Filling standard torches with mineral oil for diving?

Anyone seen this:

http://blog.makezine.com/archive/2010/09/diy_5_led_dive_light.html

Any obvious problems, other than having to refil with oil each time you want to charge the batteries?

I wonder if refraction would make it all floody?

If water can leak in to a standard light during a dive then oil could leak out while its sitting in your dive bag/car/table etc
Yes refraction will be a problem too.

Question you need to ask is...
What standard type of light is worth going to this trouble and what advantages does it have over a proper dive light.

Packhorse said:

If water can leak in to a standard light during a dive then oil could leak out while its sitting in your dive bag/car/table etc
Yes refraction will be a problem too.

Question you need to ask is...
What standard type of light is worth going to this trouble and what advantages does it have over a proper dive light.

Click to expand...

The torch would not be under pressure when above water though, and it's the pressure which makes the rudimentary seals on most torches fail under water isn't it?

What effect do you think the refraction would have?

I'm not going to be doing it myself, I've got proper dive lights, just wondering about the theory.

The advantage is it would be a lot cheaper to buy something like a standard P7 light or thrower like this http://www.dealextreme.com/details.dx/sku.3356 and do this to it (IF IT WORKS) than a proper dive torch, or even something like the ugly light.

The Magpie said:

The torch would not be under pressure when above water though, and it's the pressure which makes the rudimentary seals on most torches fail under water isn't it?

What effect do you think the refraction would have?

I'm not going to be doing it myself, I've got proper dive lights, just wondering about the theory.

The advantage is it would be a lot cheaper to buy something like a standard P7 light or thrower like this http://www.dealextreme.com/details.dx/sku.3356 and do this to it (IF IT WORKS) than a proper dive torch, or even something like the ugly light.

Click to expand...

Be easier to wrap the damn thing in a ziploc bag and cover it in duct tape...
Then you could buy about 10 of those and have the ultimate "redneck" dive light for about $350...

No, the pressure does not cause the O rings to leak. In fact in many cases its the opposite. The pressure helps the O rings seal.
Oil filling is used to stop materials deforming under pressure. You might do this to a dive computer to stop extreme pressures from crushing the case.

While it's true that pressure would help the O-rings to seal, you're assuming the light is equipped with O-rings and not just deformable plastic lips on mating surfaces. Also, deformation under pressure could very easily open tiny gaps in surfaces that appear to be sealed properly under normal conditions, so preventing deformation would definitely help ensure a good seal. And if nothing else, the oil will occupy space so water can't get in even if there is a pinhole leak somewhere.

I see no reason not to give it a try. Assemble a light inside an aquarium filled with mineral oil, then tie it to a fishing line and cast it into the water, possibly with some ballast to help it sink. Wait a while, then reel it back in, dry it off, and open it inside the same aquarium and look for water droplets in the oil.

As for refraction making the light floody, it will certainly change the focus of the beam, but it shouldn't turn it into a floodlight.

Interesting idea, using the concept that liquid is essentially incompressible, to oppose positive pressure.

You will have water under pressure outside, acting against liquid at 1ATM inside. I'm only surmising but I imagine the oil will not be able to compress enough to damage the torch (e.g front lens).

What it would also be good for is offsetting buoyancy. Most people's preference is for a negative buoyant torch and reducing air pockets is a good thing.

Depending on how you work the battery recharge contacts it could be messy, or fine. Depends.

With the oil in the front of the torch messing with the beam quality, again I don't know from experience but surely it will. Micro bubbles, opacity, surely must degrade the light. Again, happy to be corrected.

You could make an internal waterproof bulkead for the front optics and fill the rest of the torch up.

I'd be interested to know how it goes.

Interesting comments, thanks chaps.

Of course the other thing it could help with is switching as I always found that to be a bit of a stumbling block, but there'd be no need for fancy piezos or reeds with relays. A normal toggle could be sealed well enough to stop the oil coming out even without a rubber jobby.

Any water resistant flash light will work this way. It is not the o-rings that fail. As Pack stated it is deformation that causes the leaks. Most of the cheap lights have very thin lens that we change out. They implode under pressure. Oil filled light will not require them to be changed. Regardless of the exterior pressure the lens will not deform as there is no place for it to flex to.

If four atmospheres of pressure is present outside the lens, 4 atmospheres of pressure will be present inside the case. As long as there is no air inside. The external pressure against the lens will apply the force to the oil inside. The oil inside is non compressible and electrically non-conductive. It will conduct heat and pressure but will not short out the electronics.

Loss in lumen output will depend upon the Mineral oil used and is equal to the distance from the emitter to the lens. Much like the loss from the lens to point x in the water.

It is a mess to use and is really best suited for the umbilical lights where you are not opening and replacing batteries. The Mag conversions are a prime example. By filling with Mineral oil and sealing it up, there is no need for custom tail caps, thicker lenses, or thicker o-rings. All the factory pieces will work. As the oil heats up it will transfer the heat to the case and then out to the water so a super tight fitting heat sink becomes less critical.

idk about oil, but glycerin should work, my ac gauges filled with glycerin about 4\5 of volume, it is pretty clear liquid, you'll just have to find a way to get ALL air out.

