More power to your battery
- Autor:Ella Cai
- Solte em:2017-07-13
How complicated can a boost converter be? Meng He explains how the latest power converters can help make the most of your batteries.
The typical specs that designers look for while selecting a boost converter include: input and output voltages, input current limit that dictates the output current level, switching frequency, and peak efficiency.
So how complicated can a boost converter be? If it simply boosts voltage up from one level to another, how many magic tricks can it hide up its sleeve to help solve complicated system design problems?
Doubling up
A boost converter can allow multiple power sources. Some ultra-low power applications, such as remote sensors, rely on multiple power sources; however, not every boost is optimised for these types of applications.
Imagine an IoT sensor that is powered from a long-life primary cell battery, an RF energy harvester, USB for field diagnostics, and even a solar panel for long life (Figure 1, above). These high-impedance and low-voltage power sources typically need a boost converter to regulate the voltage to a usable level. In this case, reverse current leakage from the output to the input or output to ground is highly undesirable for the disabled boost sources.
Zero output leakage current is the goal in multiple source applications, and there are several reasons why a typical boost converter can’t be used here. The first is that some boost converters discharge the load on purpose while in shutdown, to implement a feature called active load discharge. The second reason is that most boost converters do not have True Shutdown mode, which is a true bidirectional disconnect from the output to input. Last, traditional adjustable boost converters use a feedback resistor string to set the output voltage.
Unfortunately, the benefit those resistors provide in output adjustability creates a detriment in terms of reverse leakage current in this case.
Other options include fixed-output voltage boost converters that do not use external feedback resistors, although in reality, all they do is pull the resistors inside the chip and, unless they disconnect the internal feedback resistors on shutdown, there will be reverse current leakage from the output to ground through those internal resistors. Maxim’s latest NanoPower boost converter, MAX17222, has zero reverse current because it features True Shutdown mode, does not implement active load discharge, and disconnects all feedback resistors during shutdown.
Converter can have zero current
Startup Circuit diagram
Figure 2: A start-up circuit with zero current in the off state and flexible feature toggling
A boost converter can have a zero current off state and still have a flexible startup (Figure 2). To turn an end product on and off, everyone knows that a simple single pole single throw (SPST) mechanical on/off switch can be used. But equipment makers also know that customers have user interface preferences for this function, like the use of a momentary push-button instead of an SPST switch. In addition, OEMs often want customers to be able to select multiple features of their product by holding down a push button for different durations of time.
Take a BLE temperature sensor, for example. Its power source is a primary cell battery, and a boost converter is used to regulate the output voltage that powers up the rest of the system. You can push the button once to turn on the system, and push it again to turn it off. The button can disconnect the battery and the boost converter; therefore, there is no current leakage when the temperature sensor is not in use.
The problems come in when false start-up happens, if the user should just accidentally push or touch the button. The jittering from the mechanical switch can then cause the boost converter to start waking up the central command of the system, which is typically an MCU, and the whole system would start to drain the battery power. To overcome this issue, designers typically require a longer push time to turn on the product. To do this the MCU needs to be programmed to recognise the real start-up condition, and turn off the boost converter if false start-up happens.
The same can apply to turning off the product, and the microcontroller can either turn off after a three-second press once the microcontroller is already on, or react to a different duration press.
This type of implementation can, for example, be used to enable different button-press duration features, such as a low-power temperature readback versus a fast, continuous temperature readback. In this use case, a hard shutdown to disconnect the battery from the boost converter by an SPST switch would not work. The start-up circuit shown in Figure 2 allows a zero current off state while still giving flexibility by allowing different features to be enabled using the same push button pressed for different durations.
More power for less battery
A boost converter can extend the battery life and shrink product size.
The consumer internet of things (IoT) is growing at 33% CAGR (2017 to 2021), according to a Gartner forecast. With all of the buzz that it has generated in the past five years we are seeing an increasing number of battery‑powered consumer devices in use, from home automation systems to wearables.
