Protection of electronics which was flooded with seawater

Community General Discussion
1

Hello!

I have a device (in a sealed aluminium enclosure) which is located in the sea, at depth of 7meters, connected by a cable to an another device which is located on the rock.

If/when UW device leaks saltwater this could be detected by the device located on the rock.

Question is: If we detect water in our enclosure, we

just power the device off, and report to the mainland or

we report to the mainland and use power to protect the schematics (at least for a few hours) - by connecting Vcc and GND of the device to "-" of the power supply while "+" of the powersupply is connected to an aluminum enclosure where all the electronics resides.

PCBs were dipped in some laquer before mounting, it resists rain but I doubt it will survive saltwater. This laquer is to protect copper/solder joints (seawater turns them into batteries and they corrode at speed of light)

Thanks!

2

Sorry, but you are not being clear. What is the question?

3

Thinking out loud, don't know what is possible within the project

  1. prevention, fill the aluminum enclosure with a non conductive gel (or whatever) so water cannot leak in.
  2. early warning system, make a 2 stage enclosure, the device is in the inner one, in the outer enclosure you have a leak detector
  3. self healing enclosure, there are bicycle tires that heal themselves when leaking. Don't now the details but it could possibly repair the leak of such enclosure too.
  4. inflate balloons when water is detected, so the device goes to the surface before getting all wet?
4

This part makes little sense to me.

A schematic is a diagram printed on paper or stored in a computer. You would not put that in the sea!

Did you mean to say circuit? How can power protect the circuit, exactly?

That sounds like it would cause a short-circuit which could damage the power supply. I would expect the metal enclosure to be connected to ground.

5

yes, sorry I mean circuit. PCB.

6

No ifs about it, it will leak. Aluminum is a very poor choice.
Nothing you can do but to cut power and that will only save the power supply.

7

This can be worked around on the power supply side. The idea is to use cathode protection (not sure if it is right word for that). From my experience if PCB was in conact with seawater - it gets damaged. If there were no power supply then copper dissolves slow, if there was any voltage - then in less than 1 min. I would like to protect PCB for a few hours so device could be recovered and fixed with little effort

8

Don't use aluminum.

9

You can't just consider corrosion, salt water is conductive. So how you manage undesired current paths on your electronics?

I would put my effort making watertight enclosure and pressure test it. If possible use suitable potting compound to protect it completely.

10

If you bothered to add a pressurized atmosphere inside the enclosure, the sea water will not get in. If your enclosure will not hold the gas, then your enclosure is not sealed. Plan ahead when designing the enclosure. You can use CO2 or nitrogen gas cartridge for pressure.

11

You seem to be assuming ground is the negative connection of the power supply, which does not have to be the case.

I think @vvb333007 is thinking of using the aluminium case as a sacrificial anode, which seems to me to be a reasonable thing to attempt, at least in theory. Whether it will work I have no idea. I worked in telecoms all my life, telephone exchanges had positive grounds, which protected the exchange earth connection. I used to know the chemistry but can't remember it now, maybe someone here knows and can explain it. Sea water itself has ions in it and will form a battery with all the different metals in a typical circuit. I guess the hope is to create a potential greater than can be created with the seawater battery. However, I agree that aluminium is a poor choice of container. If you don't need the metal screening use plastic. If you need the screening use plastic with an internal screen. Interested to know what others think.

@vvb333007 Build a test article with the positive of the supply connected to a piece of aluminium and another with it not connected. Put each in a salt water bath and see which one suffers the worst damage.

12

I would echo the comments about aluminium and seawater conductivity/corrosion.
Some professional experience of deep ocean water instrumentation and first choice is going to be marine grade stainless steel with accessories to IP68 minimum.
Heavy duty GRP possibly.

Every detail has to be right, seals cable glands etc.
Even on dry land with rain to deal with, if possible I would go for a two layer (enclosure within enclosure) approach.

Some sort of moisture indicator useful, or include a bag of self-indicating silica gel.
Blue when dry, pink when not dry.

7-metres depth represents a significant pressure.
Just for info, the real deep sea stuff had to be good for 4000-metres depth.

13

Your depth is 7 meters and figure that 10 meters of depth is about 1.0 atmosphere. The pressure is not that great.

Since you want a cable entering the fixture you will want waterproof fittings. You can roll your own or buy off the shelf waterproof bulkhead fittings. I suggest PAVE Bulkhead Fittings or just roll your own. The link provides an idea.

When choosing an enclosure you want waterproof you do not want water-resistant. Read up on IP Specifications.

