Underwater light head question.

I'm trying to design an underwater light with a separate light head and battery canister.

I've read the adventures of DM51 with his "moderator" and I'm trying to avoid a similar situation.

After reading about his experience, I'm of the opinion that the main problem wasn't the dissimilar metals per se, but rather that the electrical circuit was using the metal flashlight body as a conductor to ground (this might have been concluded in his thread, I apologize if it was but I skimmed over some of it).

The drop in emitter I bought grounds to the metal piece it is mounted in, and if this was mounted in a metal light head body, the whole body becomes part of the circuit, resulting in all kind of galvanic corrosion and/or deposition. This happens on a micro scale due to slight variations in metal content and is what leads to pitting corrosion in pipes and such (so it is dissimilar metals in this sense). This is not a problem in flashlights not submerged, but in underwater flashlights.

So, my question is, do you agree? and

Should I somehow isolate the light head body from the circuit?

This could be accomplished by using a plastic body, but I loose a lot of the cooling effect of the metal body.

I do not believe putting some kind of coating on the body would be effective as any slight chip would result in contact and corrosion.

I am leaning toward a design that has a plate that the emitters are mounted in to act as a heat sink. The plate would then be mounted in the metal light housing but electrically isolated from it. The down side would be there would be no direct metal thermal path for the heat to the light head body. I'm not sure if they make a material that is a good thermal conductor and poor electrical conductor.

So, waddya think?

Please hurry, because I have a buddy at a machine shop ready to make me whatever I send him for a design! :thumbsup:

All of mine are isolated, all are aluminum. Just a side note I had the exteriors anodized as the salt water was tearing them up.

Just isolate the circuit from the heat sink and make the heat sink as snug a fit to the body.

If I remember correctly you were using R2's that are made for a drop in, is that correct?

R2's is correct. I ordered one to test out how bright it is and it seemed to pretty bright.

Do you think 7 of these emitters is overkill?

Part of me think yes, but there is this nagging voice that keeps saying, MORE POWER IS BETTER!!!!

When you are using dis-similar metals underwater with electricity running through them, one will oxidize and the other will be reduced or use the energy to reduce something.

You can use a thin layer of something(Mica? Tape? Hair?) to isolate the metal plate and thermal epoxy/glue to ensure a good thermal path between the metal plate and the body.

For this to happen wouldnt you will have to have both a + and - exposed to the water??


Unless you have two diffrent metals exposed that are connected. Then they make their own battery and you get DMGC.
I have this on my maglite conversion where the nickle plated gland contacts the aluminium body.
The conversion is 2 + 200 dives old and the corrosion is very slight and has not caused any problems.

You can fairly easily isolate the metal base of a P60 dropin. You just need to desolder the board from the brass and epoxy the board back in. Then you solder to the ring on the board for negative.

I hadn't though of that Luke, pretty good idea if you are good at soldering! I'm afraid I'm all thumbs and would be afraid to smoke the board.

Right now I leaning toward mounting the drop ins in a copper plate, which is mounted and electrically isolated inside an aluminum light head. Then I can use some some globs of arctic alumina epoxy around the perimeter of the plate to join the plate to the body thermally but not electrically.

Surface Tension said:

I hadn't though of that Luke, pretty good idea if you are good at soldering! I'm afraid I'm all thumbs and would be afraid to smoke the board.

Right now I leaning toward mounting the drop ins in a copper plate, which is mounted and electrically isolated inside an aluminum light head. Then I can use some some globs of arctic alumina epoxy around the perimeter of the plate to join the plate to the body thermally but not electrically.

Click to expand...

I bet you could do it. DX sells desoldering braid (they call it desoldering wire). You just flux the braid, melt the solder on the board and put the braid in the solder letting the solder flow into the braid.

Really, you don't even need to do that. You can just push the solder around with the hot iron until it forms blobs on either side of the gap that the solder bridges.

Or you can use a file and just file away the solder until there's no visible metal between the brass and the board.

The threading on all the 1-mode 3-18V DX modules is M20x1. You can find that size tap at use-enco.com. (I use that piece of info a lot more than I ever thought I would.)

Big globs of AA aren't really that good at conducting heat. I really suggest at least filing the dropins to electrically isolate them.

That probably would be an easier way to go. I know the braid you are talking about.

I could test them with a multimeter to make sure there is no conductance between housing and negative pole right?

Yes, If there is no connection, you should be good to go.

Packhorse said:

For this to happen wouldn't you will have to have both a + and - exposed to the water?

Click to expand...

Yes

Packhorse said:

For this to happen wouldnt you will have to have both a + and - exposed to the water??

Click to expand...

Really only a +. The water acts as ground.

LukeA said:

Really only a +. The water acts as ground.

Click to expand...

No you need both + and - to complete the circuit.

Packhorse said:

No you need both + and - to complete the circuit.

Click to expand...

Um no. All you need is a difference in voltage potential between the earth and your source and a reasonable connection between the two and the earth will act as ground. This concept has been in use for nearly two centuries.

LukeA said:

Um no. All you need is a difference in voltage potential between the earth and your source and a reasonable connection between the two and the earth will act as ground. This concept has been in use for nearly two centuries.

Click to expand...

Isnt that what I said?

voltage potential between the earth and your source

Click to expand...

=a + and a -.

reasonable connection between the two

Click to expand...

= complete the circuit.

What I understood from when you said

Really only a +. The water acts as ground.

Click to expand...

is that if the case is positively charged then the water will conduct the current to ground (as in the earth as opposed to negitive or-) which it cant possibly do unless your light is plugged into the mains.

Packhorse said:

Isnt that what I said?

Click to expand...

No. You said that one would have to provide both a positive and a negative. You only need a source of voltage potential. The water will absorb the difference and attempt to make the potential zero.

Packhorse said:

What I understood from when you said
is that if the case is positively charged then the water will conduct the current to ground (as in the earth as opposed to negitive or-) which it cant possibly do unless your light is plugged into the mains.

Click to expand...

Not quite. If you connect hot mains line and ground, you get current flow. Ground from mains leads to the same voltage sink as the water does, the earth itself. The case of the light should be ground (circuit -) if it isn't electrically isolated so that it has the same voltage potential as the water. Also, the case should be made of one type of metal so it doesn't go galvanic, and sacrificial zinc (It's zinc, right?) never hurt.

Talking in circles here, but a voltage potential will be a + and -.
If you have 2 metal pieces with a slight resistance beween them and they are part of a circuit then there will be a voltage potential accross them. One being + and the other being - inregards to each other.
We are saying the same thing. I just think some may have misinterpreted your quote of

Really only a +. The water acts as ground.

Click to expand...

(The water must be grounded to the - some where right?)

I think the best solution is to just make sure the body is electriclly isolated from the power circuit. And as you say use only one type of metal for the body.
I use a maglite with a nickle plated cable gland and after over 2 years and 200 salt water dives it shows slight signs of galvanic corossion on the aluminium near the gland. I could have gone with plastic glands.

Yes it is Zinc.