Comments on the Watts/Gallon Survey
- The "Watts-per-gallon" Survey finished
by Neil Frank <nfrank/mindspring.com> (Sun, 17 May 1998)
- Erik's lighting results
by rjw/aluxs.micro.lucent.com (Ronald Wozniak) (Wed, 20 May 98)
- more thoughts on light vs volume
by "Roger S. Miller" <rgrmill/rt66.com> (Sat, 30 May 1998)
- blame my editor!
by "Roger S. Miller" <rgrmill/rt66.com> (Sat, 30 May 1998)
- Next project
by George Booth <booth/frii.com> (Fri, 25 Sep 1998)
- Aquatic Plants Digest V3 #537
by Justin Collins <weaslvil/rocketmail.com> (Fri, 25 Sep 1998)
by Neil Frank <nfrank/mindspring.com>
Date: Sun, 17 May 1998
>From: Erik Olson <(e-mail)>
>Subject: The "Watts-per-gallon" Survey finished
I really like your lighting analysis. Very thought provoking. You have
clearly shown that Amano's data conforms to a watts per area formula.
I used to think that one of my large tank data points was a clear outlier:
80w per 280L. Now I see it fits nicely into Amano's data set. Same for the
other two higher light tanks: 160w per 300L and 280w per 500 L. This may
add more "ammo" to the argument that some people are using more light than
However, another variable which probably needs to be considered is general
hardness and other water chemistry parameters. My water is very soft
(20-40ppm Ca). Some people say that they can't grow plants at my (or
Amano's) watttages. It would be interesting to stratify your data into the
lowest and highest quartiles of wattage per area and then find out if there
are systematic differences in the type of water that people have.
Another small point, not related to your general conclusions: In the
article you say "incandescent light, however, is about 1/4 as efficient as
flourescent." Although incandescent may be 1/4 as efficient in the spectrum
used by the human eye, is it the same for plants (i.e. considering PAR)....
Incandescent is heavily weighted towards the red and far red which we
can't see, but plants can use and enjoy.
Thanks for making an important contribution to aquatic gardening.
Neil Frank, AGA
by rjw/aluxs.micro.lucent.com (Ronald Wozniak)
Date: Wed, 20 May 98
>Subject: Erik's lighting results
>I agree with Frank that Erik's survey results are thought-provoking and an
>interesting contribution to the hobby. I have some observations and
Ditto from me. Good job Eric. A real thought provoker.
> First, the difference between the data from Amano's books and the
>Usenet data is striking. I've only read one of Amano's three volumes and
>my recollection is that the smaller tanks feature a lot of stem plants
>with very dense plantings that shade each other, and that the larger tanks
>feature more moderate and low-light plants, with relatively little
>shading. If my recollection is true and his other books are consistent,
>could the trend in Amano's data be a product of his unique style?
I'd agree with Rogers observation. Just looking at the Usenet data, It seems
that the 2 watts/gal to 4 watts/gal better fit the data than the "estimated
surface area" (Volume to the 2/3 power). As Roger pointed out, Amano data
may be skewed to his style of planting.
> Second, when I visualize a trend through the four data points from
>commercial fixtures it parallels the trend in Amano's data, but is offset
>downward by a factor of about 4. Commercial fixtures I've seen are
>usually single-tube lights, equipped with whatever tube is close to but
>shorter than the tank the fixture will fit on. So the trend (if there
>really is one) is determined mostly by the *length* of the tank and the
>wattage of standard fluorescent tubes, not by area or by volume. You
>might create the trend in Amano's data just by using the largest 4
>commercial fixtures (or a fixture with 4 fluorescent tubes) that fit over
Again, another very good point. With fluorescent lighting, you can only put
in bulbs that can fit over the top (surface area) of the tank. Commercial fixtures
would be skewed to the cheapest solution that would light just across the length
of the tank. (i.e. One tube that fits most of the length of the tank.)
Perhaps there's a better equation to fit the data. Eric, is it possible to
plot Watts versus actual surface of the corresponding tank where the data
came from? It seems to me that estimate surface area by taking the volume to
the 2/3 power applies if all the tanks have about the same ratios of length:width:height.
Where's this all heading? Well, here's my 2 cents worth of contribution:
DEPENDANCY ON PLANT GROWTH RATE ON WATTS, VOLUME, AND SURFACE AREA version 0.1
0. Assume that the lighting intensity is always less than the plants light saturation
levels. (Light saturation levels for most aquatic plants is around 10% of full sunlight.)
(A lot of lighting [i.e. mucho watts] is needed to reach this lighting level.)
Also, assume that plants need about 12 hours of little or no light for their rest period.
1. A plant's mass (size & weight) is proportional to it's leaf surface area.
2. A plant's growth rate is proportional to the surface area of it's leaves.
3. A plant's growth rate is proportional on the amount (intensity) light shining on it's
Therefore, the taller a plant gets, the more total amount of light the plants needs.