The designers of these products all demand long battery life. In many cases, the end product size is driven by the battery size, or equipment makers are filling any spare space in the product with the power source to increase the running time. A boost converter with ultra-low quiescent current can benefit these designs. Smaller products need smaller batteries.
Take a small-size, low-capacity silver-oxide battery with a 16mAH capacity, for example. With a tiny battery like this, we need a boost converter with the matching nA quiescent current to truly deliver unmatched battery life. Quiescent current is critical because most of these products are awakened occasionally by the sensor and pulse communication to the hub through wireless protocols.
Converter cuts failure rate
Fresh batteries
Figure 3: VIN and VOUT during fresh battery replacement
VLX(max)=2x(VBAT–0.7V)=2x 4.2V–0.7V)=7V
A boost converter with True Shutdown and soft-start capabilities will not let this happen. In this converter, when the battery is connected, the body diode disconnect switch prevents the output from ringing high. As the boost regulator turns on, the body switch is first allowed to soft-start, ramping the OUT voltage to VIN minus a diode drop. After the switch soft-start is complete, the boost regulator continues in soft-start, ramping the OUT pin until it achieves regulation. The output is allowed with enough ramping time to regulate to the desired output level instead of damaging the power circuit itself, or the rest of the system.
When you put a step function into an LC filter, you get ringing as the energy bounces back and forth between the inductor and capacitor. The resonant frequency is determined by the values of LX1 and COUT, where the bigger their values, the slower the frequency.
A boost converter can reduce inventory management cost. To stock one part that covers different voltage rails in the system, vendors typically provide the resistor feedback string to allow designers to customise the output. It requires two resistors and it drains current whether the system is in shutdown, active, or standby mode.
MAX1722x diagram
Figure 4: RSel implementation
The MAX17222 boost converter has a single-resistor output selection method known as RSel (Figure 4). At start-up, the part uses up to 200μA to read the resistor value, and it typically takes 600μs. After reading the value, the Sel pin does not drain any current during operation. It carefully protects the current leakage from the battery. Its benefits include lower cost and smaller size, since only one resistor is needed.
RSel eliminates continuous waste of current through feedback resistors for ultra-low power battery-operated products. A power supply is the beating heart of the system, bringing energy to every functional block. A carefully designed power system can minimise the end product size, increase the battery life, and reduce the failure rate.
Case study: A small 16mAH sliver oxide battery, by the numbers
A small silver-oxide battery with a nanoPower boost converter (MAX17222) that features 300nA quiescent current can achieve:
Three years (26,280 hours) shelf life using a 33MΩ pull-up resistor on EN while in True Shutdown mode .
10% battery self-discharge for three years results in 16mAH–4.8mAh=11.2mAH from just self-discharge alone. In addition, total system shutdown is 71nA. 71nAx26,280hrs=1.86mAH; therefore, after three years on the shelf in True Shutdown mode, 9.34mAH (11.2mAH–1.86mAH=9.34mAH) will remain.
When enabled, how long can the system run? If for every 3,997ms the circuit averages 10µA and for 3ms it averages 5mA, then the continuous average load current is (3,997ms/4,000ms)x10µA+(3ms/4,000ms)x5mA=13.75µA c.
Total continuous average battery current = [(13.75µA+300nA)/(eff x (1-D))] continuous average battery current, 1-D=VIN/VOUT. In the use case where VIN=2.4V, VOUT=3.3V, and the conversion efficiency is 90%, the continuous average battery current = 57µA. 9.34mAH/57µA=163hrs (1 day/24hrs) = 7 continuous days of operation.
A step-up switching power converter, also known as a boost converter, is a class of switched-mode power supply. The output voltage is regulated, as long as the power draw is within the output power specification of the circuit and generally when the input voltage is lower than the output voltage.
The typical specs that designers look for while selecting a boost converter include: input and output voltages, input current limit that dictates the output current level, switching frequency, and peak efficiency.
