Pond filter with UV lamp

[Bob McParticipant @bobmc91481]

Hi All, Tea Room again!

Replacing a fluorescent type UV tube with an LED.

The problem/s. … I fitted an 80W solar panel on my shed roof to power a small 12v DC pond pump, it works great using the battery regulator supplied with the panel, however the ultra violet lamp used in the filter housing is 230v AC and I have rigged up a circuit (relay) to switch on the UV lamp only when the pond pump is being fed with power and water is then flowing over the UV lamp.

This also works a treat but it would be more efficient to use an LED UV light, (I assume), this would give me a completely free pumping and filtration setup using the solar panel.

The problem is that I don't know if the UV LED striplights have the same effect as the glass tube type lamps and there does not seem to be much information about these lamps, I have seen some led UV lamps for aquariums but these appear to be for growing plant conditions.

If anyone could give me advice about this it would be very much appreciated.

rgds..Bob.

[Bob McParticipant @bobmc91481]

Any advice about using LED UV lights

[SteviegtrParticipant @steviegtr]

I've had a Koi pond for years. I did not know you could get LED UV. for Algae control.

Oh well learn something everyday.

Steve.

[JasonBModerator @jasonb]

Not heard of any germicidal LED lamps either and like Steve had a koi pond for years.

[Nigel Graham 2Participant @nigelgraham2]

It would seem UV lamps can have two effects, perhaps by wavelength and intensity, that may control choice.

Some will encourage algae and I have seen this used in glass-tubular "green-houses" to produce the food for baby scallops in a hatchery. (The shellfish are filter-feeders who dine on plankton.)

Others will kill bacteria and are used in water-purifying. (A lot of pathogens that need anoxic, dark conditions, die off when subject to fresh air and bright sunlight – as well as those conditions probably starving them as well.)

I don't know if both effects can occur together: the UV radiation good for photosynthesising plants but bad for the more animal-like protein of a bacterium.

I think this an area needing a certain amount of research before buying anything.

[Bob McParticipant @bobmc91481]

Thanks all for your replies…so quick!!

Steve thanks for reply… perhaps my wording was a little bit ambigous….no I am not saying you can get UV LED lamps for algae control….this is what I am looking for!

And if Jason hasn't heard of it … its not looking as if they exist! Nigel is probably right that it needs some research, bacterial growth and things of a biological nature is the area I'm not that conversant with.. will have to ask at the fish shop…

..Bob…

[Michael GilliganParticipant @michaelgilligan61133]

395nm is probably the shortest wavelength LED you will readily find: **LINK**

https://www.ebay.co.uk/itm/271840695294

… and it is only just into the UV part of the spectrum

>> If you decide to play with these, take care … they could blind you <<

UVC is the antibacterial, etc. range

Ref. http://solar-center.stanford.edu/about/uvlight.html

MichaelG.

Edited By Michael Gilligan on 30/04/2021 14:54:07

[SillyOldDufferModerator @sillyoldduffer]

Though unusual they do exist, like this one on Amazon.

No idea if they're any good – my fishy neighbour moved away years ago.

Dave

[SamsarandaParticipant @samsaranda]

In respect of UV lamps for ponds they must be the types that generate the wavelength necessary to kill bacteria, I.e. germicidal types. As far as I am aware you can only generate those wavelengths in a traditional glass/quartz type enclosure as in traditional UV lamps, as far as I am aware leds do not have the power to generate the required wavelengths in the quantity required. I may be wrong on this and technology may have progressed in which someone will be along soon to shoot me down. Dave W

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Michael Gilligan on 30/04/2021 14:45:17:

395nm is probably the shortest wavelength LED you will readily find:

[…]

.

O.K. … I’m wrong

The lamp in Dave’s Amazon link claims 365nm

… and ebay seems littered with UV-C adverts [shown with violet beams for artistic effect I presume]

MichaelG.

Edited By Michael Gilligan on 30/04/2021 15:08:07

[JasonBModerator @jasonb]

The 395 and 354 are UVA but as mentioned UVC is what is used to make the algae flocculate and that's 100-200nm

[Bob McParticipant @bobmc91481]

Wow..!

thanks to all for your inputs, it seems I am asking for something at the edge of technology, will be having a look at what Dave found on Amazon..

thanks again…Bob..

[Robert Atkinson 2Participant @robertatkinson2]

I've done some reseach on UV sterilisation (there is a patent on the subject out there with my name on it). The required waveband is has been mentioned UV-C which is wavlengths below 280nm The "classic" UV fluorescent lamps emit mainly on the mercury "line" t 254nm and are very effective. Note that even these have quartz glass tubes rather than normal glass as UV-C will not pass through normal glass.

