high light intensity.[/quote]

Larry, I don’t know how many years ago that study was, but cool white fluorescent tubes (4100K light temperature) still give about the best bang for the buck.

When I started growing rose seedlings under lights (1972), the recommendation was to use cool white tubes along with some small incandescent bulbs (the incandescent was to supply more light in the red part of the spectrum). However, in those days it was common for the 48-inch shop light fixtures to come with bulbs, the standard cool white. And they worked well even without addition of incandescent bulbs to a growing area.

In the 1970s it was possible to buy specialty fluorescent grow lights which were supposed to make plants grow better. The most famous of these (still available) is Sylvania’s GroLux. It emits a pinky-purple light that makes African violets look spectacular, and grow well. It is also sold for lighting of fish tanks. Another specialty light was Westinghouse’s Agro-Lite–it gave a pale pink light. Yet another was Duro-Lite’s Vita-Lite which claimed to combat depression by approximating the light temperature of mid-day sunlight, 5500K (there was no mention of the value of rose seedlings in the battle against mid-winter depression). I used both Agro-Lite and Vita-Lite (got them for clearance prices), and they did OK but only slightly better than Cool White. Agro-Lite is no longer made, but the tradename is used for some incandescent spotlights which are claimed to make plants grow better but probably are not worth the price. Vita-Lite is no longer made. GroLux is still available, but of course specialty lights are sold for specialty prices.

Nowadays it’s possible to get fluorescent tubes in a variety of light spectra. If you want to use cool white with warm white to boost the level of red, that is fine. One of each in a 2-tube fixture does the job, and your seedlings will grow and bloom well. But the warm white tubes cost more. You can also buy tubes which come close to daylight (5000K and 5500K), and you can get these (such as the Litetronics L-359 F32T8 Energy-Lite) for only a slightly higher price than the cool white bulbs ($3 or a bit less per bulb in a box of 25) rather than $2 or less for the Sylvania Octron in contractor pack). Generally they produce slightly more lumens (about 15% more) and have a longer bulb life (up to 50% longer), so maybe they are about the same cost in the long run–if you don’t pay much more for them.

Studies are also being conducted on the use of narrow-spectrum LEDs. If you want to experiment with them, go ahead, but the studies I’ve seen suggest that we still do not know enough about them to make using them economically justifiable for you and me. Their cost will be dropping and their efficiency increasing, and maybe in a few years we’ll know the best colors to use. Till then–use fluorescents.

Fluorescent tubes are coded by diameter in T units: each T unit is 1/8". The standard old linear fluorescent tube, 1 1/2 inches in diameter, is called T12. T10 tubes (1 1/4 inch diameter) are also available but mostly for specialty fixtures of odd lengths. The T5 tube is the most efficient at present (diameter 5/8 inch), but it’s only slightly more efficient than the T8 tube, and costs a lot more initially (more than you’ll save in electrical costs over several years of use). It runs hotter and is brighter, so you’ll need to keep it farther away from the plants (thereby losing part of the value of greater efficiency) and avoid looking directly at it. At least, that is what I’ve read. The value-for-price sweet spot right now is the T8 tube, either 48" long (for most of us) or 96" long (if you have the space in a special grow area).

Larry mentioned that bulb output decreases after some time of use. That is true. T8 bulbs last longer and have a much better average-over-life light output than T12 bulbs, however, and do not darken as quickly at the ends of the tubes. The 48 inch long Sylvania T8 Octron is rated at 24,000+ hours with an initial light output of 2800 lumens and an output (at 1/3 of its rated life) that is still 90% of the original. If you run the lights 24/7, a set of bulbs would be expected to last longer than 3 years and have an acceptable light output at nearly the end of their life. When you are buying bulbs, always pay attention to the color temperature (rated in degrees Kelvin–K) and the rated life. There are some deceptive names, such as Cool White Deluxe and Daylight, which are not as good for our purposes. There are some lower-priced lines (such as UtiliTech and Econo-whatever) that may cost slightly less per bulb but do not last as well and do not give as much light–not a bargain. T8 bulbs can be bought in contractor packs of 25-30, usually at significant savings, in the big box stores such as Lowe’s and Home Depot, but remember to pay attention to the brand, the rated life, the name of the color, and the color temperature.

