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.

TPS62745: High Load Transient Response

Part Number: TPS62745

Hi

The system's input voltage is 6-4 V, and the TPS62745 output is fixed to 3.3V. The system includes nRF52832 (SOC) and SPI Flash.

The peak load is around ~24 mA (nRF52832 advertising + read/write Flash).

when the TPS62745 input is 4V, the Load Transient Response increase to ~ 270mV and more- see attached picture .

when the TPS62745 input is above 4V, the output ripples are around 20mV which it's OK.

1.Is the TPS62745 can't work without ripples while input 4V , output 3.3V with ~24 mA load ? do i need to replace the TPS62745 ?

2. Is it possible to reduce the Transient Response high ripples? If so, how? 

Please advice

Thank you  

  • Hi Nir,

    Is it possible to add the Vin and load current to the scope, as well as Vout? There may be significant input source impedance which accounts for part of the drop.

    Is your maximum load current 24 mA? Does this include the load during the peaks shown in your waveform?
  • Hi Chris,

    Unfortunately i can't add the load current to the scope, but i'll measured it with DMM and will send it to you.

    The maximum current measured is much lower than 24 mA - evaluation of about ~10mA 

    From my opinion, the focus needs to be on the advertising , because during the advertising the voltage drop is the highest and more impotent thing is that the STBY current before the advertising is few micros Amperes...  

    Do you have maybe the information regarding load transient response from few micros to 10 mA? 

    Please see below the nRF52 current consumption:

  • Hi Nir,

    It's not clear to me if the 24 mA is the peak current or some sort of average current. We really need to know what the peak current is--the current that the load is drawing during the dips in your waveforms. If you have a current probe or differential probe, you can measure the output current/voltage through/across a 1 Ohm resistor.

    Is it possible to share your schematic and a waveform of just Vin and Vout?
  • Nir,

    I took some transient measurements on one of our EVMs and I did not see the large transient responses that you are seeing. These were all taken with 3.3Vout,  0mA-30mA transients with a 1A/us slew rate.

    3.8Vin:

    4Vin:

    4.5Vin:


    Thank you,

    Nick

  • Hi Chris ans Nick

    Thank you for the reply, appreciate your help with that.

    Nick - It is very clear that there is a big difference between the results of your measurements compared to mine.

    please review design and schematic below:

    *please ignore the (nc) at the schematic , all components are connected 

  • Nir,

    I do not see anything wrong with your layout or schematic, assuming that R40 and R41 are not both populated. Your GND pins should also be shorted to the thermal pad, but I don't think that would be causing such a big transient.

    Would it be possible to see a waveform of Vin and Vout at the same time? If there is a drop with Vin that could be adding to this problem.

    What would also be nice is if you could find some spot to solder in a resistor or wire that could be used to measure the current since I believe that would get us the most accurate measurement of current load. Unfortunately I don't see a good spot to do this in the section you have shown here.

    You could also try connecting an EVM to this board and see if that solves the problem.

    Thank you,
    Nick
  • Nir,

    Were you able to solve this issue? If not, were you able to measure Vin during these transients, or try connecting an EVM to see if there is any improvement?

    Thank you,
    Nick
  • Hi Nick

    Apologies for the late response

    unfortunately i don't have EVM

    Please see the Vin (4V) & Vout measurements  below:

    DC measurements  :

    AC measurements  :

  • Nir,

    You input voltage looks fine.

    Are you measuring the output voltage using a low inductance probe across the C24? Your probe should use a small wire wrapped around the barrel of the probe to connect the GND instead of the typical wire clip that the probes come with.

    Is there anything else besides the nRF52832 connected to the 3.3V rail? Or anything else between the load and this converter such as a ferrite bead?

    Thank you,
    Nick
  • Hi Nir,

    Have you been able to measure the actual current during your bursts? This app note describes one method: infocenter.nordicsemi.com/.../nan_27.pdf