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.

TPS56221: TPS56221 vs TPS53353

Part Number: TPS56221
Other Parts Discussed in Thread: TPS53353, , TPS543B20, TPS53355

Hello,

My customer would like to design Vin=12V, Vout=5V and Iout(max)=16A buck convertor at Ta=85oC.
They are considering either TPS56221 or TPS53353, take a look at datasheets and implemented WeBench simulations.
The results are as follows.
                                                        TPS56221           TPS53353

Theta-ja in the datasheet                34.6oC/W            27.2oC/W
Theta-ja at WeBench                      17.3oC/W            27.2oC/W
Tj at Ta=85oC                                 119.272oC           130.385oC

Looking at datasheet, TPS53353 Theta-ja is lower than TPS56221.
But, looking at WeBench simulation results, TPS53353 Tj is higher than TPS56221.
This confuses my customer.
WBDesignTPS56221.pdfWBDesignTPS53353.pdf
Q1:
Why TPS56221 Theta-ja is different between datasheet spec and Webench simulation?
JEDEC spec PCB is used for both datasheet and Webench at TPS53353, but different PCB is used for only TPS56221 WeBench simulation? Why?

Q2:
Actually, which device has more margin thermally?

Which does TI recommend?

Best regards,

K.Hirano

  • The switching frequency for the webench designs is 300kHz (TPS56221) and 500kHz (TPS53353)

    The datasheet uses thermal simulations for the thermal impedance numbers.

    The TPS56221 ThetaJA   states Tjaeffective which implies the ThetaJA was measured using a board.

    The TPS53353 is using the simulation numbers for the webench. 

    The packages are same size, but the fet reisistances are lower for the TPS56221.   

    So I think the TPS56221 will be better.

    I would rank the choices 

    1.  TPS543B20

    2.  TPS56221

    3.  TPS53355

  • David,

    Thank you for your response.
    But I am still confused.
    The datasheet Theta-ja for the three devices are as follows

    1. TPS543B20          29.1 oC/W
    2. TPS56221            34.6 oC/W
    3. TPS53355            27.2 oC/W

    Even if TPS56221 FET resistance is lower than TPS53355, the TPS56221 datasheet Theta-ja is higher than TPS53355.
    I think datasheet Theta-ja is simulated or measured using the same board specified JEDEC for all devices. So, I thought the lower datasheet Theta-ja device, the more thermal margins. But am I wrong?
    Please explain more detail why you think TPS56221 has more thermal margin than TPS53355.
    Also why do you think TPS543B20 is the best?

    I am sorry ask many questions, but I need more clear and specific explanations to persuade my customer for your rank.

    Best regards,

    K.Hirano

  • The datasheet thermal impedance numbers are the simulations using a jedec standard for the size of board, copper thickness, vias, etc.

    When a device is used on a board, the thermal impedance will be lower.   The size of pours, numbers vias and thickness of copper is higher for a board than the jedec assumption, so the thermal impedance will be lower.

    The TPS56221 used the measured thetaja in webench.   

    TPS56221      5x6       4.5/1.9 rdson     2011 release

    TPS53355     5x6       5/2  rdson            2011 release

    TPS543B20  5x7     4.1/1.9  rdson          2017 release

    The power stage resistance of the tps56221 is lower than TPS53355, so the conduction loss will be lower on TPS56221.   

    The packages sizes are the same, so I would expect the actual thetaja on a customer board will be similar.   

    You can use the effective thetaja to calculate a new Tj for the TPS53355.   I think comparing power loss in the IC, will be best method to compare the devices.  

    The TPS543B20 has lower rdson fet (lower conduction losses), larger exposed pad for improved power dissipation  and was release in 2017. 

  • David,

    Thank you for your detailed explanations!

    Best regards,

    K.Hirano

  • Hi K.Hirano,

    Thanks for your confirmation.

    Thanks.