Home > Community > Forums > PCB Design > High(er) current and thermal reliefs - design philosophy

Email

* Required Fields

Recipients email * (separate multiple addresses with commas)

Your name *

Your email *

Message *

Contact Us

* Required Fields
First Name *

Last Name *

Email *

Company / Institution *

Comments: *

 High(er) current and thermal reliefs - design philosophy 

Last post Thu, Dec 20 2012 9:46 AM by TH Designs. 4 replies.
Started by TH Designs 20 Dec 2012 07:23 AM. Topic has 4 replies and 970 views
Page 1 of 1 (5 items)
Sort Posts:
  • Thu, Dec 20 2012 7:23 AM

    • TH Designs
    • Top 25 Contributor
    • Joined on Fri, Apr 13 2012
    • Warminster, PA
    • Posts 269
    • Points 4,370
    High(er) current and thermal reliefs - design philosophy Reply

    Hypothetical question, but somewhat relative to a project I'm working on:

    Suppose you have a design where a trace needs to carry 25 amps DC at 24 volts. This trace interfaces two components. One, a high current MOSFET and the other a surface mount fuse. The connection between the two is very short, on the order of 0.25" or so and will be made using 2oz cu. The trace width is smaller then the tab of the FET, but much larger then the pad of the surface mount fuse.

    I have done many circuits like this in the past and have made the connetcion to both components a solid, flooded copper connection, usually tapered down to the size of the smaller pad. There have been no issues with soldering either part.

    I have been discussing a new design where it has been suggested to use a thermal relief connection to the fuse to aid in soldering. I don't particulary like this approach due to the amount of current involved having to go through the smaller traces of the thermal relief. I suppose it may be possible to size the width of the relief traces such that the sum of their widths comes close to the required trace width, but I don't know if the pad is big enough to support the big relief trace.

    I'd be curious to hear others philosophies on using thermal reliefs in high current situations.

     Tom 

    • Post Points: 35
  • Thu, Dec 20 2012 8:20 AM

    • Roger BFS
    • Top 100 Contributor
    • Joined on Fri, Mar 4 2011
    • Morrison, CO
    • Posts 67
    • Points 965
    Re: High(er) current and thermal reliefs - design philosophy Reply

    Tom,

    This sounds like the classic trade-off between circuit functionality and manufacturing/assembly difficulties.  From my 30yrs of design experience and situations similar to this, the circuit function has to win.  I've always used the rationale: "Do you want it to be easy to build or do you want it to work?"

    In the case of wide traces for current capacity, they are there for a reason.  If overly restricted, trace heating can easily become a problem and thermal reliefs could themselves serve the purpose of the fuse!  Perhaps that might be acceptable in some cases, but not normally.  Even if the narrowed traces can handle the current, they can introduce higher resistance and thus voltage drops that can cause circuit problems.

    With the advent of smaller SMT parts handling higher and higher power dissipations, that has become a challenge to reflow assembly for sure.  That is why I have always insisted that my assembler has to develop a thermal profile for each and every board to adjust for the varying mix of parts from power FETs to 0402 chips and BGAs all often in close proximity.  Sometimes that means using a manual local reflow of heavy components using a "rework station" after or before all the other components have been assembled.

    I have had too many disasters materialize from assemblers just thowing a board thru their generic "one profile works for everyone" reflow ovens.   Those assemblers are only used once by me!

    IMHO - - Regards,

    Roger

    Roger Green - B F Systems, LLC
    • Post Points: 20
  • Thu, Dec 20 2012 8:25 AM

    • oldmouldy
    • Top 10 Contributor
    • Joined on Tue, Jul 15 2008
    • Woking, Surrey
    • Posts 1,408
    • Points 24,030
    Re: High(er) current and thermal reliefs - design philosophy Reply
    Not sure about the philosophy but, if the assembly folks think that they need some reduction from a "solid" connection, you can set the Pin Properties for Dyn_Fixed_Therm_Width, or Dyn_Oversize_Therm_Width to get some larger "spokes" on just this specific pin to improve the current capacity without affecting the thermal "spokes" for the whole design.
    • Post Points: 20
  • Thu, Dec 20 2012 9:43 AM

    • TH Designs
    • Top 25 Contributor
    • Joined on Fri, Apr 13 2012
    • Warminster, PA
    • Posts 269
    • Points 4,370
    Re: High(er) current and thermal reliefs - design philosophy Reply
    Roger BFS:

    Tom,

    This sounds like the classic trade-off between circuit functionality and manufacturing/assembly difficulties.  From my 30yrs of design experience and situations similar to this, the circuit function has to win.  I've always used the rationale: "Do you want it to be easy to build or do you want it to work?"

    In the case of wide traces for current capacity, they are there for a reason.  If overly restricted, trace heating can easily become a problem and thermal reliefs could themselves serve the purpose of the fuse!  Perhaps that might be acceptable in some cases, but not normally.  Even if the narrowed traces can handle the current, they can introduce higher resistance and thus voltage drops that can cause circuit problems.

    With the advent of smaller SMT parts handling higher and higher power dissipations, that has become a challenge to reflow assembly for sure.  That is why I have always insisted that my assembler has to develop a thermal profile for each and every board to adjust for the varying mix of parts from power FETs to 0402 chips and BGAs all often in close proximity.  Sometimes that means using a manual local reflow of heavy components using a "rework station" after or before all the other components have been assembled.

    I have had too many disasters materialize from assemblers just thowing a board thru their generic "one profile works for everyone" reflow ovens.   Those assemblers are only used once by me!

    IMHO - - Regards,

    Roger

    Roger,

    I agree 100% and I was hoping to see responses similar to yours before broaching this subject again with my customer. I have done some pretty high current designs over the years and have never had a board issue........... and I don't want to start.

    • Post Points: 5
  • Thu, Dec 20 2012 9:46 AM

    • TH Designs
    • Top 25 Contributor
    • Joined on Fri, Apr 13 2012
    • Warminster, PA
    • Posts 269
    • Points 4,370
    Re: High(er) current and thermal reliefs - design philosophy Reply

    oldmouldy:
    Not sure about the philosophy but, if the assembly folks think that they need some reduction from a "solid" connection, you can set the Pin Properties for Dyn_Fixed_Therm_Width, or Dyn_Oversize_Therm_Width to get some larger "spokes" on just this specific pin to improve the current capacity without affecting the thermal "spokes" for the whole design.

     Good info (as usual). You intercepted a future "how do I............" question.

    Tom

    • Post Points: 5
Page 1 of 1 (5 items)
Sort Posts:
Started by TH Designs at 20 Dec 2012 07:23 AM. Topic has 4 replies.