Potting electronics with silicone

If you want to improve the reliability of your electronics, just cover them in silicone or epoxy. Potting, as this technique is called, protects your project from impact and loose wires, but most of the time ends up as a gloppy mess. [Charles] figured out an awesome way to make pro-looking potted electronics using silicone and a few supplies from the emergency room.

On one of [Charles]‘s trips to the ER, a nurse put Tegaderm on a wound and he noticed his skin could breathe. Figuring Tegaderm would also allow silicone to breathe, he asked for a sample and went back to the lab. After putting a few globs of silicone on a PCB, [Charles] wrapped the whole thing in Tegaderm. Much to his surprise, the silicone cured and made a smooth, glossy protective covering on his board.

[Charles] put up a Picassa album of his Tegaderm/silicone potted electronics, and we’ve got to say they’re looking very professional. This technique could also be applied to [Charles]‘ glass PCB touch sensor, allowing for the creation of clear (or at least tranlucent) PCBs.

After the break, you can check out [Charles] potting his ATtinyISP board in silicone with Tegaderm. One word of warning, though: don’t use a silicone that uses acetic acid to cure – that’s a death sentence to electronics. Luckily, there’s an easy test to see if your particular brand of silicone produces acetic acid while curing. Just squeeze a bit of silicone on a piece of scrap, and if it smells like vinegar after a while, don’t use it.

Comments

  1. sureshotstudio says:
  2. selim ozatici says:

    but why he use tegaderm, he can use hot glue or normal silicon without tegaderm.

  3. If anyone plans on doing this I highly recommend finding a non-acetic-acid cure chemistry (so NOT GE Silicone II) as these can etch some of your metal traces and other components as they cure. This is the primary difference between electronics grade and non-electronics-grade silicone: the curing process does not release vapors which can etch the metals in most electronics. This has caused actual problems when some manufacturers have tried to skimp on this.

    • C says:

      Can you just spraypaint the board before putting cheap silicon?

      • nes says:

        A cheap easy to get silicone caulk guaranteed not to give off acid while curing is the stuff sold for repairing and sealing lead roof flashing. It can be found in any building supplies store for the same cost as the regular bathroom sealant type stuff. It’s almost always medium grey and opaque though so doesn’t look as pretty.

        I have used it for potting electronics to go in car engine bays and it has survived years of heat, oil and weather just fine.

    • whoops, take that back, GE Silicone II is safe. Point is, check your label (or MSDS, but it’s harder to tell from that) and if it says it releases acetic acid, you’ve got a problem.

  4. truthspew says:

    The first widespread user of potted electronics in the United States was Western Electric. Find any old phone and look at the network riveted to the base. Everything in there is potted.

    Western Electric built gear to last for minimum 30 year lifespans and more commonly a lot longer. I have one of their pre-potted 302’s that dates to 1949 and is 63 years old and working fine. The potted phones I have are going on 40 years and still functional.

    Another thing Western Electric did to prevent corrosion was plate with cadmium. That’s why the bases on some of their phones have a yellowish hue to them.

  5. Charles says:

    About the whole acetic acid thing, I’m very sensitive to vinegar, and silicones that use vinegar will drive me out of the house, I was relieved when I found GE Silicone II. It uses ammonia as the main agent. Additionally, I’ve used stuff gooped up in GE Silicone II for years without problems.

    • charles says:

      “I’m very sensitive to vinegar”

      Am not. I love pickles!

    • Marko says:

      That’s ‘alodine’. Chromium-conversion coating (It makes stuff a little iridescent, varies from almost clear to near black depending on amount. goldish is pretty standard in commercial gear). It It can be done to Al, steel, zinc/cad plated steel, etc.

      So, I suppose it could be Cd plated and then alodined, but I’m not sure that is actually the case here.

      It was almost surely hexavalent chromium back then anyway, which is up there with Cd for toxicity anyway.

  6. Chris C. says:

    +1 respect to Brian Benchoff for knowing about and explaining the issue with acetic acid cure silicones!

  7. Oliver Heaviside says:

    Look up conformal coating – it isn’t expensive.

