Power dissipation rating

Discussion in 'General Electronics Chat' started by Veracohr, Apr 6, 2015.

  1. Veracohr

    Thread Starter Well-Known Member

    Jan 3, 2011
    552
    76
    When a transistor package specifies a power dissipation, that's probably average power right? I'd like to use a MMPQ6700 quad transistor package for something, which is rated at 1W total package dissipation. In simulation each transistor is dissipating over 500mW peak, but average around 215mW. Think it might be OK?
     
  2. BobTPH

    Active Member

    Jun 5, 2013
    785
    114
    Depends on how long they might stay at the max.

    Bob
     
  3. joeyd999

    AAC Fanatic!

    Jun 6, 2011
    2,692
    2,756
    Notice that you also have to derate 8 mW for each degree Celsius at ambients above 25C.
     
  4. ScottWang

    Moderator

    Aug 23, 2012
    4,855
    767
    Total device dissipation : 1000mW
    Each bjt dissipation : 1000 mW/4 = 250mW

    For the usage life and concern the heat.
    Vceo = 40V, using 80%, Vce = 40V*80%=32V
    Ic = 200mA, using 20%, Ic = 200mA*20%=40mA
    Wmax = Vce*Ic= 32V*40mA = 128mW
     
  5. crutschow

    Expert

    Mar 14, 2008
    13,052
    3,244
    What is the duration of the peak power?
     
    Lundwall_Paul likes this.
  6. Veracohr

    Thread Starter Well-Known Member

    Jan 3, 2011
    552
    76
    Well the circuit is a headphone driver so that depends on what the input is and the level of the output. Simulating with a sine wave at max output power it's around peak for about 28% of a cycle, and it's above 250mW for about 43% of a cycle (using 250mW per transistor as a maximum).

    That was at maximum level before an opamp in the circuit starts clipping. I don't really know what a comfortable output level will be, but reducing the output power to what might be a reasonable listening volume (~50mW load power) gives about 17% of a cycle within 10% of the peak, and about 31% of a cycle above 250mW.

    Temperature isn't a concern, this will just be a temporary breadboard thing exposed to the room. I'm only building it so I can listen to the output of other audio circuits I'll be working on. If need be I can just go with single TO-92 transistors instead of the quad.
     
  7. crutschow

    Expert

    Mar 14, 2008
    13,052
    3,244
    So just feel the transistors when they are operating to see if the temperature becomes excessive.
    Chances are you'll be okay.
     
  8. Veracohr

    Thread Starter Well-Known Member

    Jan 3, 2011
    552
    76
    Thanks.
     
  9. Veracohr

    Thread Starter Well-Known Member

    Jan 3, 2011
    552
    76
    In case anyone is interested, I found out 50mW is WAY too much power for headphones. A reasonable listening level is more like 100 μW. With peaks around 50mV into 36Ω headphones the level was on the high side of comfortable listening (for me), and peaks around 100mV was OK only for short term.

    I was going off the fact that my headphones are rated for 500mW. A person would have to be mostly deaf to stand that much!
     
Loading...