Building a waterproof light

16 posts / 0 new
Last post
devman
Offline
Last seen: 9 months 1 week ago
Joined: 04/27/2011 - 06:24
Posts: 825
Location: Hong Kong / Vancouver, Canada
Building a waterproof light

Between learning about TIR from match's mod (and others, but it's the one that sticks), how to pot the circuitry (*cough* match again) and reading up on IPX ratings, it seems like it should be fairly straightforward to build a cheap, waterproof flashlight with little or no internal air pockets that should be able to handle significant depths.  Has anyone tried this? 

scubaduderon
Offline
Last seen: 8 years 6 months ago
Joined: 06/25/2011 - 03:10
Posts: 108

It does make sense to minimize internal air pockets in my mind, and I even remember some talk of filling the entire dive light with a non conductive liquid to nullify all pressure effects, but my guess is that any stout lens and body would take the pressure folks normally dive at.

I'm not an expert, but I do own a number of the dive lights and I don't believe any were designed to minimize internal air volume. The thing they have that is different from normal lights is the switches do not penetrate the bod and multiple O ring seals. Normally they have only 2 parts, the screw on lens piece with an o ring seal at the mid point and one that the lens piece seals against. 

Ron   

devman
Offline
Last seen: 9 months 1 week ago
Joined: 04/27/2011 - 06:24
Posts: 825
Location: Hong Kong / Vancouver, Canada

I've seen a youtube video somewhere of someone filling their reflector with a non-conductive liquid (gel?), but since TIR's are a solid piece of either acrylic or polycarbonate, it seems like it would handle the pressure just as well, without the additional messiness.

I thought that in you can reduce the pressure differential by, say, eliminating as much of the easily compressible gasses as possible, then you would be able to get away with a simpler, build with less robust seals.  Plus, if all the electronics are sealed in block of epoxy, any water that did get in would be harmless.

scubaduderon
Offline
Last seen: 8 years 6 months ago
Joined: 06/25/2011 - 03:10
Posts: 108

Basically saltwater eats up any metal objects that have charge on them. They will literally disintegrate. Any amount of water rolling around the insides is bad. I have seen some folks insert absorbent pads into their lights in case of leaks. This trick works to varying degrees of success. Really, the leakproof multiple o ring seal is the key with a 2 piece outer body including the lens. They screw together near the center, and have 2 o rings there. One which is sandwitched between the 2 pieces, and one that is thrust by one piece into a seal in the other.

 

Ron

devman
Offline
Last seen: 9 months 1 week ago
Joined: 04/27/2011 - 06:24
Posts: 825
Location: Hong Kong / Vancouver, Canada

and there I go, skipping right past the whole 'hmm, what happens when I put metal in salt water without a sacrificial anode'

Problem is that, while budget lights DO often have good sets of o-rings, they're only usually single rings and only are usually 4 or more segments (bezel / head / body / tail)

It might be necessary to go along match's solid-copper-body route, only with a tool-dip rubberized coating and some more exotic switching.

gcbryan
Offline
Last seen: 11 hours 20 min ago
Joined: 05/07/2010 - 20:42
Posts: 2567
Location: Seattle,WA

It's not all that hard to make a light waterproof. If you want to make it into a convenient dive light it takes a little more doing. It's generally not worth the effort since purpose made dive lights are not that expensive especially lower powered ones.

Just for fun however I have tested and slightly modified a few lights and they have survived depths in the 100 fsw range. As mentioned many aluminum LED lights today have long threaded sections with one or two o-rings. The lens and the tail cap are the two main issues. The lens is generally not as thick as you would like to handle great depths but most will still survive 100 fsw if you don't bang them into something at depth.


Clicky tail caps don't work because they either leak and/or turn themselves off due to water pressure and can't be turned back on until surfacing. Forward clickies (if they don't leak) turn themselves on but can't be turned back off until surfacing.

I took a cheap ($20) Rominsen 100 lumen light and removed the switch in the tail cap which of course leaves a hole in the tail cap. I used the rubber switch cover and inverted it and used it as a plug and then covered that with marine grade epoxy. In effect I just made it a solid tail cap. I turned the light on and off by just twisting the tail cap to make or break an electrical connection.


This particular light had long tight threads and plenty of o-rings and didn't leak. For the lens area I had to remove the o-ring due to its design it was useless in this application but used a bit of marine grade silicone and I had no leaks.

To use this method you do have to pick the right light as a candidate. Some lights can't be turned on and off by backing off on the tail cap just a bit. Some have to almost be totally unscrewed to break the connection so these don't work.

The light I used wasn't a two part body however it actually had three areas that screwed together but they all had o-rings and long threaded sections and had a tight fit. I did coat them with silicone grease just for grins.

The front end of some lights are easier to seal than others. My light, with the cost of the epoxy and silicone, was probably between $25 to $30 however there is a very well built 3 AA XR-E 170 lumen backup dive light on DX for $36 (which I have as well).

It's nice to know that you can put together a waterproof light in a hurry if you need to but ready built is just as cheap and easier and more functional usually.

The other benefit of doing a bit of testing yourself is that you know where the failure points are, and what a light can and can't do. It's good for evaluating the claims of others including dive light manufacturers as well as how to take care of your own gear.

In many cases LED lights will survive the initial leaking even in salt water and will even continue running but do need to be cleaned up soon after getting out.