    If you have a reasonable supply of water, you can produce boards as good as any manufacturer who doesn’t use a clean room.

    After you make the board and shield the IO/Power connectors, you need to obtain some saponifier.

    These make water wetter and can be bought from many places in small quantities.

    It is amazingly close to a certain dish detergent that leaves things “virtually spotless”, and which works well enough for most purposes if you have access to dishwasher. In either case, run the boards through a dishwasher and dry them.

    Then bake them in an oven @200F for a while (think about your solder’s melting point and work backwards) – or set it to 125 or 150 and leave it for a few hours. The goal is to bake out any residual moisture.

    After that, purchase any conformal coating – it’s sold as such, and comes in a spray can. One or two coats and air dry is adequate. Wait a day or more if you can – everything out-gasses.

    Then – and only then – do you pot the product.

    For high vibration, you use epoxy, for high chance of corrosion/water/etc, use silicone. For inserting it into a body, medical grade silcone.
    You can even do both.

    A little foam or rubber around the edges helps with vibration isolation – much more than simply bolting it down. If you have flexible leads that come from the board, knot them and then mechanically secure them to the board with a screw or bracket or hold-down that doesn’t permit movement. This prevents water from sneaking in.

    Finally, you might bake the whole thing at a lower temp for a day or two. After that, if you want to shield for RF, thin copper (You can use steel or brass as well) foil origami with soldered seams will have you ready to take on sky net like a pro.

    Need to verify that you’re turning out decent electronic hardware?

    Pressure, temp and steam test – old fashioned pressure cooker. Good for a couple atmospheres.

    Vacuum/High altitude is harder, but you can get close with almost any vacuum pump and an old pressure cooker body. Or find a used commercial vacuum chamber. You can make one, and use multiple pumps to overcome your sloppy workmanship.

    Vibration: Anything that shakes. Car frames work well, especially if you drive off-road or your city is too poor to care. Sweet talking someone with a paint shaker [really excellent] is easier
    if you mount your device securely in a paint bucket and stay away from big chain stores.

    Be sure to check at several regimes – the hitachi giant wand/lawn mower deck/paint shaker trio will uncover almost anything.

    Finally, RF – go find someone with a 2 stroke dirt bike not made in japan and try to run your project board attached to the frame with the gnd lead attached to the frame. That will separate the men from the boys within two minutes.

    When your project survives all this, it will be suitable for almost anything from submarines to low earth orbit. A good final test is to let your dog/children play with it for a quick game of fetch/keep away.

    Or just try to demonstrate it to your bankers while you seek a business loan to make payroll for another 90 days. If it’s ever going to fail, it will fail then.

  8. Hackerspacer says:

    Why tegaderm? Not only is it an expensive material to work with, why not just use an addition curing silicone? It doesn’t need to breathe to cure and it doesn’t contain any acetic acid or offgas.

  9. Eddie says:

    How about using Hot Melt Glue?

  10. Natalie says:

    when potting assemblies, one needs to be really careful to properly strain relive everything. It will pull some parts off the board during thermal cycles.

    BTW, i disagree with Oliver Heaviside about conformal coating before potting. There can be potting adhesion issues between the conformal coat and the silicon. But i do agree with baking out your boards before potting.

    • Oliver Heaviside says:

      Quite right, btw.

      My apologies – certain kinds of potting should be done before the conformal coat, especially if you wish to have the coat adhere to the board itself.
      For example, when you’re doing the potting for design security purposes.

      I left out a step – after the dishwasher step and the baking, it should be epoxy potting, then conformal coating, secondary potting, then any rubberized or silicon environmental layer, then rf shielding and any vibration or heat damping materials.

      But really, unless you’re sending your arduino to mars, you don’t need that kind of stuff on most things – an altoids tin is fine. Just remember to flock it and glue some googly eyes on the case.

  11. ejonesss says:

    you can do some crude potting with hot glue.

  12. Galane says:

    Syd Levine who ran that Logwell site died a couple of years ago. He was a former New Left member who grew up, got smart and went into the oil business.

