0
\$\begingroup\$

Suppose a 4-conductor cable is connected to a printed circuit board mounted inside of a waterproof enclosure. The cable may become damaged (the insulation may be breached) allowing water to seep into the cable jacket and even into the individual conductors, which now behave as water pipes that bring the moisture right down to the board. This will happen regardless of any external waterproofing measures.

Assuming the impure water does not short any of the conductors, there is now a risk of corrosion at the solder connections on the board, as well as additional moisture damage to the rest of the board. Furthermore there is a risk that over a long period of time, this water may actually fill up the enclosure, destroying the board and everything connected to it.

The cable is epoxied into the enclosure containing the board so no water can come in around the outside of the cable, but there is no protection against water coming in through the inside of the cable and the individual conductors.

This can't be a new problem, there must be a practical solution out there for this already but I have yet to find it. Sure, I could apply epoxy around the individual conductors as well where they come into the enclosure but 1) that would be a manufacturing nightmare, and 2) there could still be air bubbles or delamination of the epoxy that would allow water through.

Does someone have any experience with this and have any suggestions for how to prevent water ingress from the inside of the cable?

\$\endgroup\$
  • 1
    \$\begingroup\$ Not sure you can without doing what you suggest and applying epoxy around the individual bare wires. The best way to ensure the enclosure stays waterproof is to use connectors for the cables rather than take them through a hole, and use connectors that are sealed in their unmated state as well as when mated. \$\endgroup\$ – Finbarr Apr 11 '17 at 13:24
  • \$\begingroup\$ Non-Wicking marine cable? \$\endgroup\$ – Tyler Apr 11 '17 at 13:37
  • \$\begingroup\$ If the cable are constructed with a single solid core instead of multi-strands and with PVC insulator, I don't suppose that water ingress would happen even in the case of insulator breach. \$\endgroup\$ – vrleboss Apr 11 '17 at 14:04
  • 1
    \$\begingroup\$ @vrleboss I considered using single-core wire, the problem is that the cable still needs to be flexible. Single-core wire would break after only a few bends, which is unacceptable in my application \$\endgroup\$ – DerStrom8 Apr 11 '17 at 14:28
  • \$\begingroup\$ You probably need a buffer. Have Pins or plates in the board that come through the epoxy. Then attach the wires to those. That at least will protect the board. Ultimately, having the right hermetically sealed connector protects your box the most, but really it just moves the problem to the plug outside the connector. However, assuming a damaged cable will be replaced at some point, the plug will be too. \$\endgroup\$ – Trevor_G Apr 11 '17 at 21:29
3
\$\begingroup\$

First: Epoxy-ing a cable through a hole is no guarantee for anything at all, as epoxy is rigid and doesn't form cross-bonds with almost any other plastic and plastics work with temperature just as any other material. (and also: moving the cable may just as well open cracks)

Second there's a reason IP67+ enclosures of professional grade only employ IP67+ connectors (fully sealed) and no direct pass-through of cables.

If you intend on using direct pass-through you're likely going to need to fill the enclosure with something flexible, like potting silicone.

You can attempt more aggressive glues like poly-acrylate on holes and cable-ends, but those are dirty stuff and even their cross-bonding depends on the types of plastic.

Truly the only "good" work you can do for high moisture resistance is actually getting all of the right materials as you'd find in industrial applications.

\$\endgroup\$
  • \$\begingroup\$ Good point about the potting material, I suppose I should have used a word other than epoxy. The actual material I use may not be actual epoxy, but a more flexible material such as Henkel Technomelt. The problem I am having with the connectors is that the space they are confined to is very small, and all IP67+ rated connectors I've found are much too wide. I may need to go the custom connector route but I was hoping for an existing solution \$\endgroup\$ – DerStrom8 Apr 11 '17 at 14:01
  • \$\begingroup\$ @DerStrom8 Yeah potting material is the way to go if you can't find any good connector (what kind of space are you looking at? is a M12 connector too large already?). Another thing, could you employ a mechanism to detect a broken cable to give of an alarm? \$\endgroup\$ – Arsenal Apr 11 '17 at 14:38
  • \$\begingroup\$ The hole the connector needs to fit in is 10.8mm in diameter, so yes, an M12 connector is too wide. Additionally, length will be a problem, the entire mated assembly can only be approximately 19mm long, and that's pushing it. I actually inherited the initial design and I am trying to fix some problems with it, so that is why I have all of these constraints \$\endgroup\$ – DerStrom8 Apr 11 '17 at 14:41
  • 1
    \$\begingroup\$ @DerStrom8 There's IP68 connectors up to 6 pin circular, comparable to mini-DIN in size, male flange mount about 8mm portruding inside, hole from memory 9.5mm. Name escapes me, but there were plenty to choose from at Mouser/DigiKey when I was designing Smart Motorcycle Suitcase lighting. As you can imagine you don't want a connector sticking in 2cm either when your case exactly fits a helmet without it. \$\endgroup\$ – Asmyldof Apr 11 '17 at 15:06
  • \$\begingroup\$ That sounds very promising, though I have not been able to find one with those dimensions as of yet. I found connectors that fit your description with the exception of the dimensions... \$\endgroup\$ – DerStrom8 Apr 11 '17 at 15:15
-1
\$\begingroup\$

You use what is known as a cable gland of the proper size, and make sure it is complete with all o-rings and mounted tightly.

Phoenix makes them, and there are scores of other manufacturers of IP67 rated cable glands.

Example: https://www.phoenixcontact.com/online/portal/us?1dmy&urile=wcm:path:/usen/web/main/products/subcategory_pages/cable_glands_p-09-08/aaccc963-bd36-45bd-8e76-851f1264a585

\$\endgroup\$
  • \$\begingroup\$ Unfortunately cable glands do not prevent water ingress from the inside of the cable, only the outside. \$\endgroup\$ – DerStrom8 Apr 11 '17 at 13:58
  • 1
    \$\begingroup\$ Water shouldn't be inside the cable. If you have a problem with that, then you should be using a filled strand/filled packer type wire and cable construction which inherently blocks water transmission through it. \$\endgroup\$ – R Drast Apr 11 '17 at 14:19
  • \$\begingroup\$ The cable type is fixed, I do not have the authority to change it. That is why I'm looking for a solution at the connection level. \$\endgroup\$ – DerStrom8 Apr 11 '17 at 14:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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