Ron

14

Apart from whether the enclosure material itself is suitable, I would approach the problem from a different angle:

Why let the child fall into the well first and then think about how to rescue it? The goal should be to prevent seawater from entering in the first place.

At 7 m depth the pressure is only about 0.7 bar above atmospheric pressure. One option would be to maintain a slight overpressure inside the enclosure and monitor it with a pressure sensor.

If a leak develops, the overpressure will initially keep seawater out, at least for some time in the case of a moderate leak. The pressure sensor can trigger an alarm and the unit can then be recovered before any significant damage occurs.

In a more sophisticated setup, compressed air could even be supplied continuously from the shore station through a small tube.

I actually did something similar years ago with an underwater camera housing. It was connected to the surface with a standard polyurethane tube and a simple mechanical pressure gauge. My goal was merely to verify that the housing remained pressurized and that my expensive camera was safe for the few hours it was underwater. Looking back, it would have been trivial to add an air supply as well.

15

Wow, thanks for all the replies. I would try to run a test with a PCB and Al electrode to see who dissolves faster. Filling everything with a gel is also great idea.

16

Presumably you are using a 5000 series aluminum alloy enclosure so corrosion of the box is not an issue so your goal should be to find a way to connect the cable such that you get no leakage. A rigid stainless steel conduit might be a good choice but you must galvanically insulate it from the enclosure to prevent corrosion.

17

Don't neglect that there are dissolved salts in seawater. Sodium chloride will produce sodium at one electrode, chlorine gas at the other, in addition to hydrogen and oxygen at an ideal ratio to explode with any ignition source.

18

Just some thoughts from experience:

If you decide to encapsulate everything in gel or resin, there are a few practical issues you may have to deal with:

  1. Diagnostics and repairs can become difficult or impossible. Once everything is embedded, even simple measurements or inspections may no longer be feasible.

  2. Heat dissipation can become an issue. Components that run perfectly well in air may operate at significantly higher temperatures once encapsulated, depending on the material and geometry.

  3. Water migration inside cables is often overlooked. I am not referring to leaks at the cable entry into the enclosure, as those are usually obvious concerns. What tends to be underestimated is that water can travel inside the cable itself. A damaged outer jacket may allow water to move between the outer sheath and the individual insulated conductors. If an individual conductor insulation is also damaged, water can even migrate between the conductor strands and the insulation. Pressure and capillary action can transport water surprisingly far from the original point of entry.

Do whatever fits your application best, of course. I only mention these points because they are easy to underestimate until you have encountered them firsthand.

Personally, I have become a big fan of detecting a problem early and recovering the system before damage occurs, rather than relying solely on measures that try to limit the damage after water has already found its way in.

19

I forgot to mention, more recently we've had good success with plastic Swagelok bulkheads and tubing if your installation isn't subjected to any mechanical motion. Depending on tubing size, you can get a decent cable down into the enclosure and you don't have to worry about galvanic isolation.

You could even go all plastic if your application doesn't require any shielding.

20

Potting is the best solution, but I think you are looking for a conformal coating rather than potting. I would look at:

HumiSeal polyurethane coatings
MG Chemicals silicone or urethane coatings
Chase Corporation parylene systems

A few notes and observations:

Salt water will eventually wick down wire strands unless the cable entries are properly sealed. Take a look at the connector @EmilyJane suggested.

Connectors are usually the first thing to fail.

Exposed solder joints, especially lead-free solder, can corrode rapidly.

Even gold-plated connectors can fail if salt water becomes trapped inside them.

For a one-off project, I would usually pot the electronics rather than rely on conformal coating alone.

One of the biggest mistakes is mounting stainless-steel hardware directly to aluminum in salt water. The aluminum becomes the sacrificial metal and can corrode rapidly.

I would seriously consider a plastic enclosure. It may work better than aluminum in a salt-water environment, but only if the sealing and cable entries are done properly.

For salt-water use, look for an enclosure made from polycarbonate, ABS, PVC, or fiberglass with an IP67 or IP68 rating. IP68 is preferred if it will actually be submerged.

The weak points are usually not the enclosure itself. They are the lid gasket, screws, cable glands, connectors, and any location where water can wick through stranded wire.

For a submerged electronics enclosure, I would want:

A gasketed plastic enclosure
Plastic or nylon cable glands rated for submersion
Potted cable entries if it will remain underwater long term
No ordinary connectors below the water line
Electronics that are conformal coated, or better yet, potted
A small desiccant pack inside if the enclosure is sealed dry

For continuous salt-water submersion, I would not trust a typical hobby enclosure by itself. I would either use a proper IP68 enclosure or pot the electronics inside the enclosure.