2
\$\begingroup\$

I have been working on this personal project of water-proofing an electronic alarm clock, when I came across the idea of making electrical contacts non-conductive when produced. Specifically I have been thinking about the unique properties of passivation of aluminium. When exposed to air, solid aluminium reacts with oxygen to produce a thin layer of non-conductive (and quite inert) aluminium oxide on its surface.

So I thought it would be logical that if all exposing contact pins on electronic components are made from aluminium (or other metal/alloy that exhibits passivation), and all components are soldered to a PCB using aluminium, passivation will occur over all exposed metal surfaces of the PCB, so it would be intrinsically water-proof without additional treatment such as lamination (assuming the electronic components themselves are well insulated)(assuming no need for fingers or external connectors).

So my question is whether this is a practical thought, and if so why hasn't it seen much/any application?

\$\endgroup\$
4
  • \$\begingroup\$ In the presence of water, aluminium reacts to produce a porous layer of aluminium hydroxide, which is not effective for passivation. Along with galvanic corrosion, this leads to quite serious degradation of aluminium underwater, amounting to perhaps several millimetres/year depending on the water temperature, whether it is still or flowing, the presence of chloride ions, etc. \$\endgroup\$
    – Oleksandr R.
    Commented Jul 25, 2016 at 10:28
  • \$\begingroup\$ To start with, you never work with pure aluminum, it is always an alloy. Second, if pacification worked against water, aluminum wouldn't corrode underwater, which it does. Third, what about all of your component leads? Fourth, if you did with with pure aluminum, water would penetrate the oxide layer quickly and set it on fire. \$\endgroup\$
    – R Drast
    Commented Jul 25, 2016 at 10:29
  • \$\begingroup\$ @RDrast for practical purposes, most aluminium alloys are quite pure aluminium. And OP is not talking about burning aluminium in steam, as far as I can tell. \$\endgroup\$
    – Oleksandr R.
    Commented Jul 25, 2016 at 10:34
  • 1
    \$\begingroup\$ Plasma nitriding could produce a useful passivation, although you then have the problem of your device surviving treatment at high temperatures in a radio-frequency plasma. Most electronic devices cannot tolerate this kind of abuse. Be aware also that most epoxies are degraded by hydrolysis, so your PCBs and packages will also have to be made of some other material. On the whole, polyurethane coating seems to be the best approach for waterproofing short of placing the device inside a waterproof enclosure. \$\endgroup\$
    – Oleksandr R.
    Commented Jul 25, 2016 at 10:38

1 Answer 1

Reset to default
5
\$\begingroup\$

Ignoring the specific issues with aluminum there are production issues with the concept in general. If the components and PCB are covered with a non-conductive layer then how do you make electrical connections between them?

You need to first remove the insulating layer. Doing this mechanically would be tricky and risk damage to small parts. Doing it chemically would avoid mechanical problems but you need to find some chemical mixture which will dissolve the insulating layer but not the underlying conducting material, won't damage the parts and can easily be remove at the end of the process. Any problems with any point in this process would result in high resistances or mechanically weak joins.

To a certain extent the problem with having to remove a thin layer already exists in conventional manufacturing, there is a layer of dirt and oxidized material on all of the parts already. It may be conductive but it causes reliability problems with the joints. That is why you use flux when soldering, to chemically remove the dirt when making the joint.

For fairly clean things you can use a mild flux that will be inert fairly quickly (a no-clean flux), for dirtier items you need a more aggressive flux. If you fail to then clean any remaining flux off the board (which involves some nasty chemicals) this flux can eat away the joints over time.

There is also the metallurgy issue. The solder needs to melt at a low enough temperature that a brief exposure won't damage the components. The solder will then form an alloy with the metal on the PCB and the parts, the crystal structure of that alloy has to be compatible with the structure of the underlying metal or the joint will be weaker.

This is why 63/37 tin/lead solder was so good, the underlying metallurgy was as close to perfect as you could get. The current silver/copper/tin solders aren't too bad but they aren't as good. As soon as you add gold plating you start weakening the joints, the gold alloys that form are mechanically very poor.

Or the short answer: A method of making electrical connections that is reliable and suitable for mass production is a lot more complex than simply picking a metal to use. Rather than compromise the electrical characteristics for something that's waterproof it's far simpler to use a normal process and then add a layer of conformal coating if needed.

\$\endgroup\$
2
  • \$\begingroup\$ even conformal coating will allow some ingress of humidity over time \$\endgroup\$
    – Wesley Lee
    Commented Jul 25, 2016 at 11:21
  • \$\begingroup\$ Depends upon the material and thickness used. You could always pot it in epoxy which should be good against virtually anything and is in a sense just a very very thick type of conformal coating. \$\endgroup\$
    – Andrew
    Commented Jul 25, 2016 at 12:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.