Maximizing Battery Life in Medical Wearables

By John Varela Munoz, Systems Manager, Medical Division, Systems Engineering and Marketing, Texas Instruments

Rogelio Armino, Systems Engineer, Systems Engineering and Marketing, Texas Instruments Medical Division

This article will discuss the following questions: How advances at the chip level can extend the battery life of medical wearables. What are the trade-offs between different battery chemistries in these systems. Considerations related to transitioning from standby mode to active mode and back.

The market for wearable medical devices is expanding rapidly. Not only are more and more devices being approved by government standards to diagnose disease and monitor key vital signs, but more and more users are interested in more personal data to improve their lifestyles. The ability to track metabolic performance, stress, and sleep quality are features that are particularly popular with consumers.

Manufacturers of wearable medical devices strive to achieve smaller size, lower weight, longer operating time, and smarter feature sets. However, these features often outpace battery advances, so designers must come up with new ways to use batteries more efficiently, thereby improving overall functionality.

Regardless of the power topology, the challenges remain—wearable medical devices need to:

- Low quiescent current (Iq) and shutdown function to extend its shelf life.

- High efficiency to extend its service life in active mode.

- Ability to tolerate and manage dynamic fast transient loads common in radio transmissions.

Low-power microcontrollers (MCUs), edge artificial intelligence (AI) processing, and analog integrated circuits (ICs) are available, but it is not always possible to take advantage of these technologies in your design without optimizing power management. It is important to choose the right power architecture for your application to improve efficiency and extend battery run time.

In this article, we will highlight different power schemes and new technologies in load switches, DC-DC converters, and battery chargers that can maximize the battery life of disposable and reusable wearable medical devices, both in shelf/transport mode and in active mode.

Types of Batteries Used in Medical Wearable Devices

Devices such as heart rate monitors, multi-parameter patches, blood glucose monitors, blood pressure monitors, pulse oximeters, fitness monitors, activity monitors, and drug delivery patches can all be portable and wearable. Many of these devices are disposable or use batteries that need to be replaced. In addition, current wearable medical devices can connect to an increasing number of smart devices and support multiple protocols, resulting in higher power consumption.

Lithium manganese dioxide (LiMnO2), alkaline, and lithium-ion (Li-ion) batteries have been the mainstay of wearable devices due to their high energy density, long life, and rechargeability. But recent advances in new chemistries are enabling new possibilities and applications. Silver-zinc and zinc-air batteries enable longer active-mode capacity, while silver-oxide batteries have a low self-discharge rate of about 10% per year or less, extending shelf life.

There are also new developments in rechargeable battery technology. Lithium polymer (LiPo) batteries enable flexible battery design and can flexibly change the module shape to adapt to wearable devices. Solid-state batteries are becoming cheaper and can be applied to wearable devices, with high energy density and flexible form factors without safety issues. Nickel metal hydride (NiMH) remains popular because it can be used continuously, has a long service life, and is relatively low in cost. Low discharge rate NiMH batteries are also available.

One trend toward higher energy density battery chemistries is toward lower battery voltages (Table 1), which requires power solutions to operate at higher efficiencies at peak loads. Unfortunately, high peak currents for functions such as RF data transmission or starting a motor can cause the battery voltage to drop enough to trigger a voltage sag in critical ICs.

Optimizing Active and Standby Mode Transitions in Wearable Medical Devices

In active mode, medical patches (ECG, temperature, and blood sugar) typically spend a short time measuring, processing, or preprocessing data to send to a remote terminal, and then return to sleep. In this case, the load on the battery quickly changes from hundreds of nanoamps to tens of milliamps in a taste sensing patch, or even higher currents for drug delivery patches with motors and pumps.

Figure 1 shows this situation, where a step load in the lower waveform causes a large voltage transient in the main supply line, as shown in the upper waveform. The main challenge is that while the system is able to operate efficiently under different loads, it must also handle transients as fast as a few microseconds.

