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.

Choosing the right DC/DC Boost for low current consumption application

Other Parts Discussed in Thread: TPS610981, TPS61021A, TPS61098, TPS61099, TPS61023

Hello,

I am developing a battery powered product for IoT application and need to choose the right DC/DC boost regulator (3V3). The specifications are provided below.

The biggest problem is that I can't find proper boost converter which is capable of providing 500mA current and also having a good efficiency at low current 5uA. I have already tried WEBENCH Power Designer tool, but didn't have much success. 

The board I am developing will be used for periodic data sending to server (using cellular module). The module will consume around 500mA (peak current) when transmission/reception is active. This will happen only for a short time. Around 99,7% of time the module will be in idle mode which will consume just 5uA. The efficiency of DC/DC boost is essential at this low current.

Specifications:

  • Vin,min = 2,0V
  • Vin,max = 3,2V
  • Vout = 3,3V (fixed or adjustable type)
  • Iout,min = 5uA, (99,7% of whole time will be in this mode); I need good efficiency here (at least 50% including all Quiescent currents)
  • Iout,average = 140mA, (only 0,3% of the whole time); efficiency should be at least 93% (including all Quiescent currents)
  • Iout,peak = 500mA, DC/DC should be capable of providing this current only for a short time (e.g. 100ms), efficiency should be around 90% (including all Quiescent currents)

With the help of TI support we have already narrowed down the choice for DC/DC boost to TPS610981 or TPS61021A

TPS610981 is of course the better choice, because of very low Quiescent Current at Vin and Vout. But, because it is not available in Webench Power Designer I cannot properly calculate efficiency at 5uA current and different voltages. And also, according to datasheet I am not sure if this boost regulator can provide 500mA at Vin=2,0V as I would need. Am I reading those datasheet charts properly?

TPS61021A is preferred choice, because it is cheaper. I am trying to calculate current consumption for 5uA load with the help of Application Report (Calculating and Measuring the No Load Input Current of the Boost Converter). I am having some troubles with understanding regarding used efficiency in that equation. Should I use 13,1% (according to Webench calculation at 2,1V input and 3,3V output, 5uA) or something else?

Could you please help me with the efficiency calculation at 5uA for both regulators. Also, could you confirm if those two regulators are even ok for my specifications and if yes, which one is better. If any better regulator is available, please let me know.

Thanks in advance,

Jure

  • Hi Jure:

    How about TPS61099? It's a ultra low quiescent like TPS61098, but has larger current capability. 

  • Hi,

    I have checked the TPS61099 (adjustable) and TPS610994 (fixed) versions. This boost regulator would be good regarding the quiescent current, but does not meet my requirements for voltage range and max current.

    According to datasheet and also Webench Power Designer I just can't get 500mA peak current with input voltage range of 2,1V - 3,2V. Do you have any other solution for this requirements?

  • Hi Jure:

    The webench do the calculation based on the worst case. The 500mA is just at the corner for such voltage range.  

    TPS61023 should have enough current capability, but I afraid the Iq is a little larger.

  • Hi,

    so do you think that with TPS610994 I could get 500mA for a short time only (like 100ms)?

    What would happen if I use 3,2V input on this boost regulator, because Webench doesn't allow me to use this voltage (3,1V is max according to Webench).

    From a datasheet of TPS610994 I also encountered that the Efficiency curve starts at 9uA. What would happen if the load is 5uA like in my case?

    I checked TPS61023 with Webench and wasn't getting much success. With my parameters the design was not stable (because of Phase Margin).

    Just to make something clear. Does the calculated parameter Iin Avg (Average input current) in Webench also include Quiescent current into VIN and VOUT?

    Once again thank you for the help.

    Best regards, Jure

  • Jure Savli said:

    so do you think that with TPS610994 I could get 500mA for a short time only (like 100ms)?

    [MQ]I do the calculate with the minimal current limit of TPS610994 and typical efficiency. It can supply 3.3V/476mA, when Vin=2.1V. That's why I said it's just at the corner. But if unfortunately, the load is 500mA and it meets the worst case, the Vout will drop.

    What would happen if I use 3,2V input on this boost regulator, because Webench doesn't allow me to use this voltage (3,1V is max according to Webench).

    [MQ] It's available if Vin=3.2V or even high. The WeBench may has some error. I will note it and feedback to the webench team.

    From a datasheet of TPS610994 I also encountered that the Efficiency curve starts at 9uA. What would happen if the load is 5uA like in my case?

    [MQ] I checked the database of TPS61099, and the efficiency is 42-46% when Vin=2~3.2V, Vout=3.3V, Io=5uA. For TPS610994, the efficiency may be a lilltle higher with the modified feedback circuit integrated.

    I checked TPS61023 with Webench and wasn't getting much success. With my parameters the design was not stable (because of Phase Margin).

    [MQ] TPS61023 can handle this case. You could enlarge the capacitor to make it work.

    Just to make something clear. Does the calculated parameter Iin Avg (Average input current) in Webench also include Quiescent current into VIN and VOUT?

    [MQ] It's hard to say, especially for such light loading case. The data in datasheet is more reliable. 

  • Hi,

    thank you for the detailed response. Things are much clearer now. 

    Best regards