As the plants leaves grow wider, and the plant spread out across the gravel, the more
total light the plant(s) will need.
Also, a plant will grow in height and spreads out until the growth becomes limited by the
available light. Which is another way of saying that a plant stops growing if it doesn't
get enough light.
4. The volume of space that a plant occupies is proportional to the mass (size) of that plant.
5. The total amount of light is proportional to the intensity (wattage) of the light bulbs and
how long the lights are on.
Therefore, the final maximum amount of all plant mass (size/weight) is proportional to the
tanks volume and lighting watts. (With light duration being fixed.)
And, also, the growth rate is proportional to watts and how much space is available.
The experience of the Usenet community is that 2-4 Watts/gallons give adequate growth for the
types of plants of interest. Some types of plants need more light than others.
6. For fluorescent lighting, the number of light bulbs that you can use is proportional the
tank's top surface area. (The situation is quite a bit different for MH lighting. In MH
lighting, the light is concentrated over a much smaller area compare to fluorescent. It's also
fairly easy to reach the light saturation levels with MH lighting, which complicates any
analysis such as this.)
Therefore, for fluorescent lighting, growth rate is also proportional the tanks top surface area.
However, this rule is just a limitation of fluorescent lamp technology. Plants really just care
about light intensity & duration. Of course light quality (PAR etc...) also has a roll.
IMHO, either watts per gallon or watts per surface area will give you an estimate on how
much light you need. For fluorescent lighting, certain size tanks lend itself to putting more
lamps on top of the tanks. How many tubes you need depends on the type of plants you want to grow
and how often you want to trim them.
Comments would be appreciated. To the logical purist, I was a little foot loose and fancy free on
the above derivation. I could have been even more formal which would have made it even more
unreadable. Of course, any statement is only as good as the assumptions that the statement is
based. So, see if you agree with my assumptions.
While we are on the subject of proposing fluorescent lighting rules of thumb. Here are mine:
For fluorescent lighting that extends across the length of the tank (that are 16 to 24 inches
deep), 2-3 watts per gallon is recommended. (Depending on the type of plants you want to grow
and how often you want to trim them.)
1 tube every 3 inches of tank width gives about 3 watts/gallon.
1 tube every 4 inches of tank width gives about 2 watts/gallon.
Standard pet shop lighting, (which is about 1 tube for every 12 inches tank width,)
gives about 3/4 watt/gallon.
For 20 Gallon High tank (12 and 1/2 inches wide and length of 24 inches):
4 tubes (15 watts x 4) gives you about 3 watts/gallon.
3 tubes (15 watts x 4) gives you about 2 watts/gallon.
For 55 Gallon tank (12 and 1/2 inches wide and length of 48 inches):
4 tubes (40 watts x 4) gives you about 3 watts/gallon.
3 tubes (40 watts x 3) gives you about 2 watts/gallon.
For 90 Gallon tank (18 inches wide and length of 48 inches):
6 tubes will give you about 3 watts/gallon.
4 tubes will give you about 2 watts/gallon.
Exceeding 3 watts/gallon using normal T12 fluorescent lighting becomes difficult. Heat build up and
reflector efficiency/tube spacing become problems. T8 bulbs and electronic ballast can stretch this
rule a little, but Metal Halide (MH), VHO, "HO T8" or compact fluorescent lighting become more
practical at lighting levels above 3 watts per gallon. It is possible to successfully grow certain
type of plants at 3/4 watts/gallon, but your going to have problems finding those type of plants at
your local fish shop. Tanks deeper than 24 inches have problems of being difficult to maintain
(Unless you have 3 foot long arms) and fluorescent lighting doesn't have enough punch (light intensity)
to get through the taller plant leaves that will tend to shade the shorter plants.
Ron Wozniak Allentown PA, USA
by "Roger S. Miller" <rgrmill/rt66.com>
Date: Sat, 30 May 1998
Its been a little while now since our discussion spurred by Erik's study
of light use and tank size. I've cogitated a bit more, and have found two
reasons why smaller tanks might need higher intensity light (on a volume
basis) than larger tanks.
1. A small tank needs brighter light than a large tank so that
it will be more eye-catching. This is a purely aesthetic consideration.
Small tanks tend to get overlooked unless the light in them grabs the
viewer's eye. A large tank is inherently more visible than a small tank,
so doesn't need the brighter light.
2. Small tanks seem to a larger part of their light out their glass
sides. So in bigger tanks a larger proportion of the available light gets
used by plants and you need to provide relatively less light to get the
same effect on growth. The effect would be particularly pronounced with
tall tanks. When you get to really small tanks (less than 10 gallons) the
tanks have to be tall compared to their surface area or there's no room
for plants, and that would exaggerate the effect.
