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.

TPS63002: High temperature/current and no voltage at random times

Part Number: TPS63002
Other Parts Discussed in Thread: TPS63001, , TPS63000

I am using both TPS63001 and TPS63002 as 3.3V and 5V regulators for my battery powered design - see attached schematic for complete power section. I'm also attaching layout of the  power supply for reference. - there are no components on the other side of the board.

I have had a few boards of my design assembled by an automated assembly line, so we could see the pain points during mass production. We had failures on either device in 100% of the boards. In all of them, the area around the device becomes incredibly hot, the current draw is enormous and the voltage is well below the expected one. Resoldering the device sees some success, but the problem reappears even if the board is good for a while, which is not very good for reliability on the field, specially in mobile applications that tend to get somewhat hot during use. 

All indicates that there is some soldering issue or a thermal design problem, but I would like to check - did I do anything grossly wrong with this design in particular? Is there any other known issue that could account for the problems I've been seing? Does finishing the board with gold flash help with the soldering of these parts in any way?

I have seen this issue on previous designs that used the adjustable brother of these two devices, the TPS63000. In that case the reliability was so bad the company demanded us to rid the entire design of TI parts, which was a pain to do. I would like to avoid doing this on this device, as the power section looks so neat for the application. 

Any help is appreciated!

Thanks

 maha_dac_one-power.pdf

  • Moreno,

    the PCB layout is bad, the GND of the output capacitor should be directly connected to the power pad, instead of such long trace in current PCB. the datasheet provide a good example of layout

    could you try in current board to check if the hot is caused by the layout.

  • Hello Jasper, 

    thank you for your input. I understand you're making this comment based on the visible top layer on the pictures I provided. For simplicity I omitted the ground plane immediately below this circuitry. This is a 4 layer board and there is a solid ground plane right below the top layer. So, all connections to ground should be pretty short. I tried to replicate the DS's layout as best as I could given my board constraints. I think the only difference is the output capacitor connection.

    The effect does not seem to be caused by this, however. On a failed board, giving a some moderate force finger taps on the other side makes the chip behave well - it loses the temperature and starts operating as expected. In this particular case it is a soldering / contact issue, which is aligned to my original inquiry. 

    I would  be ok with just resoldering parts and going ahead.but this is particularly recurring in this component line and not on the remaining TI parts on this design. 

    Can you please re-evaluate your comments in light of the information above and let me know if you still maintain I should perform changes on the layout?

    Regards. 

  • Hi Moreno,

    what is the meaning of "giving a some moderate force finger taps "?

    what is part number of your inductor? what is the output current in the application? could you also try to apply an EVM for reference?

  • Hello Jasper, 

    inductor partnumber is: CIGW252010GL2R2MNE from Samsung

    By giving some moderate force finger taps I mean literally tapping with my finger on the other side of the board from the TPS63002 with moderate force. All components are on top, so the other side of the board has no components. 

    As it seems, there is poor contact between the board and the TPS device so, when I give it a slap or a tap on the oposite side, the pad makes contact and normal behavior is restored. That is my theory, unless you say there is something very wrong with the design itself.

  • Hello Jasper,

    I just realized some of your queries were unanswered as well:

    • The 5V supply is being used to power one single device - an audio DAC (AK4493 from AKM) - in worst case about 50mA, as per their datasheet. 
    • I don't have an EVM to try the part on. I have another functional board with practically the same design here from a previous revision.

    Please let me know if you need further info.

    Regards.

  • Hi Moreno,

    the inductor should be OK. it seem assembling issue is the root cause from your description. do you have a technician to help you re-solder the IC or you could re-solder by yourself? The QFN package should be not hard to handle. 

  • Hi Moreno,

    I closed the post assuming the issue is solved. if not, just reply below