• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 917 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TPS62740: Ultra low power buck to control 3.6V battery load when charging 1F supercapacitor at 3.6V

BIDAULT Hoel
Intellectual 320 points
Part Number: TPS62740
Other Parts Discussed in Thread: TPS61291, TPS61021A, TPS61099

HI,

We are looking for an ultra low power buck in order to charge a 1F supercapacitor at 3.6V, the system is powered by a 3.6V LiSOCl battery which can provide only limited current (10mA nominal, 30mA peak) our load is drawing 47 mA max for 11 second max (per 24h) rest of the time the MCU is only powering it internal RTC, the supercapacitor can accept voltage up to 5.5V and has no charging curent limit, thus the problematic is limiting the current drawn from battery during charging with best possble efficiency and eventually control the load switching during the 11 second duty period (not sure if it is really needed since the supercap can provide over 50mA and thus the battery shouldnt be sollicitated even with no load switching).

We need something similar to the TPS62740 with ultra lo Iq, 3.6V output (the TPS62740 output is 3.3V max ), current control and load switching.

The idea is to limit the load to the battery at 10mA or less when charging, avoid losses associated with a current limiting resistor, and eventually being able to switch load to supercapacitor only when the MCU enters it duty period and draw 47mA (via MCU gpio control), ideally we would like to avoid need for any MCU control during the charging, so that ithe MCU can stay in RTC standby mode everytime except during it 11 second duty.

Any suggestion will be very welcome

  • 0
    TI__Guru*** 145940 points
    Hi Bidault,

    Did you already look at this reference design on the TPS62740 product page? It looks exactly like what you describe: http://www.ti.com/tool/PMP9753

    There isn't such a low Iq device which gives a higher Vout, so you'll need to use 3.3V or lower. Have you considered the impact of the leakage current of the super cap at the higher voltages? Some of them leak more as the voltage increases, so the voltage is kept low.

    In case you need a boost between the super cap and PA, there is this ref design: http://www.ti.com/tool/PMP9763
  • 0 in reply to Chris Glaser
    Intellectual 320 points

    HI Chris,

    Thank you for the prompt reply. Yes the TPS62740 does what we need, except for the lower output voltage (but we can work at 3.3V) and yes i have seen the associated reference design (and even read your paper on Iq which is interresting).

    Using a boost has not been considered intially due to increased lost of efficiency however, it appears that given the very long sleep time (with RTC) between duty periods (over 23:59:39) the supercap leakage current is actually a most major concern for the energy budget. 

    So I think of using the advised combination of TPS62740 as supercap charging buck and TPS61291 as 3.3V boost.

    Regarding the supercap leakage current do you think charging the supercap just enought so that it can handle the 11 second duty period and end up at 0.9V (boost min Vin), then charging it only before the next duty period would be less worst than keeping it charged all the time? We dont use the Murata supercap wich is too expensive but rather a cheaper and widely available 1F 5.0V supercap, i have calculated the actual power needed for our 11 second duty period, 0.33F is enough, not sure yet if using 1F 5.0V supercap is a good choice (but i have 0.33F 3.6V to compare).

    N.B. what is the point of using a supercapacitor charging current limitiing resistor behind the TPS62740? the TPS62740 is already limiting the current, so I dont understand why we would add a current limiting resistor here, it certainy increase losses and i dont see the benefit ( R2 / R3 here http://www.ti.com/lit/df/tidrg80/tidrg80.pdf 

  • 0 in reply to BIDAULT Hoel
    TI__Guru*** 145940 points
    Hi Bidault,

    I'm glad you found those collateral useful. Here's some more to read, which should answer your questions :)

    This training video walks through the circuit's operation: training.ti.com/how-power-smart-meter-nano-power-dcdc-solutions

    This blog summarizes what this circuit does: e2e.ti.com/.../how-can-you-deliver-1w-power-from-a-10-mw-coin-cell

    Basically, the resistors limit the output current and this limits the input current into the proper range for your LiSOCl battery.

    Yes, your conclusion is what we calculated as well but it depends on the component specifics: a lower super cap voltage for most of the day is better. This is why the reference design spends most of its time at 1.9V for the MCU.

    Also, note that the TPS61291 cannot be opened, so it will always have a Vout. This might be good if your MCU is there or bad if you want to remove all leakage current.
  • 0 in reply to Chris Glaser
    Intellectual 320 points

    Hi Chris,
    Thank you for the return and link to the video. Our supercap datasheet specifies 10uA per 72h leakage current. Controlling the supercap charging via the MCU, keeping the MCU VCC low and charging the supercap just enough to handle the burst will be the key to minimize losses here.

    N.B. cwhat about adding a PMOS (for example SI2301 )before the TPS61291 in order to cut it during idle time?

  • 0 in reply to BIDAULT Hoel
    TI__Intellectual 1010 points

    Hi Bidault,

    Thank you for choosing TI ultra-low Iq boost converter to get a regulated 3.3V from a supercap input. We also have a 1uA Iq boost converter TPS61099 with load disconnect feature. With load disconnect during shutdown, you don't need to add a PMOS to cut the boost converter from the input supercap. I'm not sure if WCSP package is suitable or not for your application. If you can not use WCSP package, there is only 3.6V version TPS610995DRV available in QFN package. Considering most time the boost converter is in shutdown, I think you can also choose boost converter with a little bit higher Iq. So another choice is TPS61021A which can work with the input voltage down to 0.5V. The TPS61021A can be set to 3.3V output by external resistor divider.

    thanks

    Gavin