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.

Battery Selection for MSP430

Other Parts Discussed in Thread: MSP-EXP430FR5969, MSP-EXP430FR4133

Hello,

I'm trying to get a gauge of what type of battery I might need in order to power the MSP430.  I haven't picked a specific model yet, so it is somewhat variable, but one of the ratings I've seen is:

  • 220 μA / MHz active mode

To get the number of hours a battery will last I believe I would just take [rating in amp hours] / 220uA.  If I were to have a battery rated to 220 mAH then this microcontroller would last ~ 1000 hours in active mode.  Is "active" mode simply with the chip in active mode while it's idle?  1k hours seems absurdly long for a small battery.

Considerations to ESD and Vcc seem to indicate from the datasheet that current consumption may exceed 220 uA more often than not.  http://www.ti.com/lit/ds/symlink/msp430fr2532.pdf 

To be more specific I have to drive a potentiometer (with ADC) and a bluetooth module (likely UART).  How might I derive the power requirements of the MSP430 for this?

Thanks!

  • Hello David,

    Active mode current supply is compared to low-power mode current supply, which for you application I would definitely recommend staying in LPM3 for as long as possible. The ADC and UART can both be sourced from ACLK driven by a LFXT, this allows you to "sleep" while only consuming ~ 1 uA of current and wake up (active mode) to quickly service ADC/UART ISRs before falling back asleep (low-power mode). By utilizing MSP430 LPMs you could be capable of increasing your battery longetivity by 200x (no surprise here as our devices are designed for ultra-low power applications that need to last years on a single battery). Peripheral current consumption is also documented inside the datasheet to help generate a rough estimate, but it will be difficult to provide more precise details without physically testing the solution yourself with factors like ADC sampling rate and frequency of UART communication. If you do not have to wake up to receive data over UART then you may even be able to utilize a LPM3.5 solution using an RTC for further current consumption benefits. TI provides several cost-effective LaunchPads (MSP-EXP430FR4133, MSP-EXP430FR5969, etc.) to help evaluate an application using our MSP430 devices. Make sure to use a battery that does not go outside of the MSP430's allowable supply voltage range or violates the minimum Vcc vs DCO frequency specifications.

    Regards,
    Ryan
  • David Gooden said:
    To get the number of hours a battery will last I believe I would just take [rating in amp hours] / 220uA.


    No unless you are in the case where MCLK = 1 MHz. The spec is µA per MHz. So if running, for example, at 8 MHz, the number is 220 µA * 8 = 1.76 mA.

    David Gooden said:
    If I were to have a battery rated to 220 mAH then this microcontroller would last ~ 1000 hours in active mode.


    Wrong, for reason specified above.

    Also note that the datasheet is clear in the footnotes for the current consumption data that it is specified with no external I/O current. Additional current required for I/O depends on the specifics of your implementation. Things that factor into this include pullup/down resistor currents, I/O voltage, trace and "other device" capacitance, and frequency of the I/O signal transitions.

    David Gooden said:
    How might I derive the power requirements of the MSP430 for this?


    Remember that external signals transitioning is basically charging and discharging mini capacitors. Using information from the datasheets and your timing information, you can calculate the charge transfer and average power consumption of your external I/O.

    The easiest way is to build the capability to measure into your board and use the appropriate instruments to measure real-life data. Compare it to your theoretical calculations (which seem to always underestimate) and come up with a good calibration factor to apply to future power estimation efforts.

**Attention** This is a public forum