In Albuquerque, where sultry summer's searing sands seem soon.
by "Roger S. Miller" <rgrmill/rt66.com>
Date: Sat, 30 May 1998
This should be:
> 2. Small tanks seem to lose a larger part of their light out their
> sides. So in bigger tanks a larger proportion of the available light gets
> used by plants and you need to provide relatively less light to get the
> same effect on growth. The effect would be particularly pronounced with
> tall tanks. When you get to really small tanks (less than 10 gallons) the
> tanks have to be tall compared to their surface area or there's no room
> for plants, and that would exaggerate the effect.
by George Booth <booth/frii.com>
Date: Fri, 25 Sep 1998
>Date: Fri, 25 Sep 1998 08:31:47 -0500
>From: "Beard, Kelly" <KBeard-at-comdata.com>
>What kind of lighting would you recommend for a tank this size? I guess the
>standard dimensions for a 74 gallon are 48 x 18 x 21. Does the gallon vs
>watt rule apply here or are there different considerations?
IMHO, the watts/gallon rule rarely applies.
>watts/gallons rule applies, I guess I would have to shoot for 225-300 watts
Nope, doesn't apply here very well either.
>Some of you have tanks at least this large. I'd like some descriptions of
>your lighting, high-tech or not, if it is working well for you. Include as
>much trivial info that you can (bulb lengths, color, VHO or MH, suspended
>hood, shop lights, pendant, etc, etc, etc).
We have 4 40w (48") bulbs using old fashioned tar ballasts over our 100
gallon (60"x18"x20"tall) tanks. However, we prefer to maximize the lumens
per watt (and $ per watt :-), so we use two Triton bulbs and two Penn-Plax
Ultra Tri-Lux bulbs over each tank. Nice and bright, good color rendition
and they photograph well (all the photos in our web site were taken with
normal tank lighting and plain old daylight film).
That's 1.6 watts per gallon, but they're good watts.
George Booth, Ft. Collins, Colorado
by Justin Collins <weaslvil/rocketmail.com>
Date: Fri, 25 Sep 1998
Check out www.thekrib.com/Plants/Tech/Lighting to see an interesting
discussion on lighting levels. In my limited experience, it seems
pretty accurate. I believe it is possible to have too much light. I
currently have a 60 gallon (48x16x18, I think) with 260 watts of
flourescents (2 VHO full spectrum, 1 NO, all 48 inch, run by an Icecap
660, mounted in a DIY hood) and it is too much, even with added CO2
(15-20 ppm) and trace elements. My plants, especially wisteria seem
to like to grow out, rather than up, leaving me with sections of tank
bare at the top. Some of my crypts are doing wonderfully, but others
I am getting limited growth out of since I added the lights. Also, I
am having a hard time finding plants that will stay green under my
lights. Even my Giant Hygro and Madagascar Lace plants are turning
red at the tips. My Rotala Rotundifolia is almost entirely red. Part
of this may be that the NO tube I am using is a Coralife 50/50 bulb,
and the added blue and UV radiation may be causing some of the
problems I am having. More than anything, I think you have to decide
your light levels based on the kinds of plants you want to grow.
As far as types of lighting, I have been very happy with my
fluorescents, both NO and VHO. It doesn't cost much to run, is very
flexible, as far as size, wattage, and spectrum, and produces an
acceptable amount of heat. My Icecap ballast is wonderful. It runs
cool, doesn't make the bulbs blink 45 times before starting, and will
power basically any fluorescent I would ever want to use. If I were
going to do things over, I would seriously look into power compacts
and T-8s. Sylvania has come out with some 48" T-8 plant bulbs that
would be ideal for your setup. I am currently looking into MH or
MH/PC mix for a 100 gal I am assembling equipment for but don't know
very much about them.
I also have a couple of questions. On the 100 gallon I am setting up
(60x18x20), if I use 2 175 watt MH pendants, will I end up with dark
corners? I will also be using them over an acrylic tank, so will I
have problems with the top sheet reflecting back too much light? What
are people's experiences with this type of light? Thanks in advance.
> Date: Fri, 25 Sep 1998 08:31:47 -0500
> From: "Beard, Kelly" <KBeard-at-comdata.com>
> Subject: Next project
> What kind of lighting would you recommend for a tank this size? I
> standard dimensions for a 74 gallon are 48 x 18 x 21. Does the
> watt rule apply here or are there different considerations? If the
> watts/gallons rule applies, I guess I would have to shoot for
> of lighting.
> Some of you have tanks at least this large. I'd like some
> your lighting, high-tech or not, if it is working well for you.
> much trivial info that you can (bulb lengths, color, VHO or MH,
> hood, shop lights, pendant, etc, etc, etc).
> "Take me out tonight where there's music and there's people who are
> and alive".
> - -- Morrissey
> Kelly Beard, Cat IV, Team Allanti
> President, Allanti Cycling Club - http://www.allanti.com
> Race in Tennessee - http://www.tbra.org/98-road.shtml
DO YOU YAHOO!?
Get your free -at-yahoo.com address at http://mail.yahoo.com