Currently I have a 100 gallon pond (will be upgrading perhaps this year). The filter kit comes with filter elements, bacteria media and a UV-Clarifier lamp. Oh, and yes, of course, a pump. It's not possible to tell if the UV lamp is working without pulling the filter out of the water. I don't know if there's some sort of pressure switch that prevents the UV light from being active when not submerged. And for some reason those fluorescent type UV-Clarifier lamps seem to not last very long. Also, finding a replacement lamp - so far I've been unsuccessful. So I'm wondering if it's possible - or ill advised - to replace that fluorescent lamp with a UV lamp. Anyone have any suggestions? Recommendations? Cautions?

The current unit appears to run on some sort of ballast.

I don't know that it's critical to replace the ballast (if that's what it is) but I have no other way of verifying it's working. The unit that houses the lamp is at the bottom of the pond and difficult to reach on short notice. This (tracking disabled) appears to be a replacement lamp but I don't know how to get to the old one. And if I can figure out how to access it - I'd like to replace it with an equivalent LED type UV-C lamp.
The pump:

Somewhere in the garage - gotta go find it - is an older unit I can experiment with without putting the fish at risk. I'll update as I learn things about this pump.

 

Well, OK, I found the old pump. Bad news - it doesn't have the UV assembly.

 

Have you seen this?

 

Thanks.

 

OK! Good news! Found the old Fluorescent UV lamp holder. Seems you just pull the glass off. Will have to silicone the seal back when done. Lamp change is easy enough. But still want to replace Fluorescent UV-C with LED UV-C.


So can anyone make any suggestions on how to modify this thing? Remember, it has a ballast that will need to be chopped out of circuit and attached to some sort of DC circuitry. Depending on the specifics of the LED Data Sheet and the final voltage to be chosen along with a current limiter - - Oh, wait; Think I saw a driver on that webpage Tony gave. Hang on for dimensions of the glass tube.

[edit]. Glass tube ID is 1 .25" diameter. Depth =2.858".

 

Does this look like the lamp in your glass tube?

 

UV-C LEDs are expensive and unusual. Have you sourced them already?

 

Been looking at Tony's link. Considering that the replacement lamp for the unit AS-IS, it's $22 (US). If I go with a single "Tripple" unit, cost of $59.54 (US). Will need their constant current driver at a cost of $10.49 and their "HexaTherm Tape", costing $7.99. A cost of $78.02. Yes, expensive. But if estimated to last 30,000 hours, that's 1,250 days; or more than 3.4 years. If I'm buying new fluorescent lamps at $22 a year for the same amount of hourage (made-up word of the day) I'm looking at $75. BUT the advantage is not having to change the lamp every six months. I've had UV-C lamps burn out in as short as two weeks. I have one in my ventilation system right now that has been burning since December first (last year). The longest lasting lamp thus far. And I'm four lamps into the system. And that's with turning the lamp off in spring time and back on in the fall. Since March of 2020, four lamps. Averages suggest the lamps are short lived. I have been checking the current lamp to make sure it's up and running as long as the furnace is on. I MIGHT get more life out of it if I rigged up a switch that turned the lamp on only when there's an air stream. Now there's an idea. When the fan is off - no sense in burning the lamp. UV-C LED's may be the way to go in that system too. Never thought of that before. A little steel work, a little electronics and I have an LED UV-C system in my ventilation system. Yeah - going to think on that one too.

The cost is a wash but the benefit is confidence my pond is being managed properly and continuously. Thanks to Tony I believe I have a workable solution. For a heatsink I have some older LED lamps that I've taken apart. One heatsink is already close to the proper size. A little mill work and it will fit the glass tube nicely. Some thermal paste to make good thermal contact with the glass and I have a perpetually hydraulically cooled LED UV-C clarifier solution. Yeah, a little work for sure, but one well within the scope of my capabilities and tooling.

 

A potential source: https://www.intl-lighttech.com/applications/uvc-leds

 

OK, well, I've pulled the trigger. Getting LED UV-C for the pond. Milled a heatsink to fit inside the glass tube where the LED's will live. Parts should start arriving in about three days. Meanwhile I've disassembled the fluorescent UV lamp to see what failed. Verdict is in: Bad starter. Otherwise the tube would work.