Hey, interesting topic. Strangely this was the subject of my (unfinished) masters thesis from many years ago. Just in the last few months I've come back to the idea and have been working on proving the concept in a dive light application.

Bottom line, using a fluid filled body will eliminate the possibility of water intrusion due to pressure effects. It's just basic physics. Bell & Ross (a high end watch maker) uses this in a watch that is pressure tested to something like 10,000 FSW. A Japanese submersible manufacturer that does super deep exploration uses liquid filled batteries for the same purpose...pressure resistance. Some fiber optic manufacturers also used oil filled lines to prevent deformation on ocean floor cables.

Since flooding is the #1 source of dive light failure, eliminating that possibility seems like a good idea, to me at least Any light with proper seals should hold the fluid in without any issues...just like proper seals keep it out. Leakage would be the result of manufacturing, design, or user error. However, say you get some debris on the o ring and go for a dive. Also assume the debris would cause a flood in an air filled light. If it was filled with fluid, it still wouldn't flood because there is no place for the water to go. Pretty cool if you ask me.

As far as fluids, mineral oil is a common dielectric used for insulating and cooling high voltage transformers and switch gear. Silicone oil is another possibility. There are also a host of dielectric fluids used in liquid PC cooling. Finally, there is a proprietary fluid called Midel that is designed to be a stable dielectric insulator.

The biggest challenge in fluid filled dive lights is the design, as opposed to the physics. Hopefully I'll be working some of this stuff out over the next couple of months. I'm going to be using PC cooling fluid to start testing with but I don't think it will be chemically stable enough over the long term. My first test will be to determine the pressure generated by heating the fluid in a sealed environment and use this to determine how important additional pressure compensation is. I'll keep ya posted!

Hall,

Your idea that pressure is transferred through the body of the torch to equalise inside assumes fluid is incompressible, which it is not. You can compress liquids...granted, not by very much, but they do compress.

If the inside of the torch is at 1Atm when you descend then it can only match the outside pressure if it is compresses to the same pressure. For that to occur the lens/body of the torch has to give. It is likely that as the body of the torch will be metal the give for the whole surface area of the torch will be concentrated in the centre of the lens.

Look, to be fair I don't know how much compression has to occur in mineral oil to equalise 1Atm to 4Atm, and how much lens flex needs to occur for that to happen but I imagine it would be so small to be insignificant...but it will be there. Unless you plan to use wafer-thin brittle glass I'm thinking it won't be an issue.

BTW, great idea about cutting Lexan on the other thread, with the washers and finger guide.:thumbsup:

Another thing to consider when attempting to fill a light with fluid is that when you screw down the tailcap, lens cap, or whatever the final seal is, you will be trying to compress the fluid in the light. Once the seal is made, you have another thread or two to turn, and something's gotta give!

D

Yeah the volume displaced during closing is a tricky one and I can see that being an obstacle to filling a "standard" flashlight with fluid. The guy in the video mentioned pushing down the switch boot, which will help, but it doesn't seem like that would displace enough volume. I guess it's just a math problem.

Hrm, now that I think about it, maybe it would be best to secure the tailcap last since the O ring tends to seal very near the point where the threads would be fully closed.

I tried this with some mass market Xenon light I bought at the hardware store. It worked fine but I found it almost impossible to get all the air out because of the way it was assembled. Might be important to choose the "right" light in that sense.

You could drill and tap for a tiny screw, in the best spot on the torch for displacing air.

So, fill and close all the larger normal seals, then turn the torch to the best angle to allow all the air to displace, rising against gravity. Then top up with mineral oil with a syringe, and screw in the final seal.

Main problem with oil is that is heat isolator and it is drity.

It stores heat even if you have good for example good conduction cooling direct to enclosure .

But in other way you have to tight enclosure two times from water and from oil ... (thay may not get in contact !) that complicate light too much.

But in othercase there is many projest in that way like joints in http://www.sub-find.com/newt_suit.htm

Klem, I like the screw idea, very little final displacement.

Lucca, I don't know the numbers for specific heat capacity off hand but wouldn't oil be a better conductor of heat than air that is typically inside the housing? Still you are right, oil is not a good conductor either.

Who is going to take the first leap and fill a light and post it?

I won't :devil: and is toxic

thermal conuctivity (W/m/C)

air 0,026

transformer oil is 0,130

water 0,588

aluminium 198

cooper 388

from http://www.pwr-tools.com/

Wow I'm surprised that oil is actually pretty close to aluminum...and at least better than air. Now if we could just fill the light with copper.

Anyway, it seems like oil might actually enhance heat transfer to an acceptable level, especially assuming the bulk of the heat sink is aluminum.

archer6817j said:

Wow I'm surprised that oil is actually pretty close to aluminum...and at least better than air.

Click to expand...

where do you get that? oil is 0.130 while aluminum is 198 and water is 0.588

filling it with liquid copper will create a massive short circuit and likely melt the soldering anyway. and you can forget using many plastics