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.

TPS746: Heat Concern

Expert 2385 points
Part Number: TPS746

Hi Ti,

There is a requirement of Vin: 5V and Vout:3.3V@1A. Looking for LDO based regulator and currently looking at TPS746. Looking at the RθJA it's about 80.3°C/W where this is the concern on heat as the losses estimated at 1.7W from 5V to 3.3V.

Can you share whether can we still consider the TPS746? Thanks.

 

  • Hi ,

    While the Thermal Information table provides data that can be used as a first pass thermal check, keep in mind that with proper layout you may achieve better thermal performance than the JEDEC hi-k board on which the Thermal Information table is modeled.  The primary heatsink for TPS746, like most LDOs, is the GND plane of the PCB.  As such, maximizing the GND copper connected to the thermal pad within your application constraints will yield the best thermal performance for your application.

    That being said, 1.7 W is a lot of power to dissipate in a small package such as the 2 mm x 2 mm WSON.  You may want to consider providing a mid rail between 5 V and 3.3 V.  You can do this efficiently with a dcdc switching regulator reducing your power consumption, with two LDOs in series increasing your system PSRR, or even with a power resistor in series before the TPS746.  The last two will not reduce your power consumption; however, they will spread the dissipation and reduce the junction temperature of the TPS746.

    If you are unable to reduce the input voltage seen by the TPS746 by creating a mid rail, you should consider another linear regulator with a lower RθJA.  Due to the fact that simulates the Thermal Information table the same way on a standardized JEDEC hi-k board for every linear regulator, you can directly compare the thermal metrics between two TI linear regulators to get a better sense of which linear regulator will perform better thermally.  You can make these comparisons on our parametric search at

    Very Respectfully,

    Ryan

  • Hi Ryan,

    After evaluating through this LDO and I had faced the Vout dropping to 3.1V with the load of Iout 1A from Vin 5V. Can you advise why is the voltage dropped?
  • Hi ,

    When you say that Vout drops to 3.1 V, do you mean that Vout initially was regulating at 3.3 V as intended before the load was increased to 1 A? So that I can get an idea of how the rails are interacting, could you provide a scopeshot with Vin, Vout, Ven, and Iout?

    Very Respectfully,
    Ryan
  • Hi,

    Yes, we have tested during the load of 600mA the Vout is at 3.3V which is normal. So far we are using the EVK of TPS746 with no modifications done on it.
  • Hi ,

    Thank you for the additional information; however, for me to continue to help debug remotely I need you to provide a scopeshot with Vin, Vout, Ven, and Iout so that I can see the relationship between the rails. As we have discussed earlier in the thread, you are pushing the thermal limits of this package; however, the scopeshot is needed for me to better understand what is occurring.

    Very Respectfully,
    Ryan
  • Hi,

    We have just tied a resistor to the load to get 3.3V and nothing is change from the EVM. Can you help to share me the test report for 3.3V@1A? Else please help to provide further advise on how to achieve 3.3V. Thanks.
  • Hi ,

    I took an EVM to our lab here at room temperature and applied 5 V to the input with the output set for 3.3 V. I was able to achieve a regulated 3.3 V output with a 1 A load.

    Unless you provide the requested multi-channel scopeshot showing Vin, Vout, Ven, and Iout, I am not able to help you debug what could be going on in your system.

    Very Respectfully,
    Ryan