Why do electronics not like water? Technical explanations from an armchair...

[Beamin]

This is just common sense don't go swimming with your phone, water and electricity don't mix. But instead lets think about why they don't. I watched a video where a guy soaks a Nintendo board in his sink then dries it off and the board was fine. The comments on youtube went crazy and I realized the armchair chemists have no idea what they are talking about or have a clue other then what they heard from others using "commonsense" which can come back to bite you since things are always more complex then one would think. Right for the wrong reason...

So what do we know about water: Water is a nonpolar solvent not a conductor, BUT as soon as it gets a small amount of salts (ionic solids like NaCl or other salts or oxides of metals) it becomes more conductive even at low voltages. So if the board is off it should be fine? If you think about it everything is wrapped up in plastic, a chip should work under pure water they are sealed pretty well. But another thing I have learned is that some of the plastics, I'm not sure which ones, but I suspect the plastic that chips come in (thermostatic plastic like an epoxy, only dissolves in nitric acid and sometimes HCl) if you have other things present and a lot of time, will actually absorb a small amount of water, it's not hydrophobic it's slightly hydrophilic. If it has salts in the solution it can condense inside the chip and cause corrosion, there is no air so dissolved oxygen would have to be present.

Another way things can "oxidse" in the common parlance, is replacement reactions. This is how I can etch with HCl and hydrogen peroxide from the store. HCl won't dissolve copper, but it does when you dump it on a circuit board. I figured this out when putting old circuit boards under conc. HCl and 2% H2O2 for a week. I found the whole board turned into a bare pcb and sand like material with the covers of the electrolytics floating on top, even some plastic was dissolved! This shouldn't happen but clearly it does.

The first clue is indicated by how clear HCl turns yellow, then green, back to yellow, clear, and eventually black/gray. The HCl will attack any iron on the board. This makes ferric chloride, not sure which oxidation state though. Ferric chloride is less stable; has more energy then copper chloride. So when the ferric chloride comes in contact with the copper chloride the iron is replaced with the copper etching it away. The HCl and H2O2 participate in helping with secondary reactions, lowing the pH and creating intermediate short lived reactions. Kind of like how Aqua Regia which is HCl and nitric acid dissolve gold but neither should actually do that; one acid forms a layer one atom think on the gold then the nitric reacts with that layer, then the nitric acid reaction stops then the HCl comes in, repeats.

What does all this have to do with water in electronics? I think this is what happens when you get electronics wet. Each part is technically water proof, wrapped in plastic. Pure water, even tap water, shouldn't conduct at 3 or 5 volts. But it must dissolve salts and oxides on the board and short out sensitive chips. So as long as things are powered down you can get things wet. I would think if you dropped your electronics in water you should remove the battery (This can be done quickly and without special really hard to find tools on most newer iphones  ) before the salts dissolve and it shorts it out, as quickly as possible dunk the whole thing in 91% rubbing alcohol (you can make 98% using salt but I'm not sure if the salt will be in the 2% solution depends if the water is highly dispersed, solutions aren't little droplets, that's a suspension, and still able to dissolve it, I'll try this out next time). Don't use 70% you are just diluting the water or possibly adding more water in a more soluble form. Then dunk it again in a fresh solution one or two more times. This will get the water into the alcohol the drier (higher %) the alcohol the better.  Most of the salts won't dissolve in the alcohol, but will be carried out with each step so more water at first is a good thing, and while IPA is a polar solvent, like water, it doesn't conduct well or possibly at all, it's more like an oil then water.

But this is all speculation, what really happens? On time scale of minutes or even hours there is not time for corrosion. Low voltage doesn't conduct well through ordinary water. We could test this by building a few identical circuits each with different voltages and complexity, get them wet, then dry it out and test each part. Water proofing might be really simple if you find you don't have conformably coat a board but rather just one part or a sensor that shuts it down when humidity or water is detected. Not practical in cheap electronics but in military or medical devices this could reduce the cost of water proofing.

Has anyone tried this out?

[GeoffreyF]

I think it would be good to do a baseline in distilled water.

Functioning in water is a different category.  Electrolytic things can happen or ... as much in distilled?

Where there is a fail, what is dissolved or suspended in the water is part of a useful result. 

Finally, what is done to recover the thing after submersion?  How did that work out?

[coppercone2]

if you do an experiment you need a conductivity meter to look at how the distilled water is changing in conductivity over time. It's pretty bizzare. 

[@rt]

What is it exactly you are asking of someone has done? The test through different voltage PCBs?

The production company I worked for cleaned some boards in ultrasonic bath, so there’s some
I believe this was a mixture of demineralised water and IPA or something like that.

There are moisture sensitive devices that I suppose should not be subjected to humidity at any time,
and also some parts like buzzers that shouldn’t be wet because they retain water, not to mention transformers and inductors.
In the old days if a former got wet, the lacquer ate the copper, and it’s death was a given.

It is totally commonplace in the retro communities to give retro computer PCBs a bath, sometimes even in dishwashers.

[Cyberdragon]

The conductivity of water entiretly depends not only on what's absorbed into it, but what's going to get absorbed by contact and the distance between electrodes. Go watch Louis Rossmann's videos of liquids on a circuit.

[filssavi]

Unfortunately there is huge confusion on the matter...

H2O is actually quite a good insulator, it is in fact used in high voltage pulse generator capacitors as is self healing

Now water is the common name improperly used for the mixture of H2O and dissolved salts that can be found in nature, and because of the impurities it is a good conductor of electricity, do obviously it forms unwanted connections between nodes that shouldn’t be connected

Moisture sensitivity is a completely different phenomenon, in most of the cases (like 99.99%) the problem is not the silicon die, which couldn’t care less about the water, but it is the plastic/epoxy of the package that absorbs moisture. This by itself is not a big problem, the device functionality is unchanged, however when it is reflowed in a oven the water suddenly boils, making the whole IC explode like a popcorn.
In the end if electronics is washed by water nothing should happen, if the board if properly dried before use