The MAXM86146 is a sensor hub with embedded firmware and algorithms for wearables that seamlessly enables customer-desired on-chip processing. This highly integrated sensor hub with embedded algorithms simplifies the design-in process by delivering raw or calculated data to the outside world in an ultra-small form factor. This is achieved by minimizing the overall system power consumption while maintaining enough SNR to allow a large variety of users and body locations to be measured quickly and accurately.
Wearable health and fitness solutions are increasingly space and power constrained. These optical solutions need to maintain proper LED and PD separation distances to maximize the optical PPG signals received in order to calculate heart rate and SpO2. The longer wavelengths needed for SpO2 travel longer paths compared to the shorter path and wavelength that HRM requires. The MAXM86146 uses accelerometer data to compensate for periodic motion artifacts and is optimized for two simultaneous green PPG signals. These two green PPG signals can be achieved by either 2 x PD / 1 x LED or 1 x PD / 2 x LED. SpO2 requires the use of two different LED wavelengths, namely red and IR. They need to share the same photodiode and separation distance. The MAXM86146 can be configured for simultaneous HRM and SpO2 measurements with the addition of an external DPDT MUX, such as the MAX14689, for its low noise performance. If only HRM is desired, the MAXM86146 has the option to be configured to use an external photodiode, allowing for a single green LED to be used. This versatility makes it easy to integrate optical heart-rate monitoring and SpO2 measurements into any wearable application by offering the flexibility to support multiple optical configurations, and on-chip processing.
The device connects to a microcontroller host through a fast-mode slave I2C interface, allows for in-field updates, and provides access to raw data and processed algorithm calculations. With the addition of external LEDs and an accelerometer (either direct-connected or host-connected is required), a complete system is achieved reducing time-to-market and simplifying design. Additionally, the Maxim Integrated biometric sensor hub provides additional features such as industry leading ambient light rejection, higher signal-to-noise (SNR) ratio compared to the competition, and enters ultra-low-power deep sleep mode between samples. The MAXM86146 is shipped with bootloader software that accepts in-application programming of Maxim proprietary application code, and is designed to be used only with this application code. The application code consists of advanced algorithms, AFE controller, and an accelerometer controller. The latest application code can be downloaded from the MAXM86146 product page under the Design Resource tab.
The interface to the host is the I2C interface. It is a bidirectional, two-wire serial bus that provides a versatile medium-speed communications network. It can operate as a one-to-one, one-to-many, or many-to-many communication mediums. These engines support fast-mode I2C speeds. The devices support one slave interface (address 0x55—refer to the MAX32664 User Guide). The features for this interface are:
- One slave for communication with a host
- Supports standard 7-bit addressing or 10-bit addressing
- Restart conditions
- Interactive receive mode
- Tx FIFO preloading
- Support for clock stretching to allow slower slave devices to operate on higher speed buses
- Fast mode: 400kbps
- Internal filter to reject noise spikes
- Receiver FIFO depth of 8 bytes
- Transmitter FIFO depth of 8 bytes
The MAXM86146 ships with a bootloader that is ready to be programmed with the latest application code over the I2C communication bus. The application code contains the firmware and algorithms (version C), which controls the AFE, allowing simple communication through I2C commands. The MAX32664 bootloader firmware supports in-application programming (IAP). The MAX32664 is a variant of the MAX32660. Refer to the latest MAX32660 Bootloader User Guide for the complete programming procedure that can also be found at the MAXM86146 product page under the Design Resource tab.
For I2C programming, the following pins are required:
- RSTN
- P0.1 (MFIO)
- P0.8 (SCL)
- P0.9 (SDA)
The firmware of the MAXM86146 supports the accelerometer in Table 1 connected directly to the SPI bus. Pin-to-pin connections between MAXM86146 and the recommended accelerometer are given in Table 2. If a host-connected Accelerometer is desired, refer to the latest MAX32664 User Guide. For more detailed descriptions and requirements, refer to the MAXM86146 product page under the Design Resources tab.
