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: IC locking up when ramping input voltage

Part Number: TPS63002
Other Parts Discussed in Thread: TPS63900

Hi,

I'm trying to ramp the input voltage to this chip with enable ON or tied to the input with a 100 Ohm resistor. When I do that it gets stuck at 1.57V on the input and about 2-3V on the output, sinking any available current at the input till it current limits. Could you please advise on why this is happening and how to work around it?

I've also noticed that at any point if this part is starved with power on the input in boost mode operation, it also gets stuck in the same way as described above. My expectation in this scenario is that the part would pull the output voltage down and recover to normal output voltage  when enough power is available on the input. Could you please advise why this is not the case and if there is a way around it? 

This is a custom PCB. I should also mention that if I ramp the voltage while enable is OFF and then turn it ON, everything works as expected. 

  • Hi Marko.

    i think you test with a long cable from source and board, right?

    make the cable as short as you can, and increase the input cap. i think it will be different.

    Regards

    Tao

  • Ok I'll try adding more capacitance on the input to address the inrush / transient current and report back. 

    Could you please advise on the scenario when the device is slightly power starved on the input? Why is not able to continue operating with the output voltage slightly lower and then recover when sufficient power is available? 

  • Hi  Marko,

    i do no know what do you mean about slightly and sufficient power, with a lower or higher current limit?

    Regards

    Tao

  • Meaning the output is trying to draw more power than it's available on the input. As an example, let's say there is a 25 ohm load on the output so it needs 200 mA / 1W. If the input can only source 980 mW, the TPS63002 output drops down to 2V and stays at 2V even if the available power at the input goes back above 1W. So my question is, why does the output drop to 2V instead of respective ~4.95V (assuming 100% efficiency for the sake of the discussion) and why is it not able to recover to 5V when the input available power goes above 1W in this example?  

  • Hi  Marko,

    that seems not make sense. do you have waveforms?

    Regards

    Tao

  • Here is another example with waveforms.

    Output has a 100 Ohm load. Input has 4V and drawing 75 mA. As soon as I change the current limit on the input down to 74mA (1mA difference than what's being consumed) the voltage on the input and output drop down to 1.6V average. See waveforms attached, Input (CH2) Output (CH3). Could you please advise on why this is happening?

       

  • Hi Marko,

    that is much clearer with a waveform.Grinning

    After input current limit drop to 74mA, input can not support your output and input voltage is pulled down till UVLO OFF triggered. then device shunt down and input rise up with 74mA current till UVLO ON. finally device will work between UVLO ON/OFF repeatedly.

    Regards

    Tao

  • To me this seems like a design oversight. Why isn't there some margin here? If the input is slightly power starved, to me it would make sense if the output slightly dips, the whole point of a regulator is to to try to keep the output steady while there is fluctuations on the input. The even bigger concern here is that the device doesn't recover after the current limit is adjusted back to greater than 75mA instead it keeps sinking whatever current is available till eventually reaches thermal breakdown. Could you please comment if this is intended by design? 

  • Hi   Marko,

    what this device do is try to keep the output voltage as a set one and never care about your status of input.

    actually, we have some devices with input current limit function, their peformance will be difference with 63002.

    during your start-up with a small output voltage, device should not only support your output current but also have to charge your output cap(main part). then it will need a much more enregy to support. May be you can set a big delay on EN, this will make your device start to switching with a much higher input voltage and reduce your risk.

    Regards

    Tao

  • Could you please explain what you mean by input current limit function? Which devices are you referring to?

  • Hi  Marko,

    input current limit means device will limit the averange input current. you can take TPS63900 as ref.

    1731.tps63900.pdf

    Regards

    Tao

  • Taking this device as an example what would happen in the same scenario as I described above when the device is power starved? And how does it respond / recover after a such an event when sufficient input power is available in regards to the output voltage? 

  • Hi  Marko,

    output voltage will drop down till a balance achieved. and it can not recover unless you change the limit of input current.

    Regards

    Tao

  • Ok that makes sense and that's what I was hoping is the case with TPS63002. Could you please explain from architecture stand point why TPS63900 will recover when the current limit is increased beyond the required input current while TPS63002 stays at the low voltage 1.6V in this case and does not recover?

  • Hi  Marko,

    they will start-up with a different bias-output voltage. 

    actually, this is not only depend on device, but also components external. EX, you have a large enough input cap, you also can start-up as expected for UVLO OFF will not be triggered.

    Regards

    Tao

  • How much capacitance is needed on the input for TPS63002 to recover after it has hit a current limit? Currently I have 60 uF and the output voltage stays low even after current limit has been increased as I have described before. 

  • increase till UVLO-OFF not been triggered during start-up.

    also, you should keep  cable from power supply to your board as short as possible

  • Tao,

    based on my understanding now the answer to this whole thread should have been that between the input supply and the input capacitance you need to be able to handle the peak current for some period of time. It would be very helpful if the datasheet had a plot of what the inrush current looks like since it's fairly high compared to other devices that I have dealt with, unless it's already there and I missed it. This would allow the engineers to do the proper analysis in selecting the input capacitance. In my case and this example I had to go up to 300 uF to recover with input supply limit being at 200 mA.

  • Hi   Marko,

    it is a little complicated for inrush current not depend on device only.

    for your case, i think it is better you can have an RC delay from input to EN, then you can make sure device start to switching when input is set up ready. And it will be much less possibility UVLO-OFF triggered during start-up.

    Regards

    Tao