Consider the example in Figure 2, which illustrates the use of a Texas Instruments (TI) CC2340R5 Bluetooth low energy (LE) MCU and a low-voltage battery chemistry in a wearable application. Because the input voltage to a Bluetooth low energy device is as low as 1.7 V, you can use a boost converter such as TI's TPS61299.

Typically, when a Bluetooth LE device enters a connection event from sleep mode, the load current increases very quickly and causes a large voltage drop in the main supply rail. This is especially true when using battery chemistries with high equivalent series resistance and equivalent series inductance. To maintain these drops and not allow parts of the circuit to enter a voltage drop, the TPS61299 uses a fast mode detection control loop that stabilizes the supply voltage and reduces the settling time of a typical Bluetooth LE load to 8 µs.

In applications such as insulin pumps and drug delivery patches, load transients from dosing pumps may be too high for traditional compensation methods. In this case, large supercapacitors can quickly provide the energy needed for high-drain components. TI's TPS61094 is a 60 nA IQ buck-boost converter with supercapacitor management that seamlessly transitions between battery and supercapacitor power sources.

In addition to addressing transient response challenges, achieving high efficiency across current loads is essential. TI's DC-DC switching regulator portfolio has several devices designed to address the specific challenges found in wearable medical devices.

When the input voltage closely matches the selected output, pass-through mode allows buck, boost, and buck-boost devices to connect the load directly to the battery for improved operating efficiency. As the voltage drops, the device enters active mode with multiple operating modes. As the load changes from high to low, intelligent switching between pulse width modulation (PWM), pulse frequency mode (PFM), and burst mode helps maintain overall system efficiency above 85%, covering microamps to hundreds of milliamps.

In addition to automatically switching between PWM and PFM, TI's TPS62840 step-down converter features a 100% PWM mode that consumes only 120 nA IQ. This allows the system to maintain maximum efficiency even when the battery is running low.

Another way to address transient issues is to improve efficiency in both low- and high-current modes. TI’s TPS63900 buck-boost converter has two different programmable voltages. For example, when the CC2340R5 is about to wake up, the device can boost to 3 V, allowing the radio to operate more efficiently. Then, when only a few IC modules are turned on, the converter can drop to 1.8 V in standby mode.

Table 2 lists the devices mentioned in this section and their unique current consumption in different modes.

Extend battery life with low-power ship mode and smart load switching

Most wearable medical devices are not active when they are packaged, as it may be weeks or months before they reach the user (or patient). Dedicated low-power modes help extend battery life while the product is in transit from manufacturer to consumer.

In the so-called “shipping mode”, the devices are in a high impedance state and current leakage is effectively suppressed. Only a few devices with very low Iq remain powered in shipping mode to detect when the devices are enabled.

Additionally, a ship mode with minimal shutdown current helps maximize available power when the product is ultimately put into active use. This becomes especially important if the device is designed to run on disposable batteries.

Consider the example of a wireless patch using a 1.5 V silver oxide battery with a capacity of 150 mAh. To last for a year in an inactive state, the wearable medical device can use a load switch to minimize current consumption. If the product’s system components consume approximately 1 µA when disabled, this consumes at least 8.76 mAh of the 150 mAh budget (1 µA × 24 hours × 365 days).

In other words, inefficiencies can result in a total battery loss of approximately 5.84% (8.76 mAh/150 mAh). These inefficiencies can stem from losses such as minimum and maximum specifications due to temperature variations, power component efficiency losses, and radio module leakage.

Devices such as load switches, which feature off-state currents as low as nanoamps, can help save power in the long term. Load switches act as electrically controlled switches that can significantly reduce supply current when a module does not need to be active (Figure 3).

A load switch such as the TPS22916 (0.8 × 0.8 mm) can reduce shutdown leakage current to 10 nA, which has little impact (0.058%) on battery life when the system is disabled (Table 3).

If there are multiple independently powered modules in a single product, multiple load switches can be used to disable loads and modules. This improves the ability of the power architecture to tightly manage power consumption as needed.