IDEA: immersed LED, no fiber

I see all of these designs where people are using fiber optics or rigid rods for getting LED light into the water. Why not just put a small amount of epoxy on the electrical connections and immerse the entire LED or laser module*? You will experience the absolute minimal light falloff as the water makes direct contact with the lens. For an LED I would suggest using a fostner bit to inset it into the plastic of your housing. It would be "water cooled".

What do you think?

* For a laser module you'd need to drill a hole and allow the majority of the module to hang out the back. Make sure you seal the front lens element well.

I've thought about the same idea! However, what stopped me was the simplicity of sealing a tube with a fiber out of the end, where if it leaked, it wasn't a big deal because there is no electricity.... Another factor for me was wanting to use a high powered LED, most of which are surface mount, as opposed to an LED with just the two wire leads that you would to seal (not sure if the regular LED would provide enough light for the stream). The third reason I went with fiber was that I was trying to minimize the diameter of the "disruption" that the lighting causes, and a nice, thin fiber seemed like the best bet.

That all being said, I would love to see what kind of design you could come up with to get the LED in the water, and the water cooling is really cool...Here's an idea: have some waterproof LED with its electrical connections sealed, and mount it to a small diameter platform or something, which could be supported by three arms out to the sides... and you could run the wires along one of the support arms and out the side of the nozzle. I know this would affect the laminar flow, but I just wonder to what extent. If those support arms are thin enough, might not be too bad. Maybe its better to just still run the pipe through the center from the bottom... Not sure tho, sounds like a good experiment!

I tried that some time ago, with different variants. Didn't give good results compared to the fiber optic method. The disturbance created in the laminar stream is just too much. It's very difficult to seal the LED properly, most epoxies do not work since after some time, they suffer a transformation (when submerged in water) that causes water to leak. If water is in contact with the LED, it could be dangerous and the LED system could not work properly. The light did hit the stream and lighted half the stream up, but when the light was off, the stream was very turbulent.

After all my experiments, I went back to using fiber optic with great results.

>If water is in contact with the LED, it could be dangerous

Please clarify why. Thanks

Because there will be a voltage in a water envirorment and depending on the power supply, even death could occur. Some Switch mode power supplies like computer PSU and other 12v PSU are not isolated from the mains and there's a potential for high voltages 100v to be present in the system with respect to earth that could electrocute a person if the LED's are touching the water. That's the dangerous part (I got shocked like 5 times experimenting, before giving up). Also, if there is an electric potential between to metals and they are in contact with water, a strong quemical reaction will start that will corrode them so quickly, ending the life of the PCB.

Of course, you could use an isolation transformer and other methods to increse safety but it's never 100% safe. I don't like mixing water with electricity, it's not a good idea unless you really know what you are doing.

Ah thanks. I was worried maybe the LED would explode or something if it got water inside the housing, but couldn't figure out why.

The second you put a pump in water, you're also opening yourself up to trouble. Reminds me of a mythbusters episode where they were trying to see if peeing on the 3rd rail could electrocute you (they should have been using a laminar jet :-) )

On a very serious note, there are many of this forum who don't have an electrical background and tinkering with the stuff could be fatal.

Power safety isn't something that has been discussed on this forums. Experimenters should be using either an RCD on the power socket or at the fuse box for any outlet which is powering their outside project. And also ensure that they test it regularly.

Also any supporting electronics should be fused a correct fuse, any metalwork earthed and waterproof. Make sure the laminar jet can't soak any electronics or sockets etc.

I'll start a separate thread for this.