• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 2238 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.

Looking for a Low-Iq, high-efficiency Buck converter. 3.3V output, 3.6-9V input?

Craig Easdale
Expert 2170 points
Other Parts Discussed in Thread: TPS62745, CC1350, TPS61291, TPS62740, TPS62120

I have a project I'm working on at the moment that requires low power consumption/high efficiency.

There is uncertainty whether we will use Lithium Thionyl Chloride batteries(~3.6V), or Alkaline cells in series (~9V). As a result I'd like to be able to cater for both to avoid a redesign later.

The device will spend about 90% of the time asleep, drawing ~10uA, and in full operation is unlikely to draw more than 30-40mA. 

Some parts such as the TPS62745 look ideal on the surface (decent efficiency, low Iq, wide input), but it seems the minimum input voltage has to be 0.7V above the output which wouldn't work for the lower voltage cells.

Can anyone recommend a part that could potentially work for me?

Thanks,
Craig

  • 0
    TI__Guru*** 145940 points
    Hi Craig,

    Over last year, I believe we discussed most of the devices that provide the best solutions for this type of application. Hopefully, they were able to work for you in your other systems.

    For this request for a common DC/DC, it is not easily possible to obtain an optimal solution for 2 very different battery voltages for 2 reasons: the 9V battery needs a much higher Vin DC/DC AND the lithium thionyl battery's voltage varies so much that you need a buck-boost to get 3.3Vout.

    Here's a training I gave a few years ago which discusses many of the concerns in these systems: training.ti.com/how-power-smart-meter-nano-power-dcdc-solutions

    But yes, the TPS62745 is the ideal lowest power solution for the 9V battery. For the Li-SOCl2, we need to understand the range of output voltage required by the load, if there is a need for another voltage in the system, and the voltage range of the battery you intend to use.
  • 0 in reply to Chris Glaser
    Expert 2170 points

    Hi Chris,

    Thanks for that link, very informative.

    I only have a need for a 3.3V supply in the system, the unregulated battery voltage won't be used at all.The 3.3V will supply the Wireless SoC (CC1350), an accelerometer, and a couple of load cell amplifiers. Potentially I could get away with 3.0V for the system, but that still might not make life all that much easier...

    I was under the impression that the LiSOCl2 batteries had a very flat, fairly predictable output voltage curve until they were completely discharged. Do you mean over a range of varying current draw? To combat the capacity change/voltage drop as a result of higher currents the LiSOCl2 cells would be configured as 5 in parallel. The alternative would be 5 or 6 NiCd/NiMn/NiZn in series.


  • 0 in reply to Craig Easdale
    TI__Guru*** 145940 points
    Hi Craig,

    Yes, the voltage is flat over the life/capacity, but not over current or temperature. Here's one example: www.tadiranbatteries.de/.../SL-2790.pdf

    If the ~3.6V battery voltage needs to be reduced for the load to tolerate it, then the TPS62740 is a good device to use. If the load can accept the entire battery voltage you expect to see in your system, then just connect the battery directly to the load. If the load can accept the highest battery voltage but not the lowest, then a boost like TPS61291 to maintain 3Vout is a good solution.
  • 0 in reply to Chris Glaser
    Expert 2170 points
    It seems that the TPS62740 doesn't support an input voltage of greater than 5.5V, meaning I couldn't potentially use a battery pack configuration with higher voltage...?

    I'm actually using this part in my existing design already, but have had to look at changing due to the lower maximum input voltage.
  • 0 in reply to Craig Easdale
    TI__Guru*** 145940 points
    Correct. You will need different solutions for different types of batteries to have an optimal solution.

    TPS62740 and TPS62745 are pin to pin compatible, so you could install either IC on the same PCB for use with different batteries.

    The TPS62120/2/5 devices meet your overall Vin, Vout, and Iout requirements for both batteries (if you don't need to boost your Li-SOCl2), but have much higher Iq (compared to 360 nA).
  • 0 in reply to Chris Glaser
    Expert 2170 points
    The fact that they are pin-compatible is very convenient!

    I realise I might have to sacrifice an ideally optimal solution for more flexibility in terms of input supply voltage... Other than a marginal drop in efficiency, would there be any reason I couldn't just use the wider-input TPS62745 in both cases?

    Thanks
  • 0 in reply to Craig Easdale
    TI__Guru*** 145940 points
    I wouldn't recommend this for 2 reasons: its Vin only goes down to 3.3V, so it won't support the full lithium battery voltage AND it doesn't support 100% mode--we recommend a voltage difference between Vin and Vout for proper operation.