If you have the relatively inefficient, old T12 lights and would like to retrofit them to use the more efficient T8 bulbs (the T8 bulb uses the same socket as the T12), you probably can do that for a pretty low price if your old fixture has an electromagnetic ballast. Electronic ballasts which fit the same mounting space as the old electromagnetic ballast are frequently available through eBay and other online sources for less than $10 each in groups of 6-10, including shipping. Some Lights of America (and other) models already have electronic ballasts designed for T12 bulbs, and not much can be done for them. But they were super-cheap when you got them. Although some ballasts made recently will work for T8 bulbs and T12, not all will. Learn what you have so you can decide what you need. Don’t put T8 bulbs in a fixture that needs T12 bulbs, and vice versa.

If you are buying fixtures, look for some with good reflectors. You can buy a strip fixture to mount on a ceiling, but it doesn’t have a good reflector, so a lot of the light will go upward and won’t reach the plants. I recommend the 12 inch wide reflectors, usually called batwing reflectors, but some 8" reflectors, if they’re shiny enough (either glossy white enamel or “specular steel”, both of which are 85% efficient, or better, at reflecting the light to the plants) and have the right shape, also are good. Fixtures with batwing reflectors cost more. If you are handy with tools and materials you can make your own reflectors from some shiny material.

Recently David Zlesak and I were discussing lighting for plants (he was upgrading his lights), and I came across a thread with some interesting and informative discussion on the GardenWeb “growing under lights” forum. The thread is entitled “Ten myths of growing under lights.” The initial post was made in 2005, but most of the information is not out of date. It’s worth your attention if you grow seedlings or cuttings under lights.


Thank you Peter for all your help with updating my lights. I was spurred on when I could no longer buy the T12 bulbs I used to at Home Depot (40W 3200 lumens; cool white) and knew I needed to look into alternatives. I didn’t pay enough attention to the light fixtures in the past and realize now I could have been doing much better for light output and efficiency. I was able to buy a number of T8 fixtures recently on sale with very shiny chrome reflectors that were angled in a great way that efficiently reflects back downward most of the light from the tubes that would otherwise have shown upward away from the plant. Wow, it sure made a great difference!!! Before I would buy the cheaper T12 fixtures that were narrow so I could cram a couple next to each other above the shelves for 4 tubes wide on some shelves. I think I’m getting almost the same light intensity down to the plants with fixture with the light able to be directed back down. It is definitely noticeable how the T12 bulbs dim down fast. I’m excited for these T8’s too. Also it is nice that they are narrower and I have less boxes around in the basement where I have to deal with bulbs. I have about 30 or so shop light fixtures and finding room to store all those old T12 bulbs until I can recycle them is annoying too. The narrower T8’s that last a lot longer without dimming so bad sure seem like a good option.

Thank you again Peter for all your help and encouragement to upgrade and assembling this great post.

One might expect red light to be a potentially more efficient source of energy for plants. (Plants are green, largely because chlorophyll absorbs red light.) It’s interesting to compare spectra of red LED’s and flourescent lights and compare to absorption spectra of chlorophylls a and b. The red LED correlates with red portion of chlorophyll b’s absorption.

Flourescent lighting:

Red LED’s:

Chlorophyll a and b:

(Links to Wikimedia.org)

Dunno if studies on the topic have been done, and it’s been forever since I’ve done biology/botany, but I’ve wondered if chlorophylls a and b correlate to different functions within the plant (such as day-length determination, triggering growth in different parts of plant, etc.)

