Note: I have been working on a new mini 20 watt road light project now for about a year. I ran into some problems with the design and to this day Im still working on them from time to time. This is project is going to be more of an on going project log than a how too article.
This is just one of those projects that frustrates me more than anything because its more complecated than it looks. Its one of those things I will just pick up and work on when it hits me right to do so. Maybe someday I will get everything right and finish this one up.
January 6, 2008
A couple of years ago I made a made a set of running lights for my motorcycle. I intended to make a few sets to sell but never did because moisture kept getting into the light assembly. This didnt hurt anything but I found it very anoying and decided not to build anymore until I could get this problem worked out.
Fast forward to today.... I finished a prototype motorcycle running light tonight that I designed. This new design should be water tight so I submerged it into a bucket of water with the bulb on and it was still working when I pulled it from the bucket 15 minutes later.
My problem is that Im still getting water behind the lens. Its not much but I think its from condensation.
The light is machined from billet 6061 Aluminum and uses a 20 watt MR-11 lensed bulb.
Both sets of lights I have made look good but this moisture problem is really starting to aggravate me to the point where I thinking of scrapping my lighting project.
Here is an exploded view of the light with all the parts removed. I threaded the lens cap so that it screws on like a flashlight does. There is a black neoprene washer inside the lens cap that you cant see but it seals the whole unit.
I think the water is just condensation because when I turn the light on the lens will fog up but quickly burns it off. Submerging the light in water seems to intensify this effect to the point where there are several drops visible inside the lens after 5 or 10 minutes.
The reason I am submerging it under water is to simulate a severe storm condition. I want my lights be able to perform in bad weather.
Here is a photo of the set I made a couple of years ago which have the same moisture problem. These are also machined from 6061 T-6 Aluminum but have lensed MR-16 bulbs which are larger in size.
I was trying to better this design by making the new lights smaller and more compact and with a better lens retaining system.
The MR-16 bulb in the light below is held in with a simple snap ring.
Both the MR-16 and MR-11 bulbs Im using have non-removable glass lenses.
I'm going to modifiy my testing methods from submersion to spraying water on the test part (which would probly be more accurate anyway).
Maybe I have been going about this all wrong and the housings don't need to be completey water, maybe they just need to vent better. I may try either drilling a small hole in the bottom of the housing or milling a slot on the bottom along the threads which would angle the vent so rain could not directly enter the housing.
But first I have some cleaning up to do.
.............What a mess!
If I remove the stem from the light housing there is a 1/4 -20 threaded hole which exposes the interier of the housing. Instead of drilling new holes in the part Im going to do some experiments with the stem off and the hole pointing downward, but instead of submerging the light Im going to run it dry first to once again see if I get the brief moisture problem. Then instead of submerging the light Im going try spraying water on it for about ten minutes to see what happens.
I will post the results when I finish.
I turned the light on while it was on the work bench and this time I didnt get the fog and burn off that I got the first time.... hmmmmm.
Its starting to look like leaving an opening was a good idea!
Then I took the light outside (I was kind of excited at this point since it might work this time) and turned it on to let it heat up. Then I sprayed a stream of water over it. This time the bulb imediately started filling with water so I shut everything down. The front seal failed this time, so I guess its back to the drawing board for now untill I can re-design the front sealing area (no extra machining just different seals).
I may have to switch to the un-lensed MR-11 bulb and buy some plexiglass to make my own lens cover if I can find any around here. Un-covered MR-11 bulbs are cheaper and easier to find anyway. Im going to also try and find an old inner tube this week that I can make a new gasket out of instead of using foam rubber.
Some things were learned from this failed experiment today though. There is some benifit to having the area behind the bulb vented, but the area in front definantly has to be water tight (or so it seems anyway).
The lens covers on these bulbs seem to be just glued on but not water proof so Im starting to dislike the covered version of the MR-11 anyway.
I will make some design changes and continue this thread later in the week....
January 7, 2008
Ok, here is an update on the project. I took today off work and went to my Local True Value Hardware store and picked up a few supplies.
The man that helped me donated a small square of 0.090 thick plexiglass for the lens (It was free so I thought I would give it a try).
I also found a few cheap rubber washers that I had to cut to the right diameter for the project. I sandwiched the newly made plexiglass lens between the rubber washers.
I switched from covered 20 watt lamps to 35 watt un-covered lamps (which is what I really wanted to use anyway) and replaced the 18 gauge wire with 16 gauge wire. I switched the light on but it didnt even take two minutes before the plexiglass lens started to fail (I was kind of expecting this though). But, the rubber washers sealed perfectly and Im possitive that they will no longer fail.
Lesson learned: With out a doubt I have to use glass for the lens. Now I just have to find a place to get a glass lens which may not be easy. I will update again once I find a new lens.
Notice the plexiglass lens in the photo below. It slowly bubbled outward from the assembly as it heated up (This was kind of cool to watch but was not the result I wanted, but hey it was free).
January 12, 2008
I got my glass lenses today and I've got a really good seal on the light this time. New glass lens, two flat rubber gaskets seem to be doing a great job so far. I let the light burn on my work bench for 20 minutes and saw no sign of any moisture this time. The only vent is to the back of the lamp through the electrical wire hole.
P.S.. Im using the 35 watt lamp for these tests.
More tests to follow....
January 13, 2008
Its too darn cold to be outside spraying water today and besides that my hose is probly frozen anyway. So, I setup a place on my shop floor to do this water test.
I let the light heat up for about 15 minutes then kept spraying a steady stream of water from a spray bottle onto the light. I made sure that the light stayed very wet on all surfaces for about 10 minutes.
Rusults: The lens fogged up in one small area then quickly burned off. After this there was no sign of water getting in at all anywhere. So, I think Im going to go with this lens sealing design for my lights.
I dont think its going to get any better than this, especialy after reading the link NBeener posted on the ADV board which said even underwater lights dont stay perfectly dry all the time.
I have been thinking about powder coating the body of the light gloss black and leaving the screw-on lens cover with the polished look... but, on second thought that might just create a whole new batch of problems.
Novenber 29, 2008
Now, Im running into heat issues but I didn't discover this until after I had already started the machining process on a set of production lights. I imediately stoped machining and started looking for a solution to this latest gremlin. After running the lights in a dry state for about 15 minutes or so the internal wires were starting to melt, this is not a good thing.
I found a supplier that had high heat connectors so I purchased a handfull thinking this could be an easy fix. Turns out that the connectors are somewhat bulky and take up more space inside than I had room for. This is going to require re-machinging an area that I had already finished inside the body of the light. I did some measuring and Im not sure this will be possible since the light bodies are already machined to finished length. If re-machining works it will leave the back of the light housing thin which I don't know what the effects of this will be.
After I encountered the wire melting problem I put this project to the side and its still sitting on a corner of my work bench today.
TO BE CONTINUED.....................