UV disinfection: dose, lamp life and when it replaces chlorine
A UV unit looks simple: a pipe with a chamber and a lamp cut into it, no reagents, no change in the taste of the water. That simplicity is why it so often gets sized on a single figure — flow in m³/h — and why the result is a unit that lights up but does not disinfect. Here is what the outcome really depends on, and what you will have to do once it is installed.
The dose is intensity multiplied by time
The disinfecting effect comes from radiation in the 250–260 nm range: it damages the genetic material of bacteria and viruses, which then lose the ability to reproduce. The figure that describes the treatment is the irradiation dose. It is measured in mJ/cm² (data sheets sometimes write mW·s/cm² — the same thing) and equals the radiation intensity multiplied by the time the water spent in the chamber.
That gives three factors, and any one of them can take the result down to zero.
Emitter power. A single 25 W lamp and a bank of ten give very different intensities in the chamber. The STERILIZER range shows it plainly: UV6GPM is one PHILIPS 25 W lamp for flows up to 1.8 m³/h, UV84GPM is seven lamps and 280 W for flows up to 20 m³/h, UV120GPM is ten lamps for 32 m³/h. The irradiation dose is the same right across the series, 30 mW·s/cm²; what grows is the throughput and the connection size.
Contact time. Flow sets it: the faster the water goes through the chamber, the less dose it picks up. That is why a data sheet quotes not "capacity" but capacity under stated conditions — and the site's peak flow must not exceed it.
Water transmittance. The factor everyone remembers last. It gets its own section.
Turbidity and iron eat the dose
Transmittance is written as T and given as a percentage: the share of the radiation that gets through the layer of water and reaches the far wall of the chamber. Anything that scatters or absorbs ultraviolet — suspended solids, turbidity, colour, organics, dissolved iron and manganese — pulls T down.
How much that matters shows up on one and the same unit. DUV-1A120-N is a single-lamp 140 W chamber with a G 2" connection. At T=90% it treats 10 m³/hour, at T=85% — 8 m³/hour, on drinking water at T=70% — 5.8 m³/hour, on treated wastewater — 3.6 m³/hour. Same lamp, same chamber, nearly a threefold spread. Feed the unit turbid water instead of the water it was sized for and it physically cannot deliver the required dose — yet nothing will flag a fault, and the lamp will glow exactly as before.
Iron is worse still. It absorbs radiation throughout the body of the water and it also settles on the quartz sleeve as a film that dims the lamp from the inside. That is why UV comes last in the train: after mechanical filtration, iron removal and softening, not instead of them.
Lamp life: replace on hours run, not on failure
A UV lamp almost never burns out. It loses radiant power gradually while carrying on glowing in visible light — so you cannot tell a healthy lamp from a spent one by looking at it. That is exactly why replacement is tied to hours run.
Service life depends on the type of emitter. PHILIPS lamps in STERILIZER and Aquapro quartz lamps run for about 9000 h, after which radiant output starts to fall; Aquapro puts its scheduled maintenance interval more simply — a replacement lamp every 12 months. Amalgam lamps in DUV units last 12,000—16,000 h.
Counting those hours by hand is awkward, so the models offer two levels of monitoring. An hour counter simply shows how long the unit has run: on Aquapro that is the HT configuration, for example UV-12GPM-HT and UV-24GPM-HT. A UV sensor measures intensity inside the chamber and catches not only lamp ageing but sleeve fouling and worsening water as well: on Aquapro that is the M index (UV-24GPM-HTM, UV-48GPM-HTM), on DUV the ADVANCED and MASTER configurations.
| BAT23 | AAT23 (ADVANCED) | MAT23 (MASTER) | |
|---|---|---|---|
| Emitter | low-pressure mercury lamp | amalgam, 12,000—16,000 h | amalgam, 12,000—16,000 h |
| UV intensity monitoring | none | yes | certified system |
| Chemical cleaning | none | optional | built-in unit |
| Link to controls | — | remote switching by discrete signal | output to the process control system |
| Hydraulics for a given size | identical | identical | identical |
The hydraulic data sheet is the same for all three configurations: DUV-1A120-N BAT23, AAT23 and MAT23 all give 10 m³/hour at T=90%. The difference is only in the emitter, the monitoring and the pipework. A low-pressure mercury lamp is the choice where flows are small and short idle spells are normal: it runs cooler.
Maintenance that cannot be postponed
Ultraviolet needs no reagents, but it does need a routine. The minimum list for STERILIZER and Aquapro units:
- ✓clean the quartz sleeves every 1–3 months — a deposit on the sleeve costs you dose exactly as turbid water does;
- ✓replace lamps on hours run: 9000 h for PHILIPS lamps and Aquapro quartz lamps, 12,000—16,000 h for the amalgam lamps in DUV;
- ✓handle a lamp only with cotton gloves — a fingerprint on quartz stays there as a dark patch;
- ✓never switch a lamp on outside the housing, and never power up a dry chamber;
- ✓do not look at lamps in operation;
- ✓earth the housing with a copper conductor of at least 1 mm² cross-section;
- ✓leave about a metre of clear space along the axis of the chamber — without it the lamp cannot be pulled out;
- ✓stay inside the working range: for STERILIZER that is pressure up to 6 bar and water temperature +2…+45 °C, while the DUV chamber is rated for pressure up to 10 atm.
Check the room for lamp withdrawal before you order. The STERILIZER UV84GPM housing is 220×846 mm and hangs on a wall vertically or horizontally, and that metre of clear space at the end is needed not once during installation but every time the lamps are changed.
What ultraviolet does not do
The main limitation of UV is that it leaves no residual. Chlorine leaves a residual concentration in the water that keeps working all the way through the network to the tap. Ultraviolet acts only inside the chamber, and only at the moment the water is passing through it. Past the outlet flange the water is defenceless again.
The practical consequences:
- ✓A storage tank after the unit cancels the effect. If the tank sits downstream of the UV, disinfected water stands in it for hours and is recolonised. UV goes after the tank, not before it.
- ✓A long network with dead legs and stagnant sections is the case where chemical dosing is needed whether or not there is UV.
- ✓Secondary contamination goes unnoticed. A loss of tightness, a new tie-in, a section repaired downstream of the unit — ultraviolet will never know about any of it.
That marks out where UV belongs: point-of-use disinfection — at the inlet of a house on a borehole, ahead of a bottling line, ahead of fine filters, on utility and process water, in swimming pools and on treated wastewater. It works without reagents and leaves the taste of the water alone, which is precisely what a food plant or a coffee shop is after.
What next
For sizing, gather four figures: peak flow, not average; transmittance and turbidity from the water analysis report, along with iron and manganese; the type of water — drinking, utility, pool or treated effluent; and the connection size of the pipework. Check separately whether a storage tank sits downstream of the installation point and whether there is enough room along the chamber axis to change the lamp.
With that in hand you can turn to the catalogue: DUV units cover the industrial range and provide UV intensity monitoring with an output to the process control system, while the STERILIZER series offers multi-lamp SS304 chambers for a house, a pool or a small business. If you do not have a water analysis report yet, start there: ultraviolet cannot be sized on flow alone.
