Dose is intensity multiplied by time
A UV chamber has exactly one product: a dose, measured in millijoules per square centimetre. It is the light intensity in the water multiplied by how long the water sits in it.
Dose (mJ/cm²) = mean intensity (mW/cm²) × residence time (s)
Both halves are easy to get wrong in the same direction. Intensity is not the lamp wattage — it is what is left after the lamp has faded, after the sleeve has fouled, after the light has spread out across a widening cylinder, and after the water has absorbed some of it. Residence time is not a property of the unit — it is the chamber volume divided by whatever the house happens to be drawing at that moment, and a house draws nothing for hours and then seven gallons a minute for four minutes.
The second point is the one that catches people. Dose is inversely proportional to flow. Double the flow and the dose halves, immediately and exactly, with no lag and no warning. A unit that produces a comfortable figure at average flow produces half of it when the shower and the washing machine overlap, and there is nothing in a residential installation that tells you it happened.
UVT: the number nobody measures and everybody should
Ultraviolet transmittance at 254 nanometres, over a one centimetre path, is how much of the germicidal light survives a centimetre of your water. It is written as a percentage. Clear, low-mineral water might run in the high nineties; water carrying organics, tannins, iron or manganese can be far lower, and colour that is barely visible in a glass can move it several points.
It behaves exponentially, which is why intuition fails on it. The absorbance per centimetre is the negative base-ten logarithm of the transmittance as a fraction:
A = −log₁₀(UVT ÷ 100)
At 95 percent UVT that is 0.0223 per cm. At 85 percent it is 0.0706 per cm — more than three times as much attenuation for a ten point drop. Across a two-and-a-half centimetre water gap the difference in mean intensity is substantial, and the dose falls with it. This is why the same unit installed on two wells forty miles apart can be doing quite different things, and why UVT is a laboratory measurement rather than something you can judge by eye or with a test strip.
| UVT at 1 cm | Absorbance per cm | Light left after 2.7 cm of water |
|---|---|---|
| 98% | 0.0088 | 94.7% |
| 95% | 0.0223 | 87.1% |
| 92% | 0.0362 | 79.8% |
| 85% | 0.0706 | 64.5% |
| 80% | 0.0969 | 54.7% |
The last column is transmission across the gap alone. The calculator does something slightly more careful: it averages the intensity across the whole annulus, weighting by the water volume at each radius and including the one-over-r spreading of a line source, so water near the chamber wall — which sees the least light and is also the largest share of the volume — is counted properly.
The teaching point: what actually collapses a dose
People replace lamps. Almost nobody tracks flow, and that is the wrong way round. Run the arithmetic on any real installation and the ranking is consistent: flow is the largest single lever, transmittance is the second, and lamp age is a distant third because it is bounded — a lamp fades to perhaps 70 percent of new and then gets replaced, a factor of 1.4. Flow is not bounded at all. Going from a single tap at 2 gpm to a shower plus a washing machine plus a toilet refilling at 8 gpm is a factor of four, and it happens several times a day.
Which means the useful question about a UV installation is not what lamp it has but what the largest flow through it actually is. That figure is a fixture-by-fixture sum, and the whole house filter flow rate calculator puts a number on it. If the peak is well above what the unit was sized for, no amount of lamp maintenance fixes it.
The second thing that follows is why so many units carry a flow restrictor. It is not there for the plumbing. It exists to stop the chamber ever seeing a flow that would drop the dose below the figure the unit was validated at, by physically making that flow impossible. Removing one because the pressure seems low defeats the only mechanism enforcing the dose.
Why this is an estimate and a bioassay is not
The arithmetic here assumes plug flow: every drop enters together, moves down the chamber at the same speed, sees the average intensity and leaves together. Real chambers have a velocity profile, dead zones near the ends, and streamlines that run close to the wall where the intensity is lowest. The water that gets the least dose is the water that determines the outcome, and it is not the average.
