UV Sterilization for Purifying Water

The efficacy of a UV sterilzation system is going to be a function of how long the water is exposed to the light, and how much water the light must penetrate. Increasing the flow rate of the water above the design flow rate results in loss of protection. Increasing the size of the flow chamber has the same effect.

Saying that your UV bulbes last for years is very misleading. They may burn for years, but the light spectrum changes with time- that is why most suppliers recommend replacing them every year. If you are using the same bulb for years, you are most likely not getting the protection you think you are.

For excellent guidance on the principles of UV sterilization, I recommend the EPA Publication, EPA 815-R-99-014, "Alternative Disinfectants and Oxidants Guidance Manual". The version I have was published in April 1999 and was freely available from the EPA website.

GA Cwarner I am glad you picked up on the time of use limitation. I posted this just after you did and read your comment. Good show. My attached site on UV requirements may be simpler to read but EPA site is good.

Attached is the requirement regulations for small UV systems in North America. It would be somewhat difficult to determine the quality of the rain water after it has entered a sump. I am not sure if the water is collected from roof tops or if it is ground water runoff. Remember birds do not only "sit" on the roof. In either case you may have some difficulty in determining your UVt (UltraViolet light transmissibility) at any given time. It should be noted that the UV must be capable of a minimum of 40mJ/cm2 at all flows and you must obtain 3 log (99.9%) inactivation of bacteria and 4 log (99.99%) inactivation of viruses. I would suggest a minimum of 1 micron absolute filtration precede the UV and you may have to step the filtration to get to 1 micron. (Something like a 10 micron then a 5 micron before the 1 micron). We have used Hallet UV systems which impart a dosage of 60 mJ/cm2 after 9900 hours. UV lamps are designed to be changed annually because they lose intensity with time of use. The Hallet has a little better margin of safety with the higher dosage. The bulbs are protected from the water direct contact with quartz sleeves (UV will not transmit through glass). A smaller unit can be used as long as you use a flow restrictor to match the output of the UV system. Wattage will vary but generally about 100 watts will be required.

Hi Upsidedown,

Just leave the lamp on - as others have suggested they last for years.

I work inthe water treatment industry, and have participated in much debate over the issue of lamps on or lamps off (when not in use). Turning the lamps on/off decreases the lifespan of the lamps. I've done the numbers - cheapest just to leave the lamp on. The power use is minimal to broken bulb from off/on cycling.

The kill efficiency of the UV unit depends on the dose of 254nm light and the length of exposure (which comes down to your flowrate through a specific sized unit). The problem is in water that is not filtered, small particles provide refuge for the bacteria. This reduces the dose rate and therefore the kill efficiency. For the UV to be effective you should filter the water first to prevent suspended solids providing refuge for bacteria (maximising the kill efficiency).

Something else to note. UV light will sterilise in the immediate vacility of the unit. It doesn't prevent micro-organism growth downstream of the unit (unless that is perfectly sterile, which I'm sure it's not). And the bacteria that is killed, is decomposed into food for the bacteria that aren't killed. Within hours the bacteria count will return to pre UV levels.

But consider this. We've found the UV light causes oxidation of organic molecules. Essentially it breaks them up into smaller pieces that creates food for micro-organisms. The UV could make the problem worse, not better. It could promote bacteria growth.

Rain water is typically very low in nutrients. It may collect nutrients/organic material from the surface it lands on and collects in. In your case the sump. What is the surface it collects on??

My point is - if it collects on an inert surface, it may not have any nutrients. If there are no nutrients in the water, the bacteria that is present can't grow.

Putting a UV unit in could cause you problems.

In the industry were moving away from UV (as a sterilising agent), as the effect is so localised, and they can cause problems by creating food for bacterial. Really they're just a "feel-good" treatment option. I wouldn't waste the money.

Alternatively, consider using chlorine dosing, to prevent bacteria growth - not UV. There are slow release chlorine tablets. Periodically drop them into your tank/sump. Use test strips to monitor the free chlorine concentration. Maintain approx. 0.5 - 1.0 ppm free chlorine, and you'll be fine.

What is the end use of the water? if it's drinking water, maybe consider other filtration options, if you are still worried about the quality.

Hope this helps.

AJ I liked your answer except for; "Just leave the lamp on - as others have suggested they last for years."

The UV bulb may be on, but as pointed out, the intensity is inadequate to provide safe disinfection. There seems to be a false sense of security about the operation of UV units that provide a bulb with an on or off indicator. It is expected that the intensity of the bulb will lose at least 20%/year of the rated intensity (after 100 hours). Bulbs will last a long time but their effectiveness is diminished such that the manufacturers recommend changing them every year.

However, I agree that the UV should be operated continuously under most applications. I suspect that Upsidedown lives in an area where rainfall doesn't occur for about 8 months and thus the UV is just burning in vain. He should have a total time used meter on the UV unit and have the unit come on for about a minute prior to pumping water. OP could do this with level controls in the sump. This will warm the unit up to reach safe intensities for sterilization. Some units can be purchased with a flow indicator that activates and increases the intensity automatically. These type of units are on continuously but at a lower intensity when not in use. The intensity increases nearly instantly to the proper level.

One problem I have noticed with homeowners using a domestic UV system and sampling of water is that they reduce the flow of water during sampling. In effect, they are providing a large contact time for sterilization and an increased flux. That contact time will be reduced at normal or peak flows and so will the UV flux. A dosage of 40mJ/cm2 is required at all flows including peak flows. Sampling of domestic UV should only be done by maximizing the flow to avoid such false security. All samplers need to be better educated about sampling techniques.

If Upsidedown is using a storage chamber, then some sort of disinfection should be applied to the storage. Chlorination is usually applied except he will need to refresh the residual periodically as the water sits in the chamber. In fact, OP may be better off using chlorination as a primary treatment and then using the UV as a point of use treatment. This would provide the system with good redundancy and an easier operation.

The water used at the bothy here is groundwater collected in a well. A Draper (other submersible pump manufacturers can be found) submersible pump lifts it through a proprietary 5μm absolute filter into a 100L covered storage tank located in the loft about 6m above top well water level when a level switch in the overhead tank goes low. The pump is protected from dry-running using its proprietary float switch, which stops the pump were the well level to fall too far for the pump to operate; it never has fallen that far, fortunately.

That stored water that is used for cooking, drinking and tooth-brushing gets zapped with a Kleiber (other UV unit manufacturers are available) UV unit that is left on all the time, and takes 62W from the mains supply, contributing about £120GBP per year to the electricity bill; it is the "base load". It has 1" BSPM connections that are adapted down to 15mm copper pipework that uses regular compression fittings. The UV bulbs get changed on a planned preventative maintenance schedule once a year on or about the anniversary of occupancy of the building (easy to remember), and this can be done without draining the water from the system. At least 1 replacement bulb is always kept in stock and 1 spare glass envelope is kept for emergency in the unlikely event of its breakage; replacing an envelope would necessitate turning off the isolation valve upstream of the UV unit, opening the cold tap on the washbasin in the lavatory room, and opening the air bleed on the top of the UV housing. Getting the air out of the system again can be a bit of fun-and-games...

There is a label on the housing: "Do not look at the light from the bulb while the UV unit is on!"

The well filter cartridge gets changed when there is sufficient blockage for the water not to arrive in the loft tank. One cartridge can last two years or more.

The water gets checked for bacteriological and mineralogical acceptablility and a report provided on it every five years by the Local Authority, for which a small fee is paid. There were been no improvement recommendations on the last report to hand.

And that's it.