So how complicated can a boost converter be? If it simply boosts voltage up from one level to another, how many magic tricks can it hide up its sleeve to help solve complicated system design problems?
Doubling up
A boost converter can allow multiple power sources. Some ultra-low power applications, such as remote sensors, rely on multiple power sources; however, not every boost is optimised for these types of applications.
Imagine an IoT sensor that is powered from a long-life primary cell battery, an RF energy harvester, USB for field diagnostics, and even a solar panel for long life (Figure 1, above). These high-impedance and low-voltage power sources typically need a boost converter to regulate the voltage to a usable level. In this case, reverse current leakage from the output to the input or output to ground is highly undesirable for the disabled boost sources.
Zero output leakage current is the goal in multiple source applications, and there are several reasons why a typical boost converter can’t be used here. The first is that some boost converters discharge the load on purpose while in shutdown, to implement a feature called active load discharge. The second reason is that most boost converters do not have True Shutdown mode, which is a true bidirectional disconnect from the output to input. Last, traditional adjustable boost converters use a feedback resistor string to set the output voltage.
Unfortunately, the benefit those resistors provide in output adjustability creates a detriment in terms of reverse leakage current in this case.
Other options include fixed-output voltage boost converters that do not use external feedback resistors, although in reality, all they do is pull the resistors inside the chip and, unless they disconnect the internal feedback resistors on shutdown, there will be reverse current leakage from the output to ground through those internal resistors. Maxim’s latest NanoPower boost converter, MAX17222, has zero reverse current because it features True Shutdown mode, does not implement active load discharge, and disconnects all feedback resistors during shutdown.
Converter can have zero current
Startup Circuit diagram
Figure 2: A start-up circuit with zero current in the off state and flexible feature toggling
A boost converter can have a zero current off state and still have a flexible startup (Figure 2). To turn an end product on and off, everyone knows that a simple single pole single throw (SPST) mechanical on/off switch can be used. But equipment makers also know that customers have user interface preferences for this function, like the use of a momentary push-button instead of an SPST switch. In addition, OEMs often want customers to be able to select multiple features of their product by holding down a push button for different durations of time.
Take a BLE temperature sensor, for example. Its power source is a primary cell battery, and a boost converter is used to regulate the output voltage that powers up the rest of the system. You can push the button once to turn on the system, and push it again to turn it off. The button can disconnect the battery and the boost converter; therefore, there is no current leakage when the temperature sensor is not in use.
The problems come in when false start-up happens, if the user should just accidentally push or touch the button. The jittering from the mechanical switch can then cause the boost converter to start waking up the central command of the system, which is typically an MCU, and the whole system would start to drain the battery power. To overcome this issue, designers typically require a longer push time to turn on the product. To do this the MCU needs to be programmed to recognise the real start-up condition, and turn off the boost converter if false start-up happens.
The same can apply to turning off the product, and the microcontroller can either turn off after a three-second press once the microcontroller is already on, or react to a different duration press.
This type of implementation can, for example, be used to enable different button-press duration features, such as a low-power temperature readback versus a fast, continuous temperature readback. In this use case, a hard shutdown to disconnect the battery from the boost converter by an SPST switch would not work. The start-up circuit shown in Figure 2 allows a zero current off state while still giving flexibility by allowing different features to be enabled using the same push button pressed for different durations.
More power for less battery
A boost converter can extend the battery life and shrink product size.
The consumer internet of things (IoT) is growing at 33% CAGR (2017 to 2021), according to a Gartner forecast. With all of the buzz that it has generated in the past five years we are seeing an increasing number of battery‑powered consumer devices in use, from home automation systems to wearables.
The designers of these products all demand long battery life. In many cases, the end product size is driven by the battery size, or equipment makers are filling any spare space in the product with the power source to increase the running time. A boost converter with ultra-low quiescent current can benefit these designs. Smaller products need smaller batteries.
