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.

LP5907: Voltage Regulator for Battery based very low power consumption system - needed

Part Number: LP5907
Other Parts Discussed in Thread: TPS22916, TPS7A37, TLV757P, LP5912, TPS62122, TPS62740

Hi,

I am developing a very low power consumption product that runs purely on a battery. We have an RF communication peripheral within our product. The power consumption values of our product at various instances are as below. As you can observe from the table, our product will be in Sleep state for more than 90% of the times. For this, I started pursuing LP5907MFX-3.0 and using the same on my board, for a quick Proof-of-Concept. However I knew that LP5907 is not suitable for very low power applications. LP5907 is consuming power as shown below, and it is less efficient. For your reference, I have also put snapshot of the LP5907 circuitry used in our product.

We are flexible in selecting a suitable capacity for the battery, but its space and cost are some of the deciding factors. Basically, we want the lifetime of the product to be at least 10years, so the power consumption of the product has to be extremely lower. Hence I am looking for a very efficient voltage regulator circuitry for our needs. Can you suggest any suitable voltage regulator especially for very low power applications?

Power consumption from 3.7V battery
System State Current consumption including LP5907 (approx) Time duration per day (approx)

LP5907MFX-3.0 (approx)

RF communication 50mA 1min 5mA
Active 15mA 120min (2hrs) 1mA
Idle 10mA 5min 1mA
Sleep 1mA 22hrs 0.2mA

I have another subsequent doubt. We have a road map of our product, where-in we may add another component to the above system that can take current of 500mA for 10sec in a day. Can I get any suitable power efficient voltage regulator suited even for this higher peak current requirement, but with the system taking extremely low power (1mA) for more than 90% of the times? If so, kindly suggest such voltage regulator as well for my reference.

Regards,

Vikas Chandra Rao.

  • Basically, the input and output currents of LP5907 remains same. So, I think it might be confusing to note the LP5907 currents in the above table. Instead, I have noted the power consumption of LP5907 in the updated table below for better understanding. Kindly consider this table for your reference. At the input of LP5907, we have a battery measuring 4V, and the output voltage from LP5907 is 3V. (Changes from above table are highlighted in Red)

    Power consumption from 3.7V rated battery (4V measured)
    System State
    Current consumption including LP5907 (approx)
    Time duration per day (approx)

    LP5907MFX-3.0 power consumption (approx)

    RF communication
    50mA
    1min
    50mW
    Active
    15mA
    120min (2hrs)
    15mW
    Idle
    10mA
    5min
    10mW
    Sleep
    1mA
    22hrs
    1mW
  • Hi Vikas,

    Does your product have a primary battery or rechargeable? 10 years is a very long time for a battery to supply 1 mA, without recharging.
    10 years = 87658 hours. At 1 mA current draw, the required battery capacity to provide 1 mA for 10 years is 87.7 Ah (Amp-Hours). One 18650 Li-Ion cell battery has around 2 mAh capacity (2 Ah), depending on the cell version. To get 10 year life on a single cell, your average current needs to be 22.8 uA.

    To get the Sleep current very low, I suggest disabling the LP5907 or using a load switch to disconnect the RF communication peripheral load from the battery. This is the only way you will meet your 10 year battery life. Texas Instruments has a number of load switches:
    www.ti.com/.../overview.html

    One load switch to look at is TPS22916. I recommend the slow timing C Version to limit inrush current, which will also help battery life: TPS22916C with Quick Output Discharge (QOD) or TPS22916CN without QOD. Both switches have current draw of 10 nA typical, 100 nA max when off. You can use this load switch between the battery and LDO or between the LDO and RF communication peripheral. I would recommend between the battery and LDO, as the on resistance is lower for higher input voltage to the load switch. If you place the load switch between the output of the LDO and the load, the voltage at the load will be lower than the regulator output due to voltage drop across the load switch.

    For the 500 mA load, you can look at these LDOs: TPS7A37 or TLV757P.

    Best,
  • Hi Vikas,

    Adding onto  comment, the LP5912 may also be of interest. It can handle the 500mA load and has an Iq of 30uA which will be beneficial to extending battery life in your application.

