They definitely have a lifespan and grow less effective over that lifespan. I have a 9W bulb and it is rated at 8000hrs. I replace mine yearly at the start of spring and it does make an immediate difference when I do so. You should google the part number of your bulb and determine it's effective hour rating. As far as the expense of the bulb, look on E-Bay. My local pond shop sells mine for $36.00. I found them on E-Bay for $14.00.

It really depends how you use it. I started turning mine on only for a couple months in the spring when the water warms up and starts to turn a bit green from algae bloom. By June the plants have established themselves and the UV light is no longer needed, so I turn it off. Using it that way I can get 3 or 4 years out of one bulb.
Of course some people may have a heavy fish load and not many plants and will need to use their UV light all year long, in which case it will need to be replaced yearly.

We remove our bulb in the fall, wipe it off, wrap it in tissue and store it for the winter. Our current one is on its second year this spring. I imagine we will need a new one for next year. We reinstalled ours a week ago and our pond was so murky but cleared up in just a few days. Ours is a 45 watt in a 3500 gallon pond.

Nanahanna

Just keep track of the useage and when it hits recomended use replace it. I only use our UV(c) light at the start up of the season to get rid of any algae or until plants can start to consume nutrients so algae is not a problem. ONce that is done, we do not use it any more...sop at most it may be on 1 month.....so theoretically I should be able to get 12 years out of a bulb.....but its not hard to check because if algae starts to grow and the bulb is on, then odds are the bulb is in need of replacement. So far out one (Hozelock Bio Force unit) is in use close to 3 or more years and still kicks algae's butt!

We have a 40 watt UV. This is the 4th year for this bulb. It basically is in from April 1st to October 15th every year. We take it out and put it on a shelf. I won't put in a new bulb until I see that it isn't working and the algae starts growing.

Hi, keithw you mentioned that you buy your bulbs on e-bay,
which buyer do you use?

Here is the Ebay store where I found mine, they do not have lot of different models available. http://stores.ebay.com/Lights64

I searched quite a bit on the web about the effective life of UV bulbs and almost every reference that I can find for any bulb is 6000-8000 hours which translates to 1 year of 24/7 usage.

curb1, UV lights can fool you. They don't always go completely out. They just get dimmer and when that happens they are useless. I have no idea how to test for this though.

Here is a link that might be useful: My pond the way it was before the freeze.

Hey thanks everyone for your helpful tips and comments. I've used the same 55 watt bulb for the past 2 years with no problems and that was running it 24/7 during the summer months. I think what I'll try this year, after starting with a brand new bulb and cleaning the quartz sleeve, is only turning it on when I see the water turning murky.

Let us know how it works. I have thought about the same thing. I am going to use this same bulb (4 years old) until water starts to turn green. It clears up so fast that it shouldn't be a problem.

Most manufactures rate their bulbs in the amount of time it takes for the output to drop to about 60% of a new bulb. Usually this is about 9000 hours of use. This doesn't mean that the bulb for your pond needs to be changed at 9000 hours. Let me give you a couple examples. Lets say you have a light that is rated for a 2500 gallon pond and 9000 hours of use and you have a 2500 gallon pond. Then at the end of 9000 hours you should change the bulb. Now lets say you have the same UV light but your pond is only 1200 gallons. In this case you could probably go close to 18000 hours on the bulb. I know you are saying how is this possible because the bulb is wore out after 9000 hours. Well at the end of 9000 hours it can handle a 2500 gallon pond but your pond is 1200 gallons and because it is smaller it only takes 30% of what a new bulb will give you which is half of what a 60% bulb will give you. Now to confuse you a little this only true if you have a flow rate half of what the max flow rate for the UV. But with the UV being twice the size that is needed than the lower flow rate will work just fine. The only problem is the bulb may burn out before that many hours are on the bulb.
Mike

Mike I really would like to hear your theory confirmed by an electrical engineer. I simply cannot wrap my mind around that one. It seems to me the bulb is expending the same amount of energy and deterioration no matter the volume of water or flow rate. It is simply the hours of usage verses the fatigue of materials.
Sorry Mike I need more convincing on this one.
Charles

Here is a link that might be useful: My pond the way it was before the freeze.

Charles the bulb is usung the same amount of energy no matter the length of time but what happens is the energy at a certain wavelength drops off. So most manufactures rate their UV's at 60% of their new output. It takes so much energy to kill algae. This is done by taking the energy output of the bulb at a certain wavelenght by the amount of time that the algae is exposed. The amount of time that a UV is rated at is the max flow rate. The size of the pond comes into account by determining what point is the algae being killed off verus how fast is it reproducing. When algae is dying faster than is reproducing the population goes down and the water clears. By decreasing the size of the pond the flow rate can decreased with the same result. This is why a smaller pond can get by with a smaller unit. The only thing that I said that may not be correct is the amount of time you could use the bulb as I assumed that the deterioration rate of the bulb stays constant and I don't know if this true or not as I haven't seen any data on how a bulb deteriorates past the 60% level.
Mike

I hate to argue but I have to agree with bugsb in that the explanation above does not seem to make any sense. Bulbs are rated for a number of hours of effectiveness. The standard for that rating may in fact be based on a percentage decrease from it's initial effectiveness but the deterioration is consistent when the bulb is running regardless of the volume of water flowing through the filter. All that being said I agree with curb1 and wanderoze above about only running the light when necessary. When I first installed mine, I ran it 24/7 all year. Now I turn mine off as soon as fall starts and only turn it back on when I start noticing string algae forming in spring.

Kieth,
You say you turn it back on when the string algae appears in the Spring. The "green soup" and the string algae do have similar spring appearances in our area. But, it is my understanding that the UV lights do not have an effect on the string algae, only the "green soup".

This whole controversy can be eliminated by running the "old bulb" until it doesn't work, and then popping in a new one.

Let me see if I can explain this again. I don't want to argue with anyone either. Let me use Emperior Aquatics as they publish more information about their bulbs than does anyone else. Their 25 watt bulb produces 8 watts in the 240 to 260 range of wavelength which is the wavelength that is used to destoy the DNA of the algae. At 9000 hours it produces 4.8 watts in this range and at the max can handle a about a 2360 gallon pond at 7.8 gpm. The flow rate that Emperior gives for a 30000 uWs/cm2.It takes 22,000 uWs/cm2 to destoy the DNA. The water in the pond needs to turn over every four hours for the light to sterilize the pond. Where size comes in to play is if cut the flow down to 3.9 gpm the watergoing thru the light would see 60,000 units at 9000 hours of use. But at that flow rate it would only handle 1180 gallons. But if the pond was only 1200 gallons this would be more than 2 times the amount of radation the algae would need to be sterilized. So we could use this bulb until the radation drops down to 22,000 units. This would work out to be about 30% of the output of a new bulb. I am sorry if this is too complicated but it is hard to make it simplier.
Mike
Mike