Take a small-size, low-capacity silver-oxide battery with a 16mAH capacity, for example. With a tiny battery like this, we need a boost converter with the matching nA quiescent current to truly deliver unmatched battery life. Quiescent current is critical because most of these products are awakened occasionally by the sensor and pulse communication to the hub through wireless protocols.
A NanoPower boost converter can allow a 16mAH silver-oxide battery to support three years of shelf life and four weeks of continuous days of normal operation (see case study above). With these products, tiny low-voltage and high-impedance batteries such as silver oxide, AA, AAA, and zinc air can find homes in products that challenge conventional battery life.
Converter cuts failure rate
Fresh batteries
Figure 3: VIN and VOUT during fresh battery replacement
A boost converter can reduce the field failure rate when connected to the source for the first time (Figure 3). Many systems run off secondary cells. Before the battery is plugged in, the input and output voltages are both at 0V. When a fresh battery is plugged in, the IN voltage goes immediately to the battery voltage, and the other side of the inductor, LX, will ring high with a voltage that can be up to twice the battery voltage minus the body diode drop. With a 4.2V battery, this can be up to 7V, which could damage the Nmosfet from LX to GND (Figure 3).
VLX(max)=2x(VBAT–0.7V)=2x 4.2V–0.7V)=7V
A boost converter with True Shutdown and soft-start capabilities will not let this happen. In this converter, when the battery is connected, the body diode disconnect switch prevents the output from ringing high. As the boost regulator turns on, the body switch is first allowed to soft-start, ramping the OUT voltage to VIN minus a diode drop. After the switch soft-start is complete, the boost regulator continues in soft-start, ramping the OUT pin until it achieves regulation. The output is allowed with enough ramping time to regulate to the desired output level instead of damaging the power circuit itself, or the rest of the system.
When you put a step function into an LC filter, you get ringing as the energy bounces back and forth between the inductor and capacitor. The resonant frequency is determined by the values of LX1 and COUT, where the bigger their values, the slower the frequency.
A boost converter can reduce inventory management cost. To stock one part that covers different voltage rails in the system, vendors typically provide the resistor feedback string to allow designers to customise the output. It requires two resistors and it drains current whether the system is in shutdown, active, or standby mode.
MAX1722x diagram
Figure 4: RSel implementation
The MAX17222 boost converter has a single-resistor output selection method known as RSel (Figure 4). At start-up, the part uses up to 200μA to read the resistor value, and it typically takes 600μs. After reading the value, the Sel pin does not drain any current during operation. It carefully protects the current leakage from the battery. Its benefits include lower cost and smaller size, since only one resistor is needed.
RSel eliminates continuous waste of current through feedback resistors for ultra-low power battery-operated products. A power supply is the beating heart of the system, bringing energy to every functional block. A carefully designed power system can minimise the end product size, increase the battery life, and reduce the failure rate.
Case study: A small 16mAH sliver oxide battery, by the numbers
A small silver-oxide battery with a nanoPower boost converter (MAX17222) that features 300nA quiescent current can achieve:
Three years (26,280 hours) shelf life using a 33MΩ pull-up resistor on EN while in True Shutdown mode .
10% battery self-discharge for three years results in 16mAH–4.8mAh=11.2mAH from just self-discharge alone. In addition, total system shutdown is 71nA. 71nAx26,280hrs=1.86mAH; therefore, after three years on the shelf in True Shutdown mode, 9.34mAH (11.2mAH–1.86mAH=9.34mAH) will remain.
When enabled, how long can the system run? If for every 3,997ms the circuit averages 10µA and for 3ms it averages 5mA, then the continuous average load current is (3,997ms/4,000ms)x10µA+(3ms/4,000ms)x5mA=13.75µA c.
Total continuous average battery current = [(13.75µA+300nA)/(eff x (1-D))] continuous average battery current, 1-D=VIN/VOUT. In the use case where VIN=2.4V, VOUT=3.3V, and the conversion efficiency is 90%, the continuous average battery current = 57µA. 9.34mAH/57µA=163hrs (1 day/24hrs) = 7 continuous days of operation.
