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.

TPS61099: About current consumption when Burst mode

Part Number: TPS61099

Tool/software:

Hi team,

When this device is operated below condition. It is observed that the current consumption increase up to about 10mA in a cycle of about 20 ms.

・Using EVM circuit
・Vin=1.5V
・Vout=3.1V
・No load

I guess this is because of Burst mode operation when light load and the reason of this cycle(20ms) is that it takes time to dicrease output voltage to Vout_bst.

Is my understanding correct?

Please your opinion.

Best regards,

teritama

  • Hello teritama,

    Your understanding is correct. In burst mode, the device pushes energy from the input to the output in a short period of time, so the current consumption during this time is higher and then goes very low when it does not burst (switch). The advantage of the burst mode is to have the power dissipation due to switching for just a short period of time.

    Best regards,
    Brigitte

  • Hello Brigitte-san,

    Thank you for reply. In ths case that we want to dicrease this current spike, using LC filter on input or using larger input capacitor is suggested in the original question thread.

    Do you have any material or method of calucration the LC filter or input capacitor value?

    The customer want to know specific value to dicrease ths current spike. It would be helpful just to teach how to guess the guidline values.

    Beat regards,

    teritama

  • Hello teritama,

    Please have a look at this app note: Analysis and Design of Input Filter for DC-DC Circuit

    Best regards,
    Brigitte

  • Hello Brigitte-san,

    Thank you for sharing app note. I'm conserned about in this material said only case of buck conveter and not taken into consideration for boost converter and  burst mode(pulse skip).

    Is it enough following UNDAMPED L-C FILTER section in this AN (Input Filter Design for Switching Power Supplies)?

    Is above consideration is not required?

    Best regards,

    teritama

  • Hello teritama,

    Yes, you can follow the undamped LC filter design as well. You just need to make sure that the input impedance of the filter is low enough for the converter.

    Best regards,
    Brigitte

  • Hello Brigitte-san,


    Thank you for your opinion. I would suggest that if using LC filter, follow undamping LC filter design.

    Incidentally, in the case of using LC filter for spike current like this case, does target cut-off frequency is set around few hundreds MHz tipicaly?

    And especialy, is it correct in assuming that there are no recommended values for input capacitors to suppress spike currents?

    In my understanding in this case parasitic elements from PCB layout can affect to decide LC filter or input capacitor value.

    So, it is difficult that we decide the recommended value of these.

    Is my understanding correct?

    Best regards,

    teritama

  • Thanks Brigitte  for helping to answering !

    Hi Tertama,

    The cut off  frequency is about 5-6KHz, is less influenced by the input LC filter , and the output capacitor and power inductor. You could just refer to the app note Analysis and Design of Input Filter for DC-DC Circuit to surpress the input current spike.

  • Hi Nini-san,

    Thank you for your answer. I would suggest to use LC filter with the cut-off frequency is around 5-6kHz.

    In addition, in my understanding from Brigitte-san's answer, we don't needed that consider about Burst-mode for design LC filter.

    Is this because that burst mode is concerned to spike cycle only, not concerned spike current value?

    Best regards,

    teritama

  • Hi Teritama,

    both the cycle and spike should be considered .

    About 5-6kHz , that is the cut off frequency of the boost converter, other than the LC filter in the front of the boost. Hope this doesn't confuse you.

  • Hi Niji-san,

    >both the cycle and spike should be considered .

    ⇒This means we need to consider about Burst-mode operation to desighn LC filter or dicided input capacitor value?
     

    >About 5-6kHz , that is the cut off frequency of the boost converter, other than the LC filter in the front of the boost. 

    ⇒Sorry for my missunderstanding. You mean 5-6kHz is cut-off frequency of Boost converter AC characteristic, not LC filter requirement.
     So, we need to concern about how attenate this spike current. And when design LC filter is it enough we assume the spike frequency is around 100MHz?

    Best regards, 

    teritama

  • You could set the cut off frequency at about 20kHz

  • Hi Nini-san,

    How about this question?
    >Does this means we need to consider about Burst-mode operation to desighn LC filter or dicided input capacitor value?

    I think in cace that we should consider about current spike cycle, we should concerned Burst-mode operation(enter this mode condition etc. ) , because this current spike cycle is caused by Burst-mode operation. If this is not required I want to know about the reason to explain to the customer.

    Could you tell me about this?

    >You could set the cut off frequency at about 20kHz
    ⇒Is this value from typical use case?

    Best regards,

    teritama

  • The 20kHz is just for mitigating the 100MHz spike frequency. Just for this consideration.

  • Thank you for your kindly supports.