News

Why Comparing Low-Power MCUs Can Be So Difficult

October 05, 2020 by Jake Hertz

When vendors have different names for power modes of low-power MCUs, it can be confusing to compare options for battery-powered IoT designs.

The Internet of Things and wearable tech have made low-power circuit design paramount. Devices like these tend to be standalone systems, meaning they are often battery-powered. Naturally, the functionality of these devices is only as good as their battery life; the longer they last, the more useful the device. 

 

Comparing system power consumptions

Comparing system power consumption. Image used courtesy of Hackaday

 

For these reasons, the market has been inundated with low-power MCU options.

While a variety of options is certainly useful for designers, it can oftentimes be daunting to decide which MCU to use for which application. This is especially true when trying to sift through datasheets and promotional specifications. In this article, we’ll review different considerations when choosing a low-power MCU for your design. 

 

The Problem With Comparing MCUs 

Nowadays, it is not uncommon for MCUs to have upwards of nine power modes. To make things more confusing, different vendors tend to name comparable power modes differently. As a result, trying to compare two comparable MCUs can be confusing. 

 

The profiles of MCU power consumption

The profiles of MCU power consumption. Image used courtesy of Silicon Labs
 

In general, MCU suppliers tend to display the lowest power achievable on the first page of the datasheet. While a device may be capable of the presented power specification, the operating mode used to achieve this spec may not be practical in a designer’s specific application.

Some of the non-advertised features of the lowest power mode may include a very slow wake time, no state or RAM retention, or a reduced operating voltage range.

 

Initial Considerations for Low-Power MCUs

To compare power consumption, engineers need to consider current consumption, state retention, wake-up time, and wake-up sources. Another important consideration is the availability of peripherals that can reduce total system power. Usually, these kinds of electrical specifications are organized by vendor-specific power mode. 

Hence, to accurately compare MCUs, designers must first define operating conditions and match them to common operating modes. Different operating condition considerations may include sleep mode current consumption, RAM retention, RTC enabled/disabled, peripherals enabled/disabled, wake time, and supply voltage range, among others.

 

A snippet of an STMicroelectronics low-power MCU datasheet

A snippet of an STMicroelectronics low-power MCU datasheet. Image used courtesy of Mouser
 

Engineers must first understand the intended operation of their finished product when taking these operating conditions into account.

For example, if an engineer were designing an IoT device that made one reading and transmission daily and slept the rest of the day, it would make sense to choose the MCU with the lowest deep sleep power mode consumption because this is the most common operating condition. 

Once operating conditions are clearly defined, it should become significantly easier to evaluate and compare MCUs.

 

Other Considerations for Low-Power MCUs

Beyond the quantifiable low-power specifications, there are different functions that may be critical to achieving the lowest overall system power consumption. 

One example is deciding whether or not an MCU has the ability to allow firmware to scale the internal supply voltage. An MCU operating at a slow frequency has the ability to decrease its supply voltage and save power.

 

1-second period of waking up from deep-sleep mode

Example of a device with a 1-second period of waking up from deep-sleep mode, executing a fixed workload, and returning to a deep-sleep mode. Image used courtesy of Analog Devices
 

Another beneficial function may be selective clock gating, which allows hardware blocks to be disconnected from the active circuits, preventing inactive peripherals from consuming power. 

While there are others, these two examples clearly show that certain functionality of an MCU, beyond its raw current consumption numbers, may be crucial to selecting the best low-power solution.

 

Microchip’s MCU Plus Display Driver

Adding to the list of available low-power MCUs, Microchip released its new PIC24F GU and GL families of MCUs

Notably, these devices are meant for battery-operated applications with digital displays, evidenced by the integration of a low-power display driver.

By using an integrated LCD driver, developers can offload simple animation routines from the CPU, allowing for display animation to operate even when the MCU is in doze, idle, or sleep modes for power savings.

 

Microchip’s newest low power MCU, the PIC24F

Microchip’s newest low power MCU, the PIC24F. Image used courtesy of Microchip
 

Joe Thomsen, VP of Microchip’s MCU16 business unit says, “This scalable family of devices enables applications from cloud-connected low-power IoT nodes and sensor systems to automotive, consumer and industrial automation applications and helps developers easily add displays, robustness, and security to their designs.”

 

Choosing the Right Low-Power MCU

Choosing the right low-power MCU for your specific application can seem daunting at first. To better evaluate and compare different options, one must first understand the MCU's intended operating conditions, specifications, and functionality. 

With the demand for ultra-low-power design on the rise, developing these skills will be indispensable for practicing engineers.