You can get UV-C LEDs. The 285nm LEDs are not technically UV-C. Most common are 275nm so just in the UV-C band and not very effective. Even these are not cheap. One off from a reputable source is £2-3 for a <5mW LED. .
The only LED's that are really suitable are the 269nm ones. These are very expensive, over £100 for a single 50mW LED from a reputable source. You would need abot 100 of these to replace a single 8W 254nm tube.

The conventional UV-C tubes are very efficent, probably more than an LED, stick with what you have.If you want to run off solar find an old caravan or similar 12V fluorescent lamp that has the same size white light tube and use the inverter out of that

As others have said some wavelenghts will actually promote growth of things you don't want so buying a cheap LED unit from an unknown seller could be wors than nothing.
Anything with a plastic cover is not going to work, UV-C will not pass through it.

Robert G8RPI.

[Bob McParticipant @bobmc91481]

Thats fantastic Robert..!

this is the information I needed, so it seems that I won't be replacing the tube I have anytime soon.

I have got an inverter to power it but will leave it connected up as it is… thanks again for this… it just

goes to show that there are people on this forum who are 'dark horses' with exceptional knowledge.

nb..Dave, I looked at the one on Amazon and that one is 220v working so there would be no benefit.

thanks again rgds…Bob..

[JasonBModerator @jasonb]

Posted by Bob Mc on 30/04/2021 17:46:29:

this is the information I needed, so it seems that I won't be replacing the tube I have anytime soon.

Usual advice is to change yearly as they do seem to go off, though depends how long you have them running

[SillyOldDufferModerator @sillyoldduffer]

Posted by Robert Atkinson 2 on 30/04/2021 16:26:25:

…

You can get UV-C LEDs…
The only LED's that are really suitable are the 269nm ones. These are very expensive, over £100 for a single 50mW LED from a reputable source. You would need abot 100 of these to replace a single 8W 254nm tube.

…

The technology is worth watching because history says prices will drop once these things become mainstream. Check prices again next year.

On a side note I can't remember when LEDs suddenly went from rare and exotic to commonplace. My first LED was green, dim and expensive; my guess early 70s but possibly later. Not long after red and then yellow appeared – cheaper and brighter. Then suddenly they were everywhere. Anyone remember the exact timeline?

Dave

[Bob McParticipant @bobmc91481]

Thanks Jason… good advice as usual, I meant… change it to an LED one…

Dave, you are most likely right about cost coming down, I too remember they were, to my mind at the time an absolute marvell especially when you had been previously dealing with small filament bulbs which must have been ~100 times less efficient , I couldn't understand why they didn't make front page news!

I don't recall when they first appeared but as you say it was probably early 70's … and now look at what we have in the semiconductor display arena, along with computer storage which I understand still follows 'Moores Law' .

In the late 70's I remember going to Honeywell for an interview as a computer engineer, didn't get the job but they did tell me that they had solid state storage of 1GB….! I nearly fell off the chair! I did have a small computer .. it had a RAM of 1k… I remember it took me a few hours to get it to draw a circle and then I got a program on cassette with which you could play a tennis game…. and I think it made a 'plop' noise as the bat hit the ball… it was marvellous!

Ah well… must'nt reminisce too much I never get this posted! rgds..Bob..

[Robert Atkinson 2Participant @robertatkinson2]

Hi,
Annual lamp changes are pretty conservative. Normal flourescent tube life is 20 to 60 thousand hours depending on source and a lot of loss of output is due to the phosphor coating which of course the UV-C tubes don't have. If on 24/7 then a 2 year life is almost certanly OK.

Covid 19 might bring down the cost of 269nm LEDs due to large scale production for sanitation robots but recent research indicates that surface contact is not a significant infection route for Covid.
Even then they are not as efficent as the fluoresent tubes. Thatypical small "8W" 16mm dia 288mm long tube produces 2W of UV-C output for 7W electrical input.

http://www.osram.com/ecat/PURITEC HNS UV-C-UV-C lamps for purification-Ultraviolet lamps-Industry-Specialty Lighting/com/en/GPS01_1028570/ZMP_4021034/

In comparison a LED consuming 3W electrical input only produces 0.05W UV output.

https://docs.rs-online.com/edb9/A700000007304392.pdf

This would be 120W of power input for the same output as a 7W input tube. About £100 a year more to run on electricity consumption. I don't see tubes being replaced any time soon

Robert G8RPI.

[Author]

Posts