Use ASICs to implement functional safety in battery-powered appliances

Article By : Enrique Martinez

Designers can consider ASICs for implementing functional safety to protect battery and motor control designs in home appliances.

Recent advances in battery technologies, coupled with environmental and energy efficiency initiatives, have accelerated a move toward many household appliances going cordless. While the removal of the mains supply gives users better protection against electric shock, risk is not mitigated completely, and therefore, functional safety still needs to be a core tenet of a system’s design.

This article looks at how functional safety can be applied in home appliances, and examines the economic tipping point of taking an ASIC vs. discrete component route to do so.

Recent developments and enhancements of safety standards and legislation aimed at home appliances include IEC 60335 for attended-use devices, IEC 60730 for un-attended use, and UL 1642 for Li-ion batteries. These standards not only highlight the growing importance for these devices to adhere to the fundamental principles of protecting people and property against dangers and damage, but also bring them in line with industrial, automotive, medical, and aerospace systems, where functional safety has always been a hot topic.

Managing batteries and controlling motors

Portable battery-operated appliances typically use Li-ion technology as their preferred power source due to its favorable energy-to-weight ratio, which is up to five times more than the Ni-Cd and lead-acid counterparts. Li-ion batteries also have a much higher number of operational charge-discharge cycles.

However, there are considerations to be taken into account when using Li-ion batteries, which are sensitive and need to remain within the boundaries of their charge and discharge profile. They do not tolerate overcharging, as an excessive charge will reduce the operational life. In a worst-case scenario, it can lead to fire or explosion due to an internal thermal runaway process, where the cell temperature increases to the lithium melting point (180.5°C).

Charging Li-ion batteries at low temperature is also an issue. Below 0°C, an internal plating phenomenon occurs, which will damage the cells and compromise the battery safety.

By proactively monitoring the cell’s temperature, voltage, and current, and by activating protection mechanisms before reaching critical limits, it’s possible to manage the above-mentioned problems. For example, if we look at a portable vacuum cleaner or a cordless sander, we can see a need for both battery management, motor speed control, and environmental control. To manage this, engineers can take a discrete component approach, or they can develop an ASIC.

A typical battery/motor system

If we look first at the elements of a battery manager, it requires high-precision voltage sensing to determine the Li-ion cells charge status and temperature. It also typically requires an ADC of up to 16-bit resolution to monitor critical situations, such as overloading or overheating, that may compromise the product safety.

Furthermore, an MCU with built-in flash memory is needed to process monitoring of the Li-ion cells and to add additional intelligence, user interfaces, debugging, communications, and auxiliary functions.

Now we look at the motor. The traditional solution for motors requiring speed control is a brushless 3-phase motor with a PWM drive. However, the cost of the controlling electronics in this architecture has limited it to high-end systems based on mains voltage.

The move to low-voltage battery operation leads to a different paradigm, one that utilizes very efficient low-voltage power MOS switches, and enables bridge rectification circuitry to be integrated. This comes with an added benefit of removing the need for high-voltage-mains isolation.

block diagram of an ASIC that facilitates battery management and motor controlFigure 1 This block diagram shows an ASIC that facilitates battery management and motor control functions in a portable appliance. Source: EnSilica

ASIC vs. discrete components

ASICs have developed a reputation for being costly. This may have been true years ago, and press coverage for ASICs often reinforces the reputation by focusing on large chips developed by and for Google, Tesla, and Apple. These ASICs use leading-edge technologies that demand hefty sums for a single mask set and design costs that stretch into hundreds of millions of dollars.

However, most applications don’t need leading edge technologies, and this is certainly the case for battery-powered devices. Instead, battery-powered appliances tend to be simple, small, and low power. Therefore, by adopting more cost-effective technologies such as bipolar-CMOS-DMOS (BCD), which allows the integration of power, flash memory, and both analog and digital functions into a single device, it’s possible to drastically reduce the masking and design costs.

Different masking options and voltage ratings of up to ~80V for the high-voltage MOSFETs enable ASIC silicon solutions for Li-ion battery-based motor control. Depending on the required motor power, it may also be possible to integrate the drive stage into the same die.

Additional features are also made possible within BDC ASIC implementations at almost no extra cost, including drivers slew-rate control for better EMC performance, bridge short-circuit detection and shutdown, chip thermal monitoring, and load voltage and current monitoring.

The cost of a CMOS-dedicated production mask set is in the region of $1.3 million at 28 nm and $0.4 million at 55/65 nm, going down to sub-$100k for 180 nm (Figure 2). These values continue to reduce as processes become more mature; for example, a 55/65 nm mask set is roughly a third the cost it was when first introduced.

graph of CMOS-dedicated production mask set costsFigure 2 Using mature technologies cuts the mask set costs significantly. Source: EnSilica

Of course, the process is not the only cost implication in designing an ASIC. The IP licensing, development, and qualification costs also need to be factored into the ROI calculation.

As a reasonable rule of thumb, you should consider an ASIC over commercial off-the-shelf (COTS) if you are looking to (a) achieve a design that is smaller, more efficient, and harder to imitate, and (b) the electronic component spend per product line is in excess of $2 million.

graph comparing ASIC and COTSFigure 3 ASIC or COTS? Here is an ROI analysis for a recent anonymized real-world safety critical device. Source: EnSilica

Minimizing costs

So, whether you are designing in house or using a custom ASIC developer, there are three golden rules you can use to keep NRE costs lower and increase your ROI.

Rule 1: Plan well

A good specification is crucial to success. It is crucial to understand exactly what is available in the market, and what unique functionality or user experience a customer expects or would like. The alternative will be a custom ASIC that is missing features, or one that is over-specified. Both ways, the advantages over standard parts will be lost.

Rule 2: Use proven IP blocks

Just as using mature processes will significantly cut mask set costs, using proven IP blocks enables engineers to realize a custom ASIC solution much more quickly, and significantly reduces the risk of getting it wrong.

Rule 3: Reuse your software (and your ASICs)

Software development impacts both project NRE and time-to-market significantly. Indeed, the development of tools and software are probably the biggest investment you will make over the product life-cycle, so reusing existing applications and software libraries can greatly help control project cost and accelerate time-to-market.

It should also be noted that the NRE cost per device reduces further if the ASIC can be used across multiple product lines.

This article was originally published on EDN.

Enrique Martinez is functional safety manager at custom ASIC design company EnSilica.

Related articles:

Leave a comment