    After getting out of the drilling and pumping part of the oil business he went into building and repairing well logging equipment and other downhole gear like explosive perforators. He said his FBI file was quite thick, was quite a lengthy process to get cleared for explosives and radioactive materials handling since he’d been a member of the communist New Left group. (He said if there had been a revolution they’d have had to hold it after noon because none of them ever crawled out of bed till 12:00 PM.)

    I knew him by e-mail on a metal working e-mail list. We all missed him muchly when he died. :(

  13. Charles says:

    Not sure the best place to put this… but…

    1) This silicone does not outgas acetic acid (at least not much if at all). Most silicones I’ve dealt with that were purchased recently use other chemicals. I’ve been using this method for over a year and haven’t had ANY issues with the chips functioning.

    2) I do not use two-part silicone since I’ve tried this once with Q-sil 216. It didn’t work out well at all.

    3) Conformal coats are unnecessary. These parts have been thrown into laptop bags, dropped off everything and

    4) Using hot glue or gooping the silicone does not yield an attractive looking surface. The Tegaderm makes it smooth.

    5) Tegaderm is CHEAP! The patches are less than 60 cents each.

    • Oliver Heaviside says:

      As per item 3 – if you think a laptop bag or a drop of a few meters represents a harsh environment, you’re probably right about conformal coatings.

      But that’s like saying condoms prevent STDs, Deet can prevent Malaria, the fed can prevent an economic crash, or that a bullet proof vest will keep you alive.

      I assure you, counter examples abound. :)

  14. markS says:

    An exellent alternative to tegaderm is your finger dipped in detergent.
    Make sure to get it well coated in detergent before touching the silicone.
    You can then safely smoodge it around to get the shape you want.
    It won’t stick to you if you use detergent. You can dilute it in water a bit if you don’t want to use it neat. The final surface wil be slightly cloudy.

    The final surface will not be as perfect as tegaderm but its essentially free.

  15. Paddlebub says:

    Acetic cure silicone can affect soft lens LEDs by penetrating to LED covering…..even by just being in the same box for example sealing cable entry points.
    I imagine it would stuff up other component seals,eg caps, and really stuff up sensors.
    If you read the app note on LED manufacturers sites they should have a list of unacceptable chemicals.
    If you have a vac pump and scale to prep two part silicone carefully I much prefer that stuff myself.
    Silicone cables are contraindicated for submersible cabling, they use rubber instead. From what I’ve read silicone is more water permeable than you think.. Thick layers for anything permanently submerged. Some solvent like prep sol are bad. For fun try leaving a blob of nicely cured silicone in a puddle of prep sol for a couple of hours. Prep sol is great for stripping silicone clean off a surface.
    I’d be interested in people’s opinions on polyurethane for potting and sealing.
    Sorry no links, I’m tapping this out of rhe phone idly watching American Dad.

    • Charles says:

      I actually use a polyurethane conformal coat (after trying a few others) on the clear glass circuit boards. I’ve had such great success with spray polyurethane, since it goes on much smoother than with a brush, that I’ve basically left all brush-applied conformal coats behind. But that, like many two-part silicones (i.e. Q-sil 216, etc.) don’t “gel” to the degree you want to make a nice solid shell. They don’t “feel” as smooth and nice.

  16. steve says:

    Dow Corming sells silicone for potting electronics. Just apply with a brush and let dry. No tegaderm or dubious silicones from the hardware store involved.

    • Natalie says:

      agree, and the 3M products are great to use in a production product, Also check out Dolphs.
      But I think this article is more geared to the casual user that don’t have access to these industrial products easily or just wants to use what is flying around the shop.

    • Charles says:

      The conformal coats only coat the circuits, and don’t provide a solid grippy-feeling case. I’ve used Q-sil 216, and a few other proper coats, but I’ve never been able to get the two-part silicones to “gel” to the point needed to shape them into the neat grabby cases. I only used the Silicone II from the hardware store down the street as a last resort, only to find it was spectacular.

      If you know of another silicone that will quickly enter a pasty state, I’d be interested in trying it, since there’s a reasonable chance I could get it in _actually_ clear.

      RTV Silicone has such a bad wrap from the family of acetic acid based silicones. It’s frustrating to see people writing it off for such a small portion of the family of products.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 96,409 other followers