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.

TPS61088: TPS61088 output voltage drops to a fixed level and does not recover

Part Number: TPS61088

We used TPS61088 to convert 3.3V to 6.55V, and the schematic was generated using WEBENCH Power Designer. Another step-down DC/DC powered from 6.55V, total load less than 3W(measured at the 3.3V input).

After working continuously for more than 12 hours, the output suddenly drop to ~2.89V and does not recover. The output will return to normal after power cycle.

We have had this issue on two devices. Can someone help me?

  • Hi

    it seem like the device is disabled which causes the output drop to 2.89V. please check the VIN, EN and VOUT waveforms during normal operation and VOUT=2.89V condition.

  • Jasper

    Thank you.

    The EN pin connect to a fixed voltage,  from 3.3V input. I forgot to record the EN waveform.

    From the waveform of FSW, TPS61088 is out of work, but I confirm that EN is fixed at 1.7V. When working normally, FSW is a periodic waveform.

    Vin: ~3.31V

    Vout: ~2.9V

    FSW: ~1.2V

  • Hi,

    the device stops switching. thus there is ripple in FSW pin.

    could you try to toggle the EN? EN changes to low and High again and check if the IC can work.

    also make sure the EN pin is well soldered to the PCB. somethings solder issue causes some strange behavior.

  • This is the schematic we used. Rcomp is 28.7k, Ccomp is 1nF. Webench shows the Phase Marg is 52.39.

    The SW is fixed at 3.3V when the fault occurs, not periodic. When we test with an electronic load, the SW is periodic even under overload. The actual typical load is 3.5W, and the maximum is 5W. To leave some margin, 1A load current was selected when using Webench.

     EN is connected as follows, so it's always ~1.7V.

    When the fault occurs, SS voltage is 1.2V, which may be a problem. What will cause it? We connect SS to ground through a 0.1uF capacitor.

  • Hi,

    it is not stability issue as the device doesn't switch. it is not output power either. 

    1. please try to toggle EN if the IC can work.

    2. please measure the FB pin voltage.

    also the attached waveform doesn't show. please try again

  • Hi, Jasper

    Thank you.

    Failure takes a long time (two to four days) to occur, and only occurs at high temperatures above 40 degrees Celsius, and cannot be recovered even when the temperature drops to 25 degrees.

    The board did not work after overload, I will try to toggle EN after fix it.

    Fb drops to ~0.58V (the voltage measured by oscilloscope is a little high, actually it should be a little lower when measured by multimeter).The next time when it occurs, I'll use a better oscilloscope.

    FB waveform:

  • Hi, Jasper

    If the inductor is saturation, will the DC/DC not work, or just unstable?

    The inductor MPN is LPS6225-102M, Irms@40°C = 1.65A. The 3.3V input current is 0.9A, this may cause the inductor to saturate.

  • Hi

    i'm confused. is the IC damaged? you mentioned the device can work normally after recycling the power supply. 

    the device can work with inductor saturation, but may be unstable. but this is risk that the device got damage when inductor become saturation.  the saturation current of this inductor is high. what is the output current in the application?

  • I'm sorry, I may not have described it clearly.

    When we use the DC/DC on the device, the real load is only 3W, 3.3V input current is 0.91A, and the 6.55V output current is 0.5A. In case of failure under this condition, FSW is a fixed voltage, the DC/DC can work normally by recycling the power supply. 

    In order to do some stress testing on the DC/DC, we disconnected the real load and connect the DC/DC to an electronic load. The stress testing including undervoltage, overcurrent and over power, the DC/DC works well under these condition. When I increase the power to the point (larger than 12W) where it can't handle it, the output voltage drops to 2.89V, but FSW is still a periodic waveform. After these stress testing, the circuit appears to have been damaged, but it still works well under light load.

  • Hi 

    the device should not damage at 12W condition. could you try to replace the inductor with a larger current one, and then repeat the experience again. 

    also please share me the image of the layout. the IC damage is normally caused by layout

  • I have purchased the inductor with large current, and I am going to verify it again.

    The image of PCB will be provided tomorrow.

    Thank you.

  • PCB layout image:

  • Hi,

    the layout is not well optimized. please refer to these AN to optimize the layout 

    https://www.ti.com/lit/an/slva773/slva773.pdf 

    https://www.ti.com/lit/an/slva790a/slva790a.pdf 

    I'm not 100% sure your observation is caused by the layout. so I also would suggest you apply an EVM and connected to your system board for measurement. 

  • Thank you, we will check the PCB layout and try to verify using EVM.