[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: watts per gal

Gordon wrote:

questions about the whole "watts per gal" idea in tank lighting.

I'm not an electrical engineer, a mathematician, or a botanist, so what
thinking is most likely wrong, or something like that :)

Numerous websites have formulas to calculate how many "watts per gallon"
light you need to grow low-mid-high light plants...

The questions I have are:

Isn't watts a rather unreliable measure of light output?  I mean, a 75
standard incandescent puts off as much light as a standard 20 watt cool
fluorescent, or a 13 watt compact fluorescent twist-bulb.  Which wattage
the formula's speak of... I assume they mean watts fluorescent (since
to be the most popular)?

Are there any formula that use something more related to actual light
output, like lumens or candlepower?  Or formula that are calibrated to
wattage to light output ratio for standard fluorescent bulbs (can't
the power compacts yet).


The watts per gallon rule is fairly accurate in a lot of cases with
fluorescent and MH lamps. It can however be grossly misleading as well.
Lumens per gallon or candle power would be even worse. There are at
least 5 things that screw up the watts per gallon rule. Ballast
efficiency, lamp efficiency, fixture efficiency, PAR efficiency and how
well your lamp spectrum matches the photopic response of your plants. To
apply all these factors is quite difficult if not nearly impossible so
the watts per gallon rule is what people use. Suprisingly often the
different efficiencies in each setup cancel each other out and the watts
per gallon rule works great.

There are other ways of predicting the light output of a fixture that
work pretty good. If you ignore the lamp spectrum you can pretty much
get a good prediction just by calculating the system lumens per gallon
of your light fixture. Then you pick whatever spectrum you think will be
best. This works great for fluorescent lamp systems. You must always use
the same lamp spectrum to make this calculation. I pick the 841 spectrum
as it is pretty much the same for every type of lamp. You then find out
the lumen output for your particular lamp style multiply this by the the
ballast factor and a guesstimated fixture efficiency. If you have 200
lumens per gallon you have a high light tank. If you wish to compare MH
lamps to fluorescent lamps or as in other cases where you cannot find an
equivilent spectrum you have to convert the different lamp spectrums by
using a PAR/lumen conversion factor. Those conversion factors are
available on Ivo Busko's lamp comparison page.

To make it easy to calculate I consider an AH Supply type fixture to be
of maximum possible efficiency. All other fixtures need to be derated
according to a guesstimate of how good they are. The worst case scenario
would be T12 lamps with close lamp spacing and a semi-reflective
background. Many home made fixture suffer from this sort of problem. I
think you have to consider this fixture efficiency to be at least 1/2
that of the best fixtures available. If lamp spacing is reasonable say
with 2 lamp diameters between the lamps and you have an excellent difuse
reflector then the fixture efficiency might be 85% of the best fixtures.

An example calculation might be to compare a 2 55 watt PCs to 3 32 watt
T8s and 6 20 watt T12s on a 55 gallon tank.

If you have a Phillips ballast and PC combination the lamp lumens is
4800, the Phillips ballast factor is .88 and if you have an AH Supply
reflector you would have the best fixture efficiency. 4800 lumens/lamp*
2 lamps * .94 * 1.0/ 55 gallons = 164 lumens per gallon. The second
setup using Phillips lamps and a 1.2 ballast factor is also pretty good.
3100 lumens per lamp * 3 lamps * 1.2 * .85 fixture efficiency = 172
lumens per gallon. The 20 watt lamps are only about 1000 lumens per lamp
and the ballast factors for the ballasts that run them are often quite
low. I picked a .77 B.F You cannot quite get 2 lamp diameters between
the lamps so you fixture will function not quite as good as the T8
fixture. I am calling the efficiency .80 So, 6 lamps * 1000 lumens/lamp
* .77 * .80 equals  67 lumens per gallon.

Now you can see if everything is stacked against you the watts per
gallon rule means nothing. Nobody would try to light a 55 gallon tank
with 6 20 watt lamps or would they? I see lots of people trying to use
compact fluorescents to light some pretty large tanks. It is a hopeless

Naturally since I basically made up the fixture efficiencies based on
luminaire data I found on the net, this method is not as accurate as it
appears but it is far better than the watts per gallon rule. Someone
really needs to do a study on lamp spacing and how it relates to fixture
efficiency in an aquarium. Likewise for specular reflectors verses
difuse reflectors.