That is why manufacturers rate reactors by biodosimetry — running a known challenge organism through the actual chamber at the actual flow and measuring what came out, expressed as the dose that would have produced the same result in a collimated beam. The result is a reduction equivalent dose, and it is always lower than the geometric calculation for the same hardware. Where a unit has a validated flow and dose rating, that rating governs. Use this page to understand which input is hurting you, not to certify a unit.
What this page will not do
Every water quality number on this page — hardness, iron, solids, transmittance, a concentration, a target dose — is something you enter from a laboratory test of your own water. Use a state-certified laboratory; your state or county health department maintains the list and will usually tell you what a well in your area is commonly tested for. This calculator has no idea what is in your water, will not tell you whether a result is acceptable, and does not select equipment. Treatment follows the report, and a device that removes one thing does not remove another.
It reports a calculated dose and nothing else. It does not tell you what dose your installation needs, because that is not a number a web page gets to supply — it depends on what is in your water, what the unit is certified to, and what your health department and a water treatment professional say about the specific well. It does not tell you whether water leaving the chamber is disinfected, safe or potable. And it says nothing at all about bacteriological results, a flooded well or a newly drilled one: those go to a state-certified laboratory and your local health department, and the work that follows goes to a licensed well contractor.
UV is also entirely a downstream question. It removes nothing physically, so anything that shades the light or coats the sleeve — sediment, iron, hardness scale — has to be dealt with before the chamber, which is why UV sits at the end of a train. The sediment cartridge life calculator and the carbon bed calculator cover the two stages that usually sit ahead of it.
Questions people ask
What UV dose do I need for well water?
That is not a figure this page will give you, and you should be suspicious of any page that does. What dose an installation is built around depends on what a certified laboratory found in your water, what the unit is certified to deliver, and what your state or county health department and a licensed water treatment professional say about that specific well. The calculator takes a dose figure from you — from your equipment documentation or from the professional advising you — and tells you the flow at which your chamber reaches it. That is a geometry and flow question, which is answerable. The other one is not answerable from a browser.
Why does my UV unit deliver less dose when more taps are open?
Because dose is intensity multiplied by residence time, and residence time is the chamber volume divided by the flow. The chamber holds a fixed number of gallons in the light, so at twice the flow each drop spends half as long there and receives half the dose. The relationship is exact and instantaneous. A chamber holding 0.84 gallons at 8 gpm gives about 6.3 seconds of exposure; at 16 gpm it gives 3.1. This is the single largest variable in a residential installation, which is why sizing is done against the peak simultaneous draw rather than the average use.
How much does a fouled sleeve matter compared with an old lamp?
They multiply, and neither is visible from outside. A lamp at 70 percent of new output and a sleeve passing 90 percent of what reaches it leave 63 percent of the rated figure, before the water has absorbed anything. What makes fouling the more insidious of the two is that lamp fade is predictable and gets solved by a calendar, while a sleeve films up at a rate that depends on hardness, iron and temperature, so two identical units on different wells foul at completely different speeds. Units with a UV intensity sensor exist precisely because the two failures are invisible; the sensor sees the product of both.
Can I use tap water clarity to estimate UVT?
No, and this is worth being blunt about. UVT is measured at 254 nanometres, in the ultraviolet, where the eye sees nothing at all. Water that is perfectly clear to look at can absorb heavily at 254 nm — dissolved organics and tannins are the usual culprits, and iron and manganese both absorb strongly. Conversely water with a faint tint may transmit better than it looks. It is a laboratory measurement on a spectrophotometer, it is inexpensive, and it is the input that most often turns out to be the reason a unit is underperforming.
Does a bigger chamber always give a bigger dose?
Bigger in the right direction does. Making the chamber longer increases both the lit area and the volume, so residence time rises and the dose rises roughly with it. Making the chamber wider is not the same trade: it adds volume, which helps, but it also pushes the outer water further from the lamp, where both the one-over-r spreading and the absorption have taken more out. Past a certain gap the extra volume is holding water the light barely reaches. That is why high-output units get longer and add lamps rather than simply getting fatter, and why a low UVT pushes designs toward narrower annuli.