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.

MSP430F147 Power Consumption-URGENT Please

Other Parts Discussed in Thread: MSP430F147

I was wondering if you have a listing of how much current the peripherals on the MSP430 use.

The only thing I could find was the comparator.

We are using the MSP430F147 in a battery operated product. We would like to use the PWM with three outputs.

The main clock (external xtal) will be 1Mhz, the PWM resolution will be 256 bits which would make the PWM period about 256 micro seconds.

The processor will be in power down with only the PWMs running.

We need to know how much current the PWM will add to our battery draw.

  • Well, teh power consumption of many peripherals depends on its usage. The timers, for example, require more current if operated on higher frequencies, and no current if not clocked. Effectively, the energy to perform a tick is the factor in question. The current is just a function of the accumulated requried energy over time.

    For the PWM outputs (and output, that is), the current consumption also depends on the load current drawn from the output pins. It depends on the attached hardware. If the outputs ar enot connected to anything, the current consumption is 0. If they are tied to GND, then for the 'on' phase, the current raises to as much as the port can drive (20mA or more). The truth is somewhere between, but unknown to TI and therefore not in the datasheet.

  • Jens-Michael Gross said:
    It depends on the attached hardware. If the outputs ar enot connected to anything, the current consumption is 0.

    You forgot about the current that is present during gate switching, even if output is not loaded. I think that it is good to make an experiment and measure current at the entry  power supply on the board. It should be tested by oscilloscope to check the peeks and also by some avarage means.

    It may be important if it is a battery powered device, where each mA counts.

    Regards,
    Piotr Romaniuk, Ph.D.
    ELESOFTROM

     

  • Piotr Romaniuk said:
    You forgot about the current that is present during gate switching

    Well, yes, there is some current during transitions, but it is only oen transition now and then, so the (minimal) current is neglible. The currents drawn by the timer counter and the compares is significantly higher (and still very low).
    However, I was rather talking about the leakage currents/static currents on these output pins.which are by magnitudes larger and also permanently flowing.

    Piotr Romaniuk said:
    It may be important if it is a battery powered device, where each mA counts.

    We're talking about nA here, and only for very short periods during a long time span. More precisely, it is rather a charge than a current. I think in teh range of few fAs (femto ampere seconds), but that's just a guess.

  • Jens-Michael Gross said:
    You forgot about the current that is present during gate switching

    Well, yes, there is some current during transitions, but it is only oen transition now and then, so the (minimal) current is neglible[/quote]

    Only one transistor conducts a current only for static state. When the gate is switched both transistors (the pulling up and pulling down) conduct the current that flows from positive power supply to ground. I tested it one time, but it was not verified on msp430. Nevertheless this is general rule unles you have open colector output.

    I saw that the application uses PWM, so periodic switch may be expected.

    Jens-Michael Gross said:
    It may be important if it is a battery powered device, where each mA counts.

    We're talking about nA here, [/quote]

    I agree, I meant that the current balance must be considered with special care. 'mA' did not mean this specific units, rather quantitative character of current.

    Jens-Michael Gross said:
    More precisely, it is rather a charge than a current. I think in teh range of few fAs (femto ampere seconds), but that's just a guess.


    Do you mean that battery has some energy that are split over a time? In this case, I would make a balanse by some integral means, so you are right, integration of current is a charge.

    Regards,
    Piotr Romaniuk, Ph.D.
    ELESOFTROM

  • Piotr Romaniuk said:
    Only one transistor conducts a current only for static state. When the gate is switched both transistors (the pulling up and pulling down) conduct the current that flows from positive power supply to ground.

    That's true for the typical totem-pole configuration. However, newer technologies and the need for power reduction have forced the invention of break-before-make technologies (so no transient current flows). For the output it does not make a difference. Either the output is externally pulled high, then a high-to-low transition of the driver keeps the output high for the 'dead' period, or it is pulled low, then it becomes low at the beginning of the dead period. A difference of a fraction of the switching time and usually into a constant direction (the attached hardware usualyl behaves always the same).

    Piotr Romaniuk said:
    I saw that the application uses PWM, so periodic switch may be expected.

    Yes, but a PWM has only two transitions of a few ns per PWM cycle of µs or even ms. 'periodic' does not mean 'frequent' (well, taken literally, it is 'frequent').

    Piotr Romaniuk said:
    Do you mean that battery has some energy that are split over a time? In this case, I would make a balanse by some integral means, so you are right, integration of current is a charge.

    Not only that. In case of FETs, it is literally a charge. The gate forms a capacitor that is charged.
    And yes, the battery is also just a charged capacitor which will be (may be irreversibly) discharged. But the (usable) capacity depends on the load. teh igher the load, the smaller the usable capacity. OTOH, there's the self-discharge. Below a certain point, this is more than the actually used energy, and fighting for every picocoulomb makes no sense anymore.
    But that definitely gets too far.

**Attention** This is a public forum