The electronics in cell phones don't survive salt water (since they are lower powered circuits and can't handle shorting out) but higher powered dive light circuit boards usually do have thicker tracing that do survive. I know this the hard way!

agenthex
agenthex's picture
Offline
Last seen: 7 years 1 month ago
Joined: 07/14/2010 - 04:49
Posts: 3070
Location: Merica

The lenser d14 frogman clone that pop lite does is only $30. I've seen crappier versions on ebay for 15. The design seems good as there's only one joint to leak out of (which you can just fill with o-rings) and a magnetic switch. Personally I wouldn't want to trust something I built myself down in the depths.

That w200 dive light on dx doesn't seem to get good reviews.

Reading this makes you smarter: http://lesswrong.com/

devman
Offline
Last seen: 9 months 1 week ago
Joined: 04/27/2011 - 06:24
Posts: 825
Location: Hong Kong / Vancouver, Canada

Thanks Bryan,

  It's more out of curiosity about solving the problems than actually building one at this point.  A ziplock baggie or a condom with the end tied off will more than handle any depths I can handle on lung power alone.  The whole thing started with a middle-of-the-night thought 'I wonder just how shockproof you can make a light' and ended up with 'how else could I abuse something like that?'

devman
Offline
Last seen: 9 months 1 week ago
Joined: 04/27/2011 - 06:24
Posts: 825
Location: Hong Kong / Vancouver, Canada

agenthex wrote:

The lenser d14 frogman clone that pop lite does is only $30. I've seen crappier versions on ebay for 15. The design seems good as there's only one joint to leak out of (which you can just fill with o-rings) and a magnetic switch. Personally I wouldn't want to trust something I built myself down in the depths.

That metal w200 dive light on dx doesn't seem to get good reviews.

I thought about magnetic switching, but all I could come up with is reed switches... and local suppliers only carry them for 0.5A max.

I suppose I could try to trace the driver and look up the chips to find something I could use that only runs on TTL levels.  Hmm...

agenthex
agenthex's picture
Offline
Last seen: 7 years 1 month ago
Joined: 07/14/2010 - 04:49
Posts: 3070
Location: Merica

Yeah, the frogman switch is supposedly only for 0.5A, too. You're supposed to use alkalines which limit current (along with the resistor they put in). The cheap chinese version supposedly has an even smaller resistor, and if you use nihm's... they're not very safety oriented over there, if it's not obvious enough from their lithium chargers.

Reading this makes you smarter: http://lesswrong.com/

Tas62
Offline
Last seen: 3 months 4 weeks ago
Joined: 12/26/2010 - 17:57
Posts: 494
Location: Tasmania, AUS

Supposedely waterproof to 100m.......with a tail clicky?

http://www.manafont.com/product_info.php/ultrafire-diver-light-ufdt1-q5-...

gcbryan
Offline
Last seen: 11 hours 20 min ago
Joined: 05/07/2010 - 20:42
Posts: 2567
Location: Seattle,WA

I think most people use a relay with the reed switch. I have a dive light that uses magnetic switching and I think it's done with a hall effect sensor.

gcbryan
Offline
Last seen: 11 hours 20 min ago
Joined: 05/07/2010 - 20:42
Posts: 2567
Location: Seattle,WA

I hadn't seen that one before. I see it has the all important crenelated bezel so important for a dive light Smile

The button must be magnetic?

bushytails
Offline
Last seen: 7 years 6 months ago
Joined: 06/11/2011 - 21:53
Posts: 167

I don't think minimizing air pockets in the light is beneficial...  any air at all is about the same as it being completely hollow.  By the time the light has decreased in volume even a tiny bit, it's because it's buckling, or the lens has cracked, or it's otherwise too late.

Rather than putting the reed switch in the main power, put it in the control line of a 7135-based regulator, or the mosfet gate on a buck regulator, or the enable pin of the boost controller, or similar.  Then it has to handle virtually no current, and the tiny draw of the light with the switch off is negligable.  Just solder it with some short wires to the driver board, in the space between the driver and the top of the pill (may not work with all lights), and invent some type of lanyard with a magnet and a rubber band, or something.  Or put a big magnet in the holster, so the light turns on as soon as it's removed, and turns off when put away.

There's some cheap hall-effect based lights these days, might be easier to start with one of those, assuming the selector ring doesn't penetrate the body.

For quick-and-dirty fixes, don't forget the miracle that is teflon tape on threads...

--Bushytails

agenthex
agenthex's picture
Offline
Last seen: 7 years 1 month ago
Joined: 07/14/2010 - 04:49
Posts: 3070
Location: Merica

http://www.priceangels.com/D5_Cree_Q5_LED_Diving_Flashlight__Black_1*186...

 

This looks pretty cool (magetic ring selector) if it works.

Reading this makes you smarter: http://lesswrong.com/

Budgeteer
Budgeteer's picture
Offline
Last seen: 7 years 2 months ago
Joined: 08/24/2010 - 02:19
Posts: 3087
Location: Slovenia (EU)

Or consider the XTAR D01 which is a bargain for what it provides. Not exactly budget priced but very budget for serious scuba gear. Probably one of the best in price/prefrormance with high quality manufacturing. You have to hold one to believe it, pictures does not do justice. See if you can handle one somewhere before attempting any "serious" purchase.

kragmutt wrote:

They're gonna send you a green redcat with a black LED.