• Locked Locked
  • Replies 4 replies
  • Subscribers 30 subscribers
  • Views 717 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Audio Amplifier replacement Help

Carlos Silva85
Expert 4915 points

Other Parts Discussed in Thread: LM4951, LM4954

One of my customers is currently using a part that has been marked as EOL by another supplier and its looking for a TI replacement.

The requirement is a mono amplifier capable of output power: 2W with RL = 8Ohm, supply voltage of 9V.

The competitor device had a thermal resistance of 32K/W when soldered to 1.5in2 copper, therefore, assuming the device was operated at 40C, the worst case scenario would be a Tj = 104C.

Seems most of our parts are very small packages with limited thermal dissipation. Do we have anything capable of handling this power level continuously, preferably on a SOIC package or something with a ground plane for thermal dissipation?

Thanks

  • in reply to Scott Bryson
    TI__Expert 4915 points

    I would like to stick with Class A-B in order not to introduce the risk of radiated or electrical noise issues caused by a switching amp. This is a redesign to replace an exiting device and hence layout effort and project risk must be minimized.

    I was briefly looking into the LM4954 and LM4951. It appears that the LM4954 is capable of 3.5W into a 8R load at 7.5V rather than 9V for the ’51 (10% THD+N). That would reduce the cooling requirement for the ’54 since PDMAX drops from 2.05W to 1.42W. How is this possible for a class-AB?

    Thanks

  • in reply to Scott Bryson
    TI__Expert 4915 points
    I can propose this, but its a bit big as a drop in replacement.

    Lets for one second, assume the LM4951.

    This fits, the application, but the thermal is still an issue. At VDD = 9V and RL = 8R the power PDMAX = 2.05W. Our TA is 40°C max. The TI part states θJ-A = 73K/W when soldered to 1in2 copper. The NXP part states θJ-A = 32K/W when soldered to 1.5in2 copper. This seems significantly better.

    But let’s work the other way around:
    With a thermal overhead of 150°C – 40°C = 110°C and a PDMAX = 2.05W we find that θJ-A must be 53.6K/W or lower. How much copper area do I need with the LM4951SD to achieve this ?

    Thanks