As you can see in the first photograph I've already installed a 2W04G Bridge Rectifier into the base of the old lamp. The reason for the BR is due to the possibility of plugging the lamp in either polarity. Plug it in wrong and there goes a bunch of cash. So with the BR, regardless of how I plug it in - the LED's will have the proper polarity.

The LED's will be driven with CC 150mA. The old housing that once had the ballast for the fluorescent lamp will now house the transformer and the CC driver. There will at least one diode to rectify the current for the driver. Exactly what transformer I'll be using - I have to peruse through my junk drawers. I'm sure I have a 24V transformer, 18V doorbell transformer that has three different outputs (18, 12, & 9VAC). Later this evening I'll pick through my transformers. As for the driver, I believe it's capable of handling 28V input. Since I'm rectifying and possibly filtering I'll have to choose a lower voltage transformer. Maybe 18V. Power consumption is low according to the data sheet. @Tonyr1084 thank you for the link. I'm getting all the parts and materials I need from them. Costing nearly $90 (US). But for the cost of four fluorescent tubes to cover the same lifespan, $22 ea. x 4 = $88. But with the LED's I'm confident they're working. I've gone through so many fluorescent lamps in the past - it's time to upgrade.

 

Sorry to be late to the party. I have two ponds that need UV to keep floating algae at bay. I buy these or similar, get a year or two out of them. I have replacement bulbs but by the time they need that the pump or water seal is shot anyway. Oh, and they come with a little window to check the UV is on.

Ebay item link

 

Thanks ErnieM expecting LED UV-C and driver module to arrive today.

That pump/UV unit is just a little over the cost of replacement lamps for my pond filter. What concerns me is deteriorating seals. Also, your unit isn't a filter. Concerned about that as well. But thanks for the link. It could be an "Add-On" thing I do when I build a bigger pond. It's just that the UV LED's would last much longer and I'd have a greater sense of security with my pond remaining free of algae.

I'll definitely be posting pictures of the build once I get the parts.

 

Finally got the parts. And they sent the wrong LED module. I've sent them an email. Let's see how long it takes for them to respond.

 

OK, just got the right LED unit. I'll put it together in the next few days and post some pic's.

 

So far I have the LED UV-C mounted on a heat sink. The connection to the old fluorescent base has a bridge rectifier in it because I don't know which way the LED will end up being plugged in. Since there is no predictable way to connect it - the BR is there for protection only.

Later today I will apply thermal grease to the heat sink and glass tube so that it radiates its heat into the pond water. I've already tested for illumination and yes, it does light up. I tested it with low power, not the 150W the driver will deliver.

The project may be delayed a day or two, there's something up with one of my pets and we're taking her to be looked at. Hopefully all is well enough and can be easily treated. Cats seem to know when they need bed rest, and that's what she has been doing.

Will post further pictures either tonight or tomorrow (likely).

[edit - adding pictures]

 

After basic assembly and testing I must say I expected a lot more light energy from this. Maybe I just don't understand how strong UV light is.

Using a 15VDC 500mA wall wart, connected to a "Buck Toot" (BT Driver) 150mA Constant Current driver I get the following results:
19V unloaded
15V loaded drawing 2.5mA
BT output into UV-C LED's 87.2mA

Using a 13.8VDC 19A power supply connected to a BT I got the following results:
13.8V unloaded
13.8V loaded drawing 2.5mA
BT output into UV-C LED's 89mA

The BT is supposed to provide a constant current of 150mA and operate from voltages ranging from 5VDC to 28VDC.


I don't know if I should expect a more energetic light source. But at 89mA when it should be 150mA - I'm unsure if I've done something wrong or if something is not functioning properly. Testing shows insufficient current. That supports the hypothesis that there should be a more energetic beam. I know UV is dangerous so I've limited my exposure and have worn UV blocking sunglasses, but the pictures don't lie. Either the BT is faulty or something is up with the LED's.

Possibly, would a black light poster reveal more energy? This is supposed to go inside the fish pond filtration. I've spent close to $100 on this and would hate to have wasted that much money.