1. This forum section is a read-only archive which contains old newsgroup posts. If you wish to post a query, please do so in one of our main forum sections (here). This way you will get a faster, better response from the members on Motherboard Point.

PCB trace temperature?

Discussion in 'Embedded' started by Dennis Clark, Dec 16, 2004.

  1. Dennis Clark

    Dennis Clark Guest

    Hi all,

    I've seen and used the trace temperature vs. current calculators and
    seen lots of information on calculating trace temps. However, I can't
    find any information about just how hot you can let a trace get. I
    know that copper melts at about 1100 deg. C, I'm pretty sure that the
    board will be damaged before that,... So, how hot can a trace get and
    still be OK on a board? Where does board damage occur and how hot are
    we "allowed" to get a trace?

    regards,
    DLC
     
    Dennis Clark, Dec 16, 2004
    #1
    1. Advertisements

  2. Dennis Clark

    Dorsai Guest

    Basic rule of thumb: if you can't stand to touch it with your thumb, it's
    TOO DAMN HOT.

    Hope this helps.

    --
    * * * * * * * * * * * * * * * * *
    Dorsai - Author of Erotic Fiction
    http://www.asstr.org/~Dorsai
    * * * * * * * * * * * * * * * * *
    If Bill Gates had a dime for every time Windows crashed...
    Oh, wait a minute, he already does.
     
    Dorsai, Dec 17, 2004
    #2
    1. Advertisements

  3. Dennis Clark

    Gary Pace Guest

    100C on FR4 with good bonding between the copper and substrate. This will
    probably discolour the board over a couple of years. 80C if you're being
    conservative.

    Much hotter on expensive substrates etc.
     
    Gary Pace, Dec 17, 2004
    #3
  4. Keep in mind that the temperature rise is above ambient temperature in
    the area where the traces are. If it's 45°C outside the housing and
    65°C inside the housing, then that's your base temperature. If you
    want to limit it to 80°C then you can't have any more than 15°C rise.
    Yes, some of the expensive substrates are MUCH better than FR4.


    Best regards,
    Spehro Pefhany
     
    Spehro Pefhany, Dec 17, 2004
    #4
  5. Dennis Clark

    Alan Holt Guest

    Alan Holt, Dec 17, 2004
    #5
  6. Unless you are using a circuit board as a heater, the goal is to keep
    the temperature rise as close to zilch as possible.

    Depends on the board: Paper phenolic, fibre glass, kapton, etc.. UL
    is the place to look.

    Now that's a question for your lawyer to answer.
    This is, in general, a very good rule of thumb. In water TDH is ~>115 F.
    and is person and situation dependant. But then in Kuwait it gets to 130F
    in the shade.

    Transformers sometimes run very hot. A class H transformer runs at 180C
    (measured by the change in winding resistance) - This is arc-welder and
    military you-don't-want-know territory.

    For a given temperature very small parts will feel cooler, large parts
    hotter, water the hottest.

    Fahrenheit 451 (~230C) is _way too hot_.
     
    Nicholas O. Lindan, Dec 17, 2004
    #6
  7. Dennis Clark

    Dennis Clark Guest

    Thanks everyone, all food for thought and most helpful. Can someone give
    me the UL spec to consider for FR4 boards? I can look up the document if
    someone can give me the number.

    BTW, that art_thermal.pdf was indeed informative. I'll check up on the
    UL-94 spec, which may be that UL document that I'm looking for, but
    regardless, knowing the FR4 delaminates at 140 C is at least a Starting
    point in my investigations.

    Regards,
    DLC
     
    Dennis Clark, Dec 17, 2004
    #7
    1. Advertisements

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.