Connect with us

Arcing in Thermal Adhesive

Discussion in 'Electronic Design' started by D from BC, Mar 5, 2007.

Scroll to continue with content
  1. D from BC

    D from BC Guest

    I'm gluing a part to a heat sink with a non-conductive thermal
    adhesive.
    After cure, testing shows no conduction between the part and the heat
    sink. Good..

    Problem: After power up, a short occurs between the part and the
    aluminum heat sink. After power down, the part is still shorted to the
    heatsink!! Bizarre..
    There is no testing mistake.
    (200VDC exists between part and heat sink during normal operation.)

    What's going on?
    Is it possible that an small arc develops in the adhesive . The arc
    vaporizes some aluminum. Then when the power is cut off the aluminum
    vapor cools and develops a conductive bridge (a short).
    Possible?
    Or maybe the arc burns the adhesive and the burnt adhesive is
    conductive..

    Anybody encounter something like this?
    The shorting effect doesn't happen when I increase the bond line.
    However, I'm trying to keep my bond line to a minimum.
    D from BC
     
  2. John Larkin

    John Larkin Guest

    Possible. What's the resistance?
    What's a bond line?

    The glued parts have to be flat, with no burrs or such. Spacers can
    define the epoxy gap; small mylar discs (from a paper punch), a bit of
    monofilament, or a tiny amount of sand or Cataphote bead filler mixed
    into the epoxy.

    Even better, hard anodize the heat sink first.

    John
     
  3. D from BC

    D from BC Guest

    The short is below 1ohm. I just can't imagine arc fried non-conducting
    adhesive getting that conductive.

    Bond line: thickness of the adhesive

    I'm just temporarily stuck with thermal adhesive to do prototype
    testing..
    A hard anodized Al heat sink is going to be used when available.
    D from BC
     
  4. Phil Allison

    Phil Allison Guest

    "D from BC"

    ** The metal part is lightly touching the aluminium heatsink at some point.

    No conduction occurs at low voltage due the oxide layer always formed on
    surface of aluminium.

    All sane folk use mica or other insulators for voltages like 200 volts.

    Even "hard anodising" will not be reliable at 200 volts.




    ........ Phil
     
  5. Eeyore

    Eeyore Guest

    Not really.

    Your thin layer of adhesive can't insulate 200V is what's going on ! You need a
    physical mechanical re-inforcing barrier. A very thin piece of paper would
    likely do it.

    Graham
     
  6. James Arthur

    James Arthur Guest

    Sure, why not? Type II anodize isn't quite as good, but is easily
    done at home.

    James Arthur
     
  7. D from BC

    D from BC Guest


    I can understand the Al oxide layer breaks down and an arc occurs at
    the closest point..

    But after power down, a new short still persists.
    I don't know the physics of very small spark gaps..
    Does the spark weld the part to the Al heat sink?
    Or maybe there's some sort of metal vapor deposit on the thermal
    adhesive after the power is removed.

    What could be bad about hard anodizing? If the oxide layer is thick
    ,it should be ok. Long ago I think I saw a microscopic picture of the
    cross section of anodized aluminum.. The cells had random anodization
    depths.. Is that the problem?
    D from BC
     
  8. D from BC

    D from BC Guest

    Type II?? Whats that?

    I was thinking of anodizing in sulfuric acid (battery acid
    concentration).
    D from BC
     
  9. John Larkin

    John Larkin Guest

    1.5 mil hard anodize is reliable at 200 volts with just silicone
    grease, provided the surfaces are smooth/flat and it's a serious hard
    anodize. With epoxy, it should be even better.

    Test hard anodize by poking it with a pair of sharp ohmmeter probes.
    You should read infinity until you start seriously pushing and digging
    into the surface.

    John
     
  10. D from BC

    D from BC Guest

    On Mon, 05 Mar 2007 17:09:24 -0800, John Larkin

    [snip]
    I was thinking of making a simple capacitance meter cct to test for a
    target anodization thickness.
    I haven't figured a test frequency yet... 1khz, 10khz, 100khz..??
    That'll be a new post if I get stuck.
    D from BC
     
  11. Jim Yanik

    Jim Yanik Guest

    maybe you're getting an arc thru a small VOID in the adhesive?
    Then it "carbonizes" the surrounding adhesive,making it conductive,or the
    arc vaporizes aluminum and plates it's own short.
     
  12. James Arthur

    James Arthur Guest

    It seems to work well in electrolytic capacitors...

    James Arthur
     
  13. James Arthur

    James Arthur Guest

    The three most popular aluminum anodizing processes produce coatings
    called Type I,II, and III. The difference is in process variables,
    particularly the process temperature and current density applied to
    the work. Type III requires cooling, is the densest and hardest, and
    is what most people mean by "hard anodize." Type II can be done at
    room temperatures, is pretty decent, and is called "hard anodizing" by
    some.

    Battery acid diluted 2-or-3:1, lowish currents, and room temps will
    get you a low-hassle Type II coating.

    I've done it using currents much lower than usually recommended, e.g.
    20-50mA/in^2 from a wall-wart, and was naive enough to be pleased with
    the results. ;)

    You might want to scan these excellent guides:
    http://www.caswellplating.com/kits/lcd_ano.pdf
    http://www.focuser.com/atm/anodize/anodize99.html

    HTH,
    James Arthur
     
  14. D from BC

    D from BC Guest

    Is it the longer it's anodized the thicker the layer?
    I'm just wondering why the type III requires cooling.
    I'll check out the links...
    Thanks
    D from BC
     
  15. James Arthur

    James Arthur Guest

    Yes, up to a limit, whereafter the coating dissolves as quickly as
    it's grown.
    The coating is porous, a hexagonal, honeycomb-like structure. Type
    III anodize uses high current densities, which decreases the size of
    the pores but also makes a lot of heat. Heat, in turn, dissolves the
    coating faster...
    The Caswell plating guide has cool pictures in the front, and a
    technical summary at the end.
    You bet.

    James Arthur
     
  16. Is this epoxy filled with anything? I've had good experience with Epotek's
    Al2O3-filled H77. Just glued two flat metal surfaces together, good to
    1000V (tested).

    robert
     
  17. D from BC

    D from BC Guest

    I'm using an acylic polymer with a 1.25 W/(mK) thermal conductivity.
    The filler material is not listed on the data sheet. It's grey.
    D from BC
     
  18. John Larkin

    John Larkin Guest

    Thermal fillers, like the stuff in silicone grease, tends to be very
    small particles, below 100 microinches, so don't provide much
    insulation spacing.

    John
     
  19. Eeyore

    Eeyore Guest

    I like the idea of little (glass?) beads that'll maintain a specific separation.
    I only came across the idea for glueing core halves together for a reliable gap.

    Graham
     
  20. D from BC

    D from BC Guest

    There's all types of glass. Finding one that conducts the most heat
    would be a interesting.
    D from BC
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-