What size UV sterilizer do I need?
It depends on what you want dead and how fast the water goes past the lamp — and only on those. A UV unit delivers a dose: the lamp's UV-C output spread over the water flowing through it, in millijoules per square centimeter (the makers' µW·s/cm² ÷ 1,000). Green water and bacteria fall at about 30; the makers rate "protozoa" — the free-swimming stage of ich and its relatives — at 180. A 75 gallon tank (284 L) run through the unit once an hour against parasites needs 20 W of lamp by the end of the lamp's year, so a 25 W unit at 75 GPH; for green water at three turnovers an hour it needs 10 W — a 13 W unit at 220 GPH. The chart runs every standard tank both ways; the tool sizes yours, or grades the unit you have at the flow you give it.
| Tank | Green water (30 mJ, 3×/h) | at | Parasites (180 mJ, 1×/h) | at |
|---|---|---|---|---|
| 5.5 gallon | 9 W | 15 GPH | 9 W | 5 GPH |
| 10 gallon | 9 W | 30 GPH | 9 W | 10 GPH |
| 15 gallon | 9 W | 40 GPH | 9 W | 15 GPH |
| 20 gallon long | 9 W | 50 GPH | 9 W | 15 GPH |
| 20 gallon high | 9 W | 55 GPH | 9 W | 20 GPH |
| 29 gallon | 9 W | 80 GPH | 9 W | 25 GPH |
| 37 gallon | 9 W | 100 GPH | 9 W | 35 GPH |
| 40 gallon breeder | 9 W | 125 GPH | 13 W | 40 GPH |
| 55 gallon | 9 W | 160 GPH | 18 W | 55 GPH |
| 65 gallon | 9 W | 195 GPH | 18 W | 65 GPH |
| 75 gallon | 13 W | 220 GPH | 25 W | 75 GPH |
| 90 gallon | 13 W | 260 GPH | 25 W | 85 GPH |
| 120 gallon | 18 W | 340 GPH | 36 W | 115 GPH |
| 125 gallon | 18 W | 340 GPH | 36 W | 110 GPH |
| 150 gallon | 25 W | 450 GPH | 57 W | 150 GPH |
| 180 gallon | 25 W | 540 GPH | 57 W | 180 GPH |
| 210 gallon | 36 W | 630 GPH | 57 W | 210 GPH |
Lamp electrical watts, next size sold, sized for 70 % output (a year-old lamp) in clear water; salt water needs about 10 % more, stained water 65 % more. Flow is through the unit — its own pump or a valved branch — not the return. Dose model calibrated to Aqua UV's published ratings; a budget unit without a reflector delivers less.
The tank's volume is not in the dose equation at all; the flow through the unit is. Aqua UV's own ratings for a 25 W unit say so: 788 GPH for 30,000 µW·s/cm², 131 GPH for 180,000 — six times the flow, one sixth the dose, the same lamp. Run that 25 W on a 300 GPH return branch and it delivers 79 mJ/cm²: fine for green water, a third of what parasites need. Feed it from a small pump at 100 GPH and it delivers 236. A lamp's watts per gallon of tank tells you nothing until you know the flow, and that is exactly the number the "10 watts per 75 gallons" rules leave out. The tool's estimate — 3,580 mJ/cm² per lamp-watt per L/h, 35 % of the electrical watts as UV-C — is calibrated to that pair; a published rated-flow-at-dose for your unit replaces it.
How fast should water flow through a UV sterilizer?
As fast as it can while still delivering the dose for the job — because flow is passes, and passes are the whole point. A tank mixed by its pumps has 92 % of its water through the unit after 2.5 hours at one turnover an hour, 99.8 % after six, 99 % in 4.6 hours; twice the flow halves those times. So a bigger lamp is not "more kill" at a given flow, it is more flow at the same kill: the table shows the most flow each lamp size can take for each dose, and it is why a 40 W unit on a 75 gallon parasite problem is money well spent. Feed a parasite-dose unit from its own small pump or a valved branch off the return; a whole return through a UV is a clarifier, whatever the box says.
