• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 374 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.

LM3478 EVAL BOARD EXTERNAL NFET OVERHEATING

Duncan Ashworth
Prodigy 20 points
Other Parts Discussed in Thread: LM3478

I am using a modified LM3478 EVAL BOARD to generate 24V @ 3A. It works OK but the external FDS6690A power NFET gets very hot so I don't run it very long. What can be done? Is there a better NFET selection: put some in parallel, add a heat sink, etc? Per WEBENCH recommendations, I'm using:

1. L= 4 uH Inductor DCR=1.94mOhm  IDC=25A (CoilCraft: SER2014-402MLB).

2. D1= Vishay VS-30CTH02SPBF

3. Rsn= 0.003 Ohm.

4. Rfb1= 16200 Ohm

5. Rfb2= 1000 Ohm

By the way, it's confusing the way Webench switches the definitions of Rfb1 and Rfb2.

 

  • 0
    TI__Genius 15810 points

    Hello Duncan,

    Is the Fairchild FET the actual part that Webench has suggested? Also, what are your design requirements e.g. Vin range and ambient temperature?

    Regards,
    Akshay 

  • 0 in reply to Akshay Mehta
    Prodigy 20 points

    Thanks for the prompt reply.
    WEBENCH recommended substituting a Vishay-Siliconix SIR812DP-T1-GE3 NFET for the FDS6690A but from the data sheets, it looks like the FDS6690A is a better choice. Maybe I'm missing something.
    The Vinput range is from 8 to 15 Volts.
    The ambient operating temperature will be from 5C to 35C.
     
    Thanks,
    Duncan Ashworth
    http://www.floorbotics.com

  • 0 in reply to Duncan Ashworth
    TI__Genius 15810 points

    Hello Duncan,

    Webench actually calculates the power dissipation in the FET and the juction temperature of the FET at the worst conditions and picks the FET only if it passes those checks. Therefore if Webench has suggested the Siliconix FET, then it would have made sure that the FET would not overheat. 

    For your design of 3A at 24V boosted from 8V, the current through the FET is going to be considerably large. Therefore the FET's RdsON would matter a lot as it would be directly proportional to the conduction losses in the FET. The Fairchild FET has 17mOhms at 4.5V Vgs and the Siliconix has 1.65mohms at the same Vgs. This is almost 10 times the difference. The Siliconix FET also has a power pad which helps heatsinking very well and consequently helps push the power dissipation limit a little higher. 

    I hope this information helps. If possibly you should switch to the other FET.

    Regards,
    Akshay