A chip designed just for you?

ASICs are also establishing their place in many industrial applications but, despite their benefits, there’s still a misconception that the technology is unattainable for many manufacturers. Fortunately, the path to commissioning an ASIC is not singular as Richard Mount, Director of Sales at Swindon Silicon Systems, explains.

If you were to sit in the driver’s seat of your car and glance at the features available at your fingertips, you’d be looking at a surprisingly large number of ASIC-driven technologies. These are often mixed-signal devices that combine the benefits of analogue and digital technologies in a single chip and are found in every system from climate control and airbag deployment to infotainment systems and tyre pressure sensing.

Continuous development in ASIC design has also driven many other industrial sectors to choose custom integrated circuits (ICs) as a cost-effective alternative to traditional off-the-shelf components. This is particularly evident in factory automation and industrial controls where integrated sensor interfaces and the latest wireless standards combine to offer designers a high-performance smart sensor in a single package.

In fact, ASICs are being used wherever capability, size, cost and, in many cases, resistance to somewhat difficult environments are key design requirements. However, despite their growing prevalence across industry, there’s still a mistaken belief that the non-recurring engineering (NRE) costs and time implications of commissioning an ASIC make them unfeasible. The onus is on ASIC designers and suppliers, therefore, to bust those myths and suggest methods that can help ASICs fit within a manufacturer’s budget.

Why choose an ASIC?

First, let’s look at why a manufacturer should choose a custom ASIC. With many existing applications, it can often be beneficial to port an existing design into an ASIC in order to introduce new functionality and performance in a much smaller form factor, while allowing backward compatibility. This approach can preserve a customer’s existing design investment while providing equipment a new and extended lease of life.

In such an instance, an ASIC comes into its own if a manufacturer’s existing component supplier is making a process obsolete, but the product itself still has decades of potential. Here, it is possible to re-engineer a drop-in replacement to enable the continuation of the product’s availability.

However, there are many instances where it’s prudent, both technically and commercially, to develop an ASIC during the initial product design. As part of the process, the engineering team provides a specification that outlines their functionality and performance criteria, and the ASIC supplier wholly designs and produces the custom IC solution. Whether a manufacturer replaces their existing componentry with an ASIC or incorporates it into a new product design, they will be able to enjoy a multitude of benefits.

One of the principal reasons why a company will commission an ASIC is to protect its intellectual property (IP). Typically, a printed circuit board (PCB) comprises a variety of mounted discrete components, which includes commercial off-the-shelf (COTS) ICs.

However, it’s very possible for anybody skilled in electronics design to reverse engineer the PCB and replicate the product, thus potentially using the IP within that product for their own benefit. The strength of an ASIC is that it is extremely difficult to reverse engineer the IC or for it to be “re-used” when compared to a board level design thus ensuring that the IP is for the owner’s benefit and theirs alone.

From an engineer’s viewpoint, the real excitement comes from the clean slate approach that ASICs offer. For the most part, the standard requirement for an ASIC is to provide enhanced and optimised performance in a minimal footprint with low-power consumption. This ensures that the commissioning company benefits from a unique and optimised performance, targeted at their exact requirement to offer more functionality in a smaller, sleeker footprint.

To see how this works in the real world, we can return to the automotive tyre pressure monitoring system example. Here, the System in Package (SiP) sensor includes an ASIC that’s capable of seeing extremely low voltage levels from a co-packaged MEMS sensor and provides signal conditioning and processing along with RF data transmission to the vehicle. Each sensor is calibrated over temperature and pressure to ensure high degrees of measurement accuracy. Its power is supplied by a lithium button battery where operational life has to be guaranteed for many years, so extreme low power methods are used extensively.

This starkly contrasts with the performance that a typical discrete design achieves, where it would not be possible to achieve the same measurement accuracy and performance, in the same small footprint, with the required battery life. Even if the discrete design could match the ASIC’s size, its relatively high current demand would almost inevitably limit operational life.

No need to start from scratch

If a manufacturer wants an ASIC, they may still assume they do not have sufficient budget or time to make their ambitions a reality. The truth is, however, that ASIC developers have many ways to make an ASIC solution accessible to a variety of application and markets.

To ensure budgets and time are of a minimum, it’s possible to reuse existing IP in new or existing projects, shortcutting much of the design and verification. An established ASIC company should have access to an abundance of acquired circuit block IP, and a design team that has specified and created numerous custom ICs will quickly recognise the architecture most suited to a particular specification. This experience jump-starts the early design process, ensuring rapid progression from initial specification to a first engineering sample of the chip.

When in-house IP is not available, designers can license IP blocks from specialist vendors or use foundry IP and incorporate them into the design, accelerating timescales and reducing time and cost. IP obtained in either way can often be tweaked in-house, allowing it to meet customer-specific requirements. This kind of flexibility means that extremely complex systems can be monolithically integrated in far shorter timescales than when starting from scratch.

Working with an ASIC design and supply company that’s capable of delivering a full, turnkey solution can also prove cost-effective. An ASIC company that outsources any of the design and the production test of the device with outside vendors may lead to difficulties, especially in the production test and supply of the ASIC. Any ambiguity on who owns the responsibility and liability of a field return may result in unexpected costs and ultimately customer dissatisfaction.

Having all these elements in house, from the design of the ASIC, verification of the engineering samples, design and production of production test software and hardware, to full production test of packaged parts and wafer probe, QA facilities and order fulfilment will ensure full visibility, and ownership, of the entire process. This offering is known as full turn key (FtK) capability and is the proven and preferred methodology for any custom IC solution. This capability also brings with it the assurance of non-obsolescence, a known issue in today’s semiconductor market.

Thanks to developments in ASIC design and manufacture, many companies are considering, and embarking on, developing their own device, customised to their product requirements. Choosing a trustworthy, qualified ASIC partner will be the key to success — one that can deliver both the design and production test of the ASIC, using only in-house capabilities, is a prudent first step.