+ Post New Thread
Results 1 to 7 of 7
  1. #1
    Advanced Member level 5
    Points: 32,037, Level: 43
    Achievements:
    7 years registered

    Join Date
    Sep 2008
    Location
    cambridge
    Posts
    6,684
    Helped
    486 / 486
    Points
    32,037
    Level
    43

    TO247 FET mounted in "SMD" fashion and pressurized against heatsink with "foam"

    We have a 4 layer PCB on which we have our offline , all surface mount, LED driver. This PCB sits on a thin thermal pad and this all lies against a big flat heatsink surface. The big flat heatsink is earthed.
    Currently we use a D2PAK power FET and have thermal vias going through the PCB to bottom layer thermal copper pour…but this doesn’t give great heat conduction…therefore….
    We wish instead to have a TO247 FET lying horizontally, with bent leads, so that it can be soldered as a “SMD” FET…….the TO247 tab will be screwed to a small rectanguloid of aluminium (a heatsink), which, through a small cutout in the PCB, will lie against the thermal pad, and thus give good thermal contact of the FET to the big flat heatsink.
    The problem is, we cannot have a threaded hole in the heatsink, so how then will we get sufficient pressure with which to hold the rectanguloid heatsink against the big flat heatsink?….
    …well, there is a flat perspex enclosure some 3mm above the top of the TO247 case (its above the whole PCB) ….and this perspex enclosure gets firmly screwed down…so yes, we are thinking we can put some kind of foam in between the TO247 case and the perspex cover, and then as the Perspex cover gets screwed down, it will pressurize the TO247 FET against the big flat heatsink……and that will do the job nicely...we hope...
    But what materials can we use for the “foam pad”?
    Preferably it will be non electrically conducting, and with a firm springiness so that it can really pressurize the TO247 against the big flat heatsink. Do you know of any such materials?

    •   AltAdvertisement

        
       

  2. #2
    Full Member level 6
    Points: 2,218, Level: 10

    Join Date
    Nov 2016
    Posts
    352
    Helped
    76 / 76
    Points
    2,218
    Level
    10

    Re: TO247 FET mounted in "SMD" fashion and pressurized against heatsink with "foam"

    Quote Originally Posted by treez View Post
    We have a 4 layer PCB on which we have our offline , all surface mount, LED driver. This PCB sits on a thin thermal pad and this all lies against a big flat heatsink surface. The big flat heatsink is earthed.
    Currently we use a D2PAK power FET and have thermal vias going through the PCB to bottom layer thermal copper pour…but this doesn’t give great heat conduction…therefore….
    We wish instead to have a TO247 FET lying horizontally, with bent leads, so that it can be soldered as a “SMD” FET…….the TO247 tab will be screwed to a small rectanguloid of aluminium (a heatsink), which, through a small cutout in the PCB, will lie against the thermal pad, and thus give good thermal contact of the FET to the big flat heatsink.
    The problem is, we cannot have a threaded hole in the heatsink, so how then will we get sufficient pressure with which to hold the rectanguloid heatsink against the big flat heatsink?….
    …well, there is a flat perspex enclosure some 3mm above the top of the TO247 case (its above the whole PCB) ….and this perspex enclosure gets firmly screwed down…so yes, we are thinking we can put some kind of foam in between the TO247 case and the perspex cover, and then as the Perspex cover gets screwed down, it will pressurize the TO247 FET against the big flat heatsink……and that will do the job nicely...we hope...
    But what materials can we use for the “foam pad”?
    Preferably it will be non electrically conducting, and with a firm springiness so that it can really pressurize the TO247 against the big flat heatsink. Do you know of any such materials?
    I guess you would prefer a thermally conductive foam. Not sure but it seems to me that a thermally insulating foam could be counter productive.



