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.

THS4631: Clarification Regarding Thermal Pad Connection on THS4631DDA and THS4631DGNR

Part Number: THS4631

Tool/software:

Hi,

We’ve designed a board for one of our products using the THS4631DDA, and initially connected the exposed thermal pad to GND, as it was specified in older datasheets that the ePad was not electrically connected to other pins. However, we recently noticed that the datasheet now specifies the thermal pad as being internally connected to V-.

That said, we’ve never encountered any issues in our previous designs, and in fact, verified the electrical connection on several components we received a few months ago — confirming no continuity between the thermal pad and V- on those units.

Has there been a change in the internal connection of the thermal pad for this device?

We’re now developing a new product based on the THS4631DGNR, and would appreciate clarification to ensure proper layout and reliable performance in our updated design. Ideally we would keep it connected to GND as we don't want to redo a PCB design and the GND connection is the largest metal connection (more heat dissipation).

Best regards,

Patrick

  • HI Patrick,

    There was a PCN that was issued for this device as there was a fab and process change. With this change, the new device requires the thermal pad to be connected to Vs-. If the device is in single supply ground would also work, but in a split supply it must be connected to the lowest potential, which in this case would be Vs-. 

    Best Regards,

    Ignacio

  • Hi Ignacio,

    Thank you for the feedback, we’ll make the necessary changes!

    Just as a heads-up: on page 21 of the datasheet (under PowerPAD PCB Layout Considerations), it still recommends connecting the PowerPAD to GND.

    Additionally, we found it a bit confusing that the best thermal metric is listed as ThetaJC(top), while ThetaJC(bot) is marked as N/A. Since the die is thermally connected to the PowerPAD, we would have expected ThetaJC(bot) to be the most relevant parameter. Could you confirm whether the heat should primarily be extracted from the top or the bottom of the package?

    We’re planning to use more than 50 THS4631s on a single PCB, so thermal management is a key consideration for our design.

    Best regards,
    Patrick

  • Hi Patrick,

    Let me confirm everything that was mentioned with our systems engineer, and I will get back to you soon.

    Best Regards,

    Ignacio

  • Hi Patrick,

    I was able to confirm with our program manager about the die attach used for the device and can confirm it is non-conductive. This means the thermal pad is electrically isolated from the die and can therefore ground the pad and does not have to go to the lowest potential. The device was redesigned and updated so there are sections in the datasheet that need editing. As far as the thermal metrics you are also correct. The majority of the heat will be dissipated through the bottom of the package not the top. For your design, setting the thermal pad to the largest plane would be ideal which seems to be ground. I apologize for the confusion.

    Best Regards,

    Ignacio