| Lamp | Green water (30 mJ), new | a year old | Parasites (180 mJ), new | a year old |
|---|---|---|---|---|
| 9 W | 280 GPH | 200 GPH | 45 GPH | 35 GPH |
| 13 W | 410 GPH | 290 GPH | 70 GPH | 50 GPH |
| 18 W | 570 GPH | 400 GPH | 95 GPH | 65 GPH |
| 25 W | 790 GPH | 550 GPH | 130 GPH | 90 GPH |
| 36 W | 1,150 GPH | 790 GPH | 190 GPH | 130 GPH |
| 40 W | 1,250 GPH | 880 GPH | 210 GPH | 145 GPH |
| 57 W | 1,800 GPH | 1,250 GPH | 300 GPH | 210 GPH |
| 80 W | 2,500 GPH | 1,750 GPH | 420 GPH | 290 GPH |
Maximum flow through the unit for that dose, clear water. "A year old" is 70 % output — the reason to replace lamps on a calendar, not when they fail. Compare with the maker's own flow rating and trust theirs where it is lower.
What dose kills what? The literature disagrees
Honest ranges, because the sources are ranges. Algae and bacteria are cheap; the parasites people actually buy UV for are not, and the published figures for marine ich span almost 3×, from an industry number to a laboratory one. The tool sizes to the makers' 180 mJ/cm² protozoa rating as the working figure and tells you where that sits.
| Target | Dose, mJ/cm² | µW·s/cm² | Source |
|---|---|---|---|
| Bacteria, general (E. coli class) | 7–25 | 7,000–25,000 | water-treatment tables; industry 25 |
| Green-water algae | 22–30 | 22,000–30,000 | industry clarification rating |
| Ichthyophthirius (freshwater ich) theronts | 100–336 | 100,000–336,000 | Hoffman 1974 (100,000); Gratzek 1983 compilation (336,000) |
| Cryptocaryon irritans (marine ich) theronts | 280–800 | 280,000–800,000 | UF/IFAS FA164: 280,000 industry to 800,000 (Colorni & Burgess 1997) |
| Makers' "protozoa / parasite" rating | 90–180 | 90,000–180,000 | older 90,000; current Aqua UV / BRS 180,000 |
Doses for a high kill fraction of the free-swimming stage in one pass, as published; a tank gets many passes, which is why units below these figures still visibly help.
Only what passes through the chamber is dosed. Ich, marine ich and velvet spend most of their life cycle on the fish and in the substrate; the free-swimming theronts that a UV can kill are the stage looking for a host, and the ones that find one before their next lap were never exposed — the University of Florida's Cryptocaryon fact sheet says it in as many words. A UV lowers the pressure between fish; it cures nothing on them. Treating the fish is a quarantine tank job — and the same lap arithmetic is why a UV on a fallow display shortens nothing: it cannot reach the encysted stage at all.
Lamp age, the quartz sleeve and water clarity
Three things quietly cut the dose. A low-pressure lamp still lights at the end of its year but puts out about 70 % of its UV-C — invisible, since UV-C is invisible — so the tool sizes for the old lamp, and the schedule is a calendar, not a bulb burning out. The quartz sleeve between lamp and water grows a film of calcium and biofilm that blocks UV-C long before it looks dirty; wipe it whenever the lamp is changed and between times if the water is hard. And the water itself absorbs UV: salt water a little, tannin-stained or turbid water a lot — the tool derates 10 % and 40 % for those. A year-old lamp behind a dirty sleeve in stained water delivers 42 % of the rating on the box, which is the difference between a parasite dose and a clarifier, and it is the usual reason "the UV isn't working". Put a sponge or floss stage before the unit; particles shadow what is behind them.
How much does a UV sterilizer cost to run?
Its lamp watts, all day, plus the pump feeding it: a 25 W unit is 0.6 kWh a day, 219 kWh a year, and the lamp itself yearly. The electricity cost calculator takes it as a 24-hour device; if it runs only during a bloom or an outbreak, set the hours. Switching it off and on shortens lamp life — run it continuously for the weeks it is needed rather than on a timer.
My UV isn't clearing the green water — why?
Usually flow: too fast for the dose (a return line through a clarifier-sized unit), or, less often, so slow that the tank's water barely passes it — the passes number above. Then the sleeve and the lamp's age, together worth more than half the output. Then plumbing: a bypass valve half open, or a unit mounted so the chamber holds an air pocket at the top where the lamp is. And once in a while the problem was never in the water column: a bacterial bloom from an ammonia spike, or diatoms on the glass, both of which a UV can't touch — the cloudy water and algae ID guides sort out which you have. A UV that clears green water in three to five days is working; one that hasn't in ten is undersized for its flow, not for its tank.