Actually it is more convenient from a biological perspective to think of the particle-like nature of photons, rather than being hung up on wave effects. They are packets of energy that can participate in chemical reactions. Wave effects are important in biooptical issues (imaging, refractive index modulation to enhance transparency, structural color, etc.) For photosynthesis and excitation of photochemical reactions in the eye, it is easier to think of them as quanta of energy.

What you wrote about thenumber of photons being related to percieved intensity is correct. Of course there are complications. The first complication is that the molecules absorbing light are different in differen systems, which gives rise to different action spectra. So for monochromatic light, what you wrote is correct, but it falls apart when considering light of different energies. The eye responds to light on more or less a log scale, not a linear scale. Photosynthetic systems display saturation kinetics and photoinhibition. So more light doesn't always give you more photosynthesis, and past a certain point, very intense light can actually start to destroy the photosynthetic apparatus. In most aquaria, though, high light organisms are being maintained at lower than saturating light fields. Low light organisms, like some species of coralline algae can easily be saturated and in some cases photoinhibited. Dana and Andy wrote a series of articles on this subject. I think they appeared in FAMA.

Although one would think from aquarium ads that PC lights emit more light per unit of input energy than linear tubes, they actually don't. If you make comparisons between similar ballasts (say, all lamps on solid state balalsts) and with the same phosphor system in all the lamps (say a rare earth phosphor on all the lamps) then wide linear lamps make more light per unit of input energy than PC lamps. There are two main reasons for that.... the plasma in the lamps loses energy at the sides of the lamps, so wide tubes are better than skinny tubes. The only advantages of skinny tubes is that you can drive the power density up, and you can use expensive phosphors in a more cost effective way in skinny lamps. The plasma also loses energy when it goes around corners.

The comparisons you read between PC lamps and linear lamps in aquarium magazines are all cooked. They make apple and oranges comparisons across ballast types and across different phosphor systems. Yes, PC lamps will give more light per unit of input energy than cool white lamps running on the crappiest electromagnetic ballast you can find. They most assuredly don't give more light per unit of input enrgy when comparably ballasted using comparable phosphor systems.

If people have more questions about lighting technology issues, I refer them to the excellent IES lighting handbook. It is a much more "enlightening" source of information than aquarium ads.

The phosphors on my monitor recently glowed in this configuration.

Ah, I think I know this one. I suspect that, like with organisms, a tube generates energy over its volume and radiates over its surface. (the excited gas inside generates UV photons which hit the phosphor on the inside surface, which absorbs a high energy photon and emits a (many?) lower-energy photons at select visible (and invisible) frequencies.

That is more or less correct. The plasma in the lamp emits UV radiation, some visible radiation and heat. Some of the visible light makes it out directly, some of it is trapped in the phosphor. Most of the light production in fluorescent lamps comes from 254 nm UV (Hg emission line) striking the phosphor and generating light by fluorescence.

A short fat tube is more like a sphere than a long skinny tube, so that's the context in which I meant the CF has more surface area.

Within the context of your argument, I don't see why being spherical vs. long and skinny would make much difference. So what if the tube is fat? What matters is how much UV strikes the phosphor, integrated over the area of the phosphor. The only way long path lengths are going to negatively affect the amount of UV ultimately striking the phosphor is if there is appreciable trapping of UV radiation in the plasma inside the lamp. And I've not heard that advanced as a significant factor. I think the plasma is optically thin at 254 nm.

To your point, you may be right - I still have to prove to myself that more surface is a good thing here and is it directly proportional to the electrical energy you can run through it.

From the perspective of paying for phosphors to coat the inside of the lamp, more surface area is bad. Because it costs more to coat the surface to the same thickness in a tube with a lot of surface area. However, in terms of losing energy from the plasma to the walls of the lamp, larger diameter tubes are better, because a smaller fraction of the excited atoms in the plasma strike the walls in a lamp with a larger cross section.

The tendency of the lighting industry to go to skinnier tubes is driven primarily by two factors.

1. phosphor costs and
2. less trapping of radiation in the FIXTURE when skinnier tubes are used

Also there's gotta be an optimum point of volume to surface area, and whatdya wanna bet the guys that engineer these things know what it is and the geometry of a tube just happens to be pretty close to that? Or, they're locked into an old legacy spec? Hmmm.

That is a complicated surface. The optimum point is going to vary depending on whether or not you are the person paying for the phosphor or paying for the electricity. It also varies somewhat based on fixture design. And yes, there are a lot of legacy factors involved.

Another point that I need to make is this you have not shown that volume to surface area is the critical metric here. One factor that you have not addressed at all is the frequency that the tube is driven at (essentially all of them do better at high drive frequencies, but the gains are low once you get past 20kHz or so) and the Length of the lamp. Long linear lamps are always better from an energy efficiency perspective because the plasma loses less energy at the ends in collisions with the electrodes and it loses less in collisions with the walls due to bends. But once you go past about 4' in length, the increases in efficiency with longer lengths diminish.

A really critical issue that you haven't mentioned is the amount of trapping of radiation that happens in the fixture itself. Skinny lamps look good there. Skinny lamps also look better in terms of power density per unit are a of the fixture, which is quite limited in aquaria. The last reason is really why PC lamps are interesting from an aquarium perspective. Oddly enough, I don't think I've ever heard that point made explicitly in the ads promoting these products.