When no-brainer decisions deserve greater scrutiny

  • Feb 03, 2020
  • UnitedSiC

By Dr. Anup Bhalla, VP Engineering at UnitedSiC

Abstract

When it comes to selecting the right power transistor technology for a new or existing design, it should never be a ‘no-brainer’, even when it appears that the key figures of merit are compelling.

The purpose of this blog is to suggest that even when the term ‘drop-in replacement’ is used legitimately, UnitedSiC will always encourage engineers to scrutinise all of the information available, because it is confident that it will be even more compelling for engineers

Blog

How many times have you been told ‘It’s a no brainer’?

“Upgrade to our large meal with drink and fries for just 29c extra – it’s a no-brainer.”

“Update to the latest version of our operating system – it’s a no-brainer.”

“Invest in our cryptocurrency – it’s a no-brainer.”

Experience teaches us that such ‘no-brainers’ often lead to unexpected consequences, which can range from straining waistbands through broken applications to devastating financial losses.

Fortunately, as engineers, we are trained to seek out shortcuts, quick fixes and elegant optimisations – but to doubt their value until we have checked them out for ourselves. It’s the engineering equivalent of Ronald Reagan’s strategy for the Cold War nuclear disarmament process – ‘trust but verify’.

This approach enables us to engage with what may be valuable opportunities, without being dazzled by the hype. It’s particularly useful when we’re being told to believe that one version of a product is as good as another. You know the sort of thing:

“These no-brand tyres are definitely as good as the ones fitted when the car was new.”

“This fast phone charger is definitely as safe as the original.”

“This printer cartridge is definitely a drop-in replacement for the vendor’s own.”

Well, maybe – but accepting such assertions is definitely not a no-brainer. You need to trust – but verify.

We at UnitedSiC face this challenge all the time as we develop new products. We have a power-switching technology, known as a SiC FET, which combines a normally-on silicon carbide (SiC) JFET and a Si MOSFET in a stacked cascode topology to produce a normally-off SiC FET device.

This has low RDS(ON) characteristics, low internal capacitances and a high maximum operating temperature – making it useful for building highly efficient power-conversion circuits with good power densities.

Obviously, we want to make this new technology as accessible to the market as possible, and so our product lines are being tailored and packaged to serve established market niches.

For example, we have just introduced a pair of 650V SiC FETs that we believe have the lowest RDS(ON) figures for any similar switching devices in the same package.

The UF3SC065030D8S has an RDS(ON) of 34mΩ, and the UF3SC065040D8S has an RDS(ON) of 45mΩ. Both devices have been engineered so that they can be driven in the same way as standard Si FETs, IGBTs and SiC MOSFETs. And we’re supplying them in the popular DFN 8×8 low-profile, surface-mount package, widely used in power-conversion circuits for applications where space and power-density considerations are key – such as the wireless and telecoms markets.

Our devices’ low internal capacitances enable fast, efficient switching, while low internal resistances reduce heating losses and so enable greater power density. This combination of headline performance figures, familiar drive characteristics and widely used packaging makes the parts a drop-in replacement for the current technology – in other words, a no-brainer.

Hmmm – a ‘no-brainer’, right? Well, maybe. At this point we know that any good engineer’s training will kick into high gear, and that they’ll ‘trust but verify’ the claim by exploring our comprehensive specification and design-support documentation to validate the advantages of our technology for themselves.

Working with partial data is never a good idea, especially for a new technology whose performance and application may vary in subtle ways from the established approaches, and for which a couple of Figures of Merit cannot tell the whole story. For example, the SiC FETs I mentioned above also have a Kelvin gate return, which enables designers to develop circuit topologies that enable cleaner drive characteristics. It’s a small feature, but it could be a real advantage in some applications.

As in life, so in engineering. When it comes to making a ‘no-brainer’ decision about trying something new, even if the choice seems clear, you should still check the small print. It may be that the new offering is even better than you first thought.