MANUFACTURER | PART NUMBER | DESCRIPTION | PACKAGE |
Kionix | KX122-1037 | ±2g/±4g/±8g Tri-axis Digital Accelerometer | 2mm x 2mm x 0.9mm LGA |
MAXM86146 PIN (DESCRIPTION) | KIONIX KX122 PIN |
P0.0 (ACCEL_CS) | CS |
P0.2 (ACCEL_INT) | INT |
P0.4 (MISO) | SDO |
P0.5 (MOSI) | SDI |
P0.6 (SCK) | SCLK |
Keeping the ripple on the VLED line as low as possible ensures the highest SNR is reached. The VLED power supply should not exceed a peak-to-peak ripple of 30mV, and the switching frequency should stay between 100kHz to 3MHz. Together with a good load transient response, where the ripple tends to be the highest, high SNR can be maintained at the heaviest loads.
The MAX20345 PMIC with Ultra-Low IQ Voltage Regulators, Buck-Boost for Optical Sensing and Charger for Small Lithium Ion Systems or the smaller MAX20343 Ultra-Low Quiescent Current Low Noise 3.5W Buck-Boost Regulator, are recommended solutions for this application. They both offer a highly efficient buck-boost regulator with very small load ripple, fast load transient responses, and have load pulse consistencies that together, enable > 90dB SNR (white card loop-back test). For the latest PMIC and switching regulator products, visit: https://www.maximintegrated.com/en/products/power/power-management-ics
The MAXM86146 integrates three precision LED-driver-current DACs that modulate LED pulses for a variety of optical measurements. The LED current DACs have 8 bits of dynamic range with three programmable full-scale ranges of 124mA, each. The LED drivers are low dropout current sources allowing for low-noise, power-supply-independent LED currents to be sourced at the lowest supply voltage possible, minimizing LED power consumption. The LED pulse width can be programmed from 14.8μs to 117.3μs, allowing the algorithms to optimize SpO2 and HR accuracy at the lowest dynamic power consumption dictated by the application. The typical VLED voltage of 5.0V for the recommended LEDs in Table 3 are selected for their low forward voltage, high light output, radiation pattern and specific peak wavelength, enabling exceptional WSpO2 and WHRM performance.
USE CASE | MANUFACTURER | PART NUMBER | DESCRIPTION | LED |
WHRM | OSRAM® | CT DBLP31.12 | FIREFLY® E2218 | GREEN |
WSPO2 | OSRAM | SFH 7015 | CHIPLED® SFH 7015 | RED/GREEN/IR |
CHIPLED, FIREFLY, and OSRAM are registered EU trademarks of OSRAM GmbH.
The VLED voltage must consider the forward voltage (VF) of the LEDs driven at the 132mA current. The sum of VF of the LED at ILED = 132mA and the 900mV LED driver headroom must be considered. Additionally, the AFE requires a minimum of 3.1V applied to the VLED pin.
The VLED voltage must be above this minimum to avoid compression, and allow for enough headroom to supply the current of 124mA (typical) and 132mA (max); otherwise, the ILED is reduced and sensitive to VLED supply changes. LEDx_DRV pins can be measured during an exposure to ensure the 900mV minimum voltage is maintained during each LED exposure. This is simplified using the formula:
where, VF is a function of ILED when driven with 134mA (max). Alternatively, LED should be selected so that: .
VLED bypass capacitor selection criteria requires careful consideration of effective capacitance of a capacitor at an applied DC voltage. The typical 0603 sized (0.6mm x 0.3mm) 10μF capacitor rated for 16V typically derates to ≤ 4μF effective capacitance at a 5.0V DC-bias voltage. The applied DC voltage, with respect to the DC rating of a capacitor with a small physical size, reduces the actual (effective) capacitance that the component can supply. Therefore, a minimum of 4μF of effective capacitance is recommended for all VLED bypass capacitors at any package size. All ceramic capacitors with high dielectric constants (X5R, X7R) derate significantly; therefore, derating must be considered when selecting a VLED bypass capacitors at reduced footprint sizes. To aid in the selection process and when considering derating as well as minimum PCB footprint, the following 22μF 0402 X5R 6.3V GRM155R60J226ME11 capacitor manufactured by muRata® should be used. This component is specifically recommended for use on this design as it maintains the similar performance as its larger counterparts at a smaller package size.
muRata is a registered trademark of Murata Manufacturing Co., Ltd. Corporation.
A master I2C interface is a bidirectional, two-wire serial bus that provides a medium-speed communications network. It can operate as a one-to-one, one-to-many, or many-to-many communication mediums. These engines support fast-mode I2C speeds. Pullup resistors are required for this interface.