Robert Huntley, for Mouser Electronics
Ultra-portable wearable monitoring devices allow patients with long-term and chronic illnesses to integrate their monitoring with their daily lives, allowing a high degree of care to be delivered with more convenience than ever before. However, powering these devices for long periods without resorting to a bulky battery is presenting designers with a challenge.
As with most ‘tech' devices, every new generation of wearable device incorporates more features and functions than the previous generation. This often compounds the power challenge, as more functionality will require more power. It is not possible to increase the battery size as this will increase the overall device size and weight, which is inconvenient for all users and a significant burden on elderly patients or infants that may be using the device. Also unacceptable is reducing the time between charges, leaving the only remaining option as finding smarter ways to manage the power being used within the device.
Wearables: a full system design
Even though they are physically small, wearable medical devices are fully functioning systems that contain many essential elements. A typical device will comprise:
· A Microcontroller Unit (MCU), to run the code/firmware as well as managing and processing data
· A rechargeable battery that provides the energy to allow the device to function
· Sensors to collect data on the physical parameters to be monitored – this may be a pulse meter or glucose monitor as well as general sensors such as an accelerometer or gyroscope
· A wireless communication interface – this will typically be a common, low energy protocol such as Bluetooth® Low Energy (BLE) or Near Field Communication (NFC).
· Security provisions – this will be a combination of hardware and software and will encrypt all transmissions as well as preventing malicious interference with the firmware.
All medical monitoring devices have a specific purpose, even if it is a general-purpose fitness band that monitors vital signs. The purpose defines the choice of MCUs and sensors, and especially parameters such as accuracy, reliability, and repeatability of measurement. The length of measurement possible on a single battery charge will also impact component selection, in some cases requiring ultra-low power consumption devices. The MCU is the heart of the system and integrates all of the peripheral elements including sensors and other devices. In many cases, the architecture organizes peripherals into domains or groups that can be powered down when not needed – for example, the RF section is only required when data is being transmitted, or a particular sensor may just take one reading per minute and be powered down during the intervening time.
Battery technology has developed in recent years, with existing chemistries being improved and new ones being released. The most common type used in wearables is Lithium-Ion (Li-Ion) that produces voltages between 3.2V and 4V from a single sell. Although Lithium-ion Polymer (LiPo) batteries are used in some wearables, Li-Ion is preferred due to the higher storage capacity and the fact that they are relatively kind to the environment.
However, the laws of physics still apply, and there is a strong link between the storage capacity of a Li-Ion battery and its physical size, which means that in the most space-constrained designs (such as wearables) there is relatively little energy available to power the device for long periods. Research is continuing into new materials and technologies to improve this situation, and multiple studies indicate that graphene is likely to improve battery capacity per unit volume vastly. Another avenue of research is supercapacitors that are benefitting from general advances in nanotechnology.
Ultra-Low consumption MCUs
To design a modern wearable device, designers are selecting the lowest power MCUs possible. Currently, the state-of-the-art is considered to be an MCU that consumes significantly less than 1mA when operating and just a few nanoamperes (nA) when in sleep mode. Apart from its own consumption, the MCU has a crucial role in overall power consumption as it controls the energy supply to peripherals, ensuring that none of the valuable battery capacity is wasted.
Maxim Integrated’s MAX32660 MCU balances performance and efficiency making it an ideal choice for wearable device design. Built around a 32-bit ARM Cortex-M4 core, the MCU includes a Floating-Point Unit (FPU) processor as well as peripheral management for sensors and other devices. The MAX32660 can control external memory devices, allowing advanced processing algorithms to be developed and run. In terms of power performance, the device leads the industry requiring just 50μW of power for each megahertz of clock speed. Physically, the device measures 1.6mm x 1.6mm (WLP package) allowing it to fit within a space-constrained wearable device easily.
Another supplier offering a variety of ultra-low power 32-bit MCUs aimed at medical wearables is Microchip. Its SAM series comprises small SAM D MCUs, based around ARM Cortex-M0+ technology as well as high-performance PIC32MX XLP based MCUs and ultra-low power devices in the SAM L range. Drawing just 200nA when in sleep mode, these energy efficient devices require less than 35µA/MHz when in active mode. While they are small and frugal with energy, they are not short on features and include LCD ports, operational amplifiers, real-time clocks, and mTouch sensing, plus USB and Direct Memory Access (DMA) interfaces.
The Silicon Labs EFM32 Giant Gecko ARM Cortex-M3 are another 32-bit MCU for medical wearables. These devices include autonomous low-energy peripherals, such as AES encryption for increased security, a UART for communications, a reduced power sensor interface, and integrated operational amplifiers.
Click image to enlarge
Figure 2: EFM32 block diagram. (Source: Silicon Labs)
Removing reliance on batteries
Battery technology can be a challenge in medical applications as the patient may forget to charge the battery regularly, meaning that monitoring is missed. In some cases, a caregiver may be given responsibility for charging, but this becomes a further burden on the system. For these reasons, the medical profession has adopted battery-based medical monitors slowly, leading manufacturers to look at alternate ways of powering them.
Energy harvesting does not rely on stored energy in the form of a battery but generates energy from sources such as light or heat from the sun or movement of the patient as they move around wearing the device. With sufficient light/heat/movement, this approach can give an almost infinite source of power, allowing the wearable to operate indefinitely without any intervention.
The average person dissipates 107J/day as a mixture of heat and movement, meaning that there is easily enough energy available to power a small wearable device, provided the movement of heat can be converted into electrical power.
The principle behind converting heat into electricity is the Seebeck effect that generates a voltage based on the thermal difference between two points. In the case of a wearable, this would be the face touching the patient (warm) and ambient (cooler). The technology used for this conversion would be a semiconductor-based Peltier cell that offers the advantage of being able to produce energy 24/7, unlike solar PV, which works less well indoors, and not at all at night.
As the patient moves, so they generate energy, and this mechanical movement can be translated into electrical energy through the use of piezoelectric elements that produce a small current in response to each movement, such as walking or exercising. Wurth Electronics offers an Energy Harvesting Solution to Go development kit that gives developers an easy way to get started in the world of energy harvesting.
Regulating on-board voltages
Wearable devices commonly have an onboard DC/DC converter that ensures that the voltage supplied to all devices in the system is constant, even if the source is fluctuating. Typically controlled by the MCU, sophisticated DC/DC converters will manage all of the energy in the device, ensuring that it is used as efficiently as possible – including not consuming energy within the DC/DC converter itself.
One highly integrated DC/DC converter solution is the LTC3107 from Linear Technology. This device is specially designed with energy harvesting mechanisms in mind and is extremely frugal with energy. Using a combination of energy harvesting from a thermal source and a battery, the device can significantly extend battery life, thereby reducing the costs and inconvenience associated with replacement.
Click image to enlarge
Figure 3: LTC3107 typical application circuit. (Source: Linear Technology)
Wearable medical devices continue to offer more features while getting smaller, requiring innovative approaches to power management. While using battery energy better is an obvious approach, energy harvesting can provide many benefits including high levels of convenience, especially when well-managed by an advanced DC/DC converter.