There have been a good many studies of these matters. Here are three of many I found last night:

***re use of narrow-band LEDs (2008)


***re lights and orchid growing (1981), used a combination of wide spectrum and cool white fluorescents


***re growing tomato transplants (1991)

Probably we should remember that “fluorescent lights” is an awfully broad term. The spectra of fluorescent lights, as of “Red” LEDs, vary considerably.

Just from personal experience and no record keeping studying done, I started out using the grow light tubes but they burned out too quickly and were pricey. I started using just the cheaper regular fluorescent tubes instead and haven’t noticed any difference in germination or growth rates at all.

Here’s a link to a poor copy of the paper I had in mind. The authors based a lot of their conclusions on a couple slightly earlier papers of theirs, plus work by several others. Basically, what Peter said. It is from 1983, while the earlier work is from late '70s. So things have changed for the better since then. I looked at HD the other day trying to buy some T12s and it’s really hard to figure the difference between some of them, but what I saw listed as cool white deluxe seemed to have both longer predicted life and a bit higher lumens. Round $2/tube in a 10-pack.

If you are starting from scratch, compact fluorescents might work out cheaper and more flexible for arranging. They come in so many sizes of lamp and package, it’s hard to be sure. If you figure $10 for a shop fixture, then add tubes and give it a lifetime of 10 years, with a few changes of tubes it is slightly cheaper to go that way. New LED flood lamps are being heavily promoted and they come close. In all cases you have to figure in the wattage per lumen. The LEDs win that one hand’s down but the initial capital input is pretty big. Depends on your electricity rates how much the claimed savings turns into. So far, not much data on the whitish LEDs in practice for plants.


We have someone here trying the red + blue LED route for several kinds of plants. For things with daylength dependent flowering, you have to add some far=red at dawn and dusk it seems. Tomatoes grow OK. Roses should. Problem is, blue light is needed for some photomorphogenic signals, red/far-red for others. And green light is perceived in shade avoidance. So rather complicated and species dependent.

Larry or others, I might have missed it in here, is there an ideal Lumen we are looking for on the tubes/box.

Thanks Peter for this great post.

I hope the post will be useful to at least a few of us. I’ve learned a lot about it myself from revisiting the research since I wrote an article on the subject for a 2008 issue of the RHA Newsletter. And in the day since the original post, I’ve found a lot more stuff. I’d add it, but it just supports what I’ve already said, with the exception of the specifics mentioned below about lumen depreciation.

The T12 fluorescent is being phased out because of its relative inefficiency. The 40W T12 bulb will soon be unavailable (it’s being replaced by a 34W tube which has correspondingly less light output), and electromagnetic ballasts for the T12 have not been legally manufactured in the USA since more than 2 years ago.

Even if you’re able to get T12 bulbs now, you’re going to get shorter bulb life and poor lumen maintenance in the bulbs. The T12 bulb has lumen depreciation of 20% at 7000 hours, and continues to lose. The 40W T12 that gives 3000 lumens at the beginning may give little more than 2000 lumens by the time it reaches the last 20% of its life. The 32W T8 loses only 10% after 7000 hrs, with little further loss for the life of the bulb, and produces more light than the T12 by the time the two reach that 7000-hour mark. And it does it with 20% less electricity.

So it makes sense to convert your fixtures to T8 if that is possible (both T8 and T12 use the same socket). If you are getting new fixtures, don’t buy deeply-discounted T12 fixtures–you may save a little on the fixture, but you’ll lose money in the first year just on greater use of electricity and more expensive bulbs that don’t last as long and don’t give as much light.

David, there is no ideal “lumens” level–the term lumens refers to the amount of light put out by the bulb. Grow lights such as Gro-Lux have a relatively low output in lumens, but they are high in the wave lengths used best by some plants, so they put out less radiation that’s visible as light to us but have very high outputs of the sorts of radiation used by plants. The big problem with grow lights is that they are quite expensive–and you can get results that are almost as good with regular cool white fluorescents. And you can read better by the light of the regular fluorescents if the mood strikes you.


Thanks Peter, for all your information. That gives me an idea, I could read to the seed/seedlings some great stories about their fore bearers.