    Best,

    Wilson

  • Dear Eric, Wilson,

    I am trying to consolidate my response, based on both your replies.

    We are going to use primary battery only, and hence my requirement for having lower-power-consumption & higher-efficiency of the voltage regulator. We are also making efforts in parallel to reduce the power consumption during various states to a minimal level, so as to have ~10yrs lifetime for the battery.

    Basically, we are powering our electronics through the battery (with LP5907 in between). We have a small micro-controller that controls RF and other tiny peripherals in our product. LP5907 is used to power this micro-controller and the rest of the peripherals. If LP5907 is disabled or Load power is disabled through a load switch, then micro-controller will be in shut-down mode. Then the system will not have any control mechanism to enable LP5907/Load-switch. So, we want to avoid disabling the voltage regulator, or avoid using a load switch. Indeed, we are already powering off the RF and other peripherals for lowering the power consumption. Once my experiments are done, I will know the exact power consumption figures.

    From TI side, I want to get suggestions especially on the low power consumption and high efficient voltage regulator that can source current of 100mA approx. Are there any LDOs with very high power efficiency at lower currents (in the range of few uA)? I am fine in using a switching regulator as well. Kindly suggest whether there are any such regulators.

    Thanks & Regards,
    Vikas Chandra Rao.
  • Hi Vikas,

    The LP5907 datasheet Figure 5, shows the ground current vs output current. When your product is in low power state with 1 mA draw from the battery, the LP5907 ground current is around 10-20 uA. Your microcontroller, RF other peripherals draw the remainder of the battery current, approximately 980-990 uA. You may find an LDO with lower ground/quiescent current, but this will not reduce the current draw from the microcontroller and peripherals. Switching regulators will have higher quiescent current than LDOs, so changing to a switcher will not improve battery life.

    My recommendation is to reduce power consumption in the microcontroller and peripherals. You can achieve this through software, or by adding load switches between the LDO and any peripherals that you can completely power off.

    Best,
  • Hi Eric,

    Thanks for your suggestions and giving insights into LP5907. Actually we are already doing various experiments in parallel to reduce the power consumption of the micro-controller and peripherals to the acceptable limits.

    Regarding LP5907 power consumption at lower currents of ~1mA, it consumes power equivalent to : [Vin*Ignd + (Vin-Vout)*Iout] = [4.2V*20uA + (4.2-3)*1mA] = 1.284mW. Is my calculation correct? Is 1.284mW power consumption the lowest that we can get with any regulator (LDO/switching) with lower output currents (~1mA)? Is there any other regulator that consumes lower power? Kindly suggest. For increasing battery life, I am trying to reduce power consumption of the regulator apart from lowering current consumption of micro-controller/peripherals. Hope you understood my persistence in getting lower power consuming regulator.

    Regards,

    Vikas Chandra Rao.

  • Hi Vikas,

    Your calculation is correct. The LP5907 with 20 uA ground current estimate is 70% efficient at 1 mA output current with 4.2 V input, 3 V output. If you were to find a 1 nA ground current LDO, the efficiency would be 71.4% for the same conditions and it would consume 1.2 mW. I use this example to illustrate that with ground currents in the 10's of uA, the load current is the primary factor in efficiency in an LDO.

    You could look at low output current, high efficiency switchers like the TPS62122. Using this kind of power device will give you a greater efficiency improvement over an LDO. Make sure sensitivity/range of the RF peripheral is acceptable when using a switcher for power.

    Best,
  • Hi Eric,

    Thanks for suggesting TPS62122. I am thinking about increasing RF power later on based on RF range testing. If I have to increase my RF power so as to get more RF range, then TPS62122 (with 75mA) may not be sufficient. In that case, can I get exact equivalent of TPS62122 with 250-300mA output current rating? Or should I go for some other regulator like TPS62740 now itself? Kindly suggest.

    Regards,
    Vikas Chandra Rao.
  • Hi Vikas,

    Please post your question to the switching regulator group at:
    e2e.ti.com/.../

    They are more familiar with the latest TI switching regulators and can help you find the best IC for your application.

    Best,