  3. #3
    Super Moderator
    Points: 79,430, Level: 68
    Achievements:
    7 years registered
    Awards:
    2nd Helpful Member
    betwixt's Avatar
    Join Date
    Jul 2009
    Location
    Aberdyfi, West Wales, UK
    Posts
    13,011
    Helped
    4345 / 4345
    Points
    79,430
    Level
    68

    Re: TO247 FET mounted in "SMD" fashion and pressurized against heatsink with "foam"

    I disagree with both of you! It should be thermally insulating since it is only intended to put pressure against the tab, if it was conductive, and survived the temperature itself, it would distort the perspex (acrylic) cover. I think this is a bad idea but then I have no idea what temperatures or profile your product experiences.

    The hole doesn't have to be threaded, use a self-tapping screw or if the heat sink is thin enough, use a rivet. If it is close to the edge of the PCB, consider a spring clip extending over the board to do it.

    Brian.
    PLEASE - no friends requests or private emails, I simply don't have time to reply to them all.
    It's better to share your questions and answers on Edaboard so we can all benefit from each others experiences.


    1 members found this post helpful.

    •   AltAdvertisement

        
       

  4. #4
    Advanced Member level 5
    Points: 32,037, Level: 43
    Achievements:
    7 years registered

    Join Date
    Sep 2008
    Location
    cambridge
    Posts
    6,684
    Helped
    486 / 486
    Points
    32,037
    Level
    43

    Re: TO247 FET mounted in "SMD" fashion and pressurized against heatsink with "foam"

    The hole doesn't have to be threaded, use a self-tapping screw or if the heat sink is thin enough, use a rivet. If it is close to the edge of the PCB, consider a spring clip extending over the board to do it.
    Thanks, the big flat heatsink isnt thin enough to be riveted. Unfortunately, we cannot put holes into it, in any case, they have already been manufactured without a hole in them, and its considered too expensive to put a hole in them.
    Unfortunately theres no way we could easily use a spring clip.

    The heat profile is that this is just an outdoor light for parking lots in Europe. We are exercising in how much power we can put through our current hardware, and so for the upper power levels, we see temperatures of nearly 130degC at the junction of the D2PAK FETs, and so we wish to change to TO247 with better thermal handling.

    The D2PAK we are currently using is SPB07N60C3...
    https://www.infineon.com/dgdl/SPB07N...12b42df065491f



  5. #5
    Member level 2
    Points: 632, Level: 5

    Join Date
    Jun 2017
    Location
    near the sea
    Posts
    49
    Helped
    10 / 10
    Points
    632
    Level
    5

    Re: TO247 FET mounted in "SMD" fashion and pressurized against heatsink with "foam"

    If you stated the power dissipation, thermal resistance of your heatsink and max Ta that would be helpful as it would be possible to calculate the allowable interface thermal resistance. Is the heatsink and pcb secured to the case or are you expecting this interface to handle mechanical forces in addition to a simple thermal interface ?


    1 members found this post helpful.

  6. #6
    Advanced Member level 5
    Points: 32,037, Level: 43
    Achievements:
    7 years registered

    Join Date
    Sep 2008
    Location
    cambridge
    Posts
    6,684
    Helped
    486 / 486
    Points
    32,037
    Level
    43

    Re: TO247 FET mounted in "SMD" fashion and pressurized against heatsink with "foam"

    Thanks, the big flat heatsink is essentially the case. The PCB is screwed to that.
    The small rectanguloid heatsink is somehow to be connected to the big flat heatink



    •   AltAdvertisement

        
       

  7. #7
    Member level 2
    Points: 632, Level: 5

    Join Date
    Jun 2017
    Location
    near the sea
    Posts
    49
    Helped
    10 / 10
    Points
    632
    Level
    5

    Re: TO247 FET mounted in "SMD" fashion and pressurized against heatsink with "foam"

    Quote Originally Posted by treez View Post
    Thanks, the big flat heatsink is essentially the case. The PCB is screwed to that.
    The small rectanguloid heatsink is somehow to be connected to the big flat heatink
    I assume by small rectangaloid you mean the TO247, why use convoluted language ?

    So back to the real questions, do you know the thermal resistance of your main heatsink/case ?

    Do you know the power dissipation of the TO247 device ?

    Do you know the maximum specified Ta ?



--[[ ]]--