Performance, power density and efficiency are goals not normally associated with electrical systems on automobiles. That was yesterday, now the pistons, valves and fuel pumps of the last 150 years have transformed into the electric motors, inverters and batteries of today.

The last 24 months will probably be charted as the most dynamic in the automobile industry since its inception. We have seen change on many fronts; mergers, acquisitions, new models launched, iconic models mothballed and even one of the Big Three enter and emerge from bankruptcy. Ironically because of all of the headline-grabbing corporate twists and turns, the most radical milestone since the invention of the internal combustion engine in the 1850s has occurred almost unnoticed: the electrification of the automobile. Fortunately, while stockbrokers and bankers were counting the losses, new electrical systems for the car of tomorrow that rely on electricity and efficiency have been developed. These innovations both reduce the fuel a car consumes and the exhaust gasses it produces.

However, these new systems present many challenges to the power electronic engineer; next-generation goals are in need of a next-generation technology. While the semiconductors used in power MOSFET technology have made huge advances over the last 40 years, the package that houses the switch has remained fundamentally unchanged. The molding, wire bonds and lead frame of the plastic power package are a fundamental barrier to reaching best performance. As the semiconductor switch gets better and better it’s a case of diminishing returns per dollar spent for the design engineer as the parasitic inductance, electrical and thermal resistance of the plastic power package remain a limiting factor.

A New Approach
To address these issues, a new approach is required that combines the latest generation AEC-Q101 qualified Trench MOSFET silicon with a unique packaging technology. IR’s Automotive DirectFET®2 product line, for example, based on the proven heritage of the DirectFET platform launched 8 years ago, combats the three key limitations of thermal resistance, parasitic resistance and parasitic inductance normally associated with traditional plastic power packages.

Minimal Thermal Resistance
DirectFET2 allows dual-sided cooling to be used (see Figure 1) to dramatically reduce the resistance of the overall thermal path to around half that of a D2Pak thereby lowering operating temperatures, minimizing heat sinks and improving efficiency. 

Figure 1: Dual-sided cooling delivers superior thermal performance

Minimal Parasitic Resistance
Increasingly at lower Rds(on) values, the majority of the Rds(on) and associated losses come from the wirebonds in the plastic power package, so paying for a large piece of silicon to reduce conduction losses is not cost effective. The new device eliminates wirebonds. Instead solderable front metal (SFM) pads on the die allow direct connection from the silicon to the circuit board, so the package only adds 150?? – small compared with the milliohms added by traditional power packages. As a result, losses are reduced and reliability improved. The ‘skin effect’ at higher frequency is also less pronounced; the removal of the wirebonds means that the package retains its low Rds(on) characteristic even in the MHz range making it an ideal candidate for the DC-DC converters needed on hybrid vehicles.

Minimal Parasitic Inductance
By removing the wirebonds, parasitic inductance is dramatically reduced. Nobody understands the hazards of parasitic inductance more than the power electronics engineer. At best this phenomena causes inefficiency, at worst it causes a system to stop functioning. DirectFET2 allows reduction or even elimination of external snubber networks as the die free package inductance of the device is around 0.6nH compared with 5nH produced by a D2Pak (see Figure 2). As a result, a more elegant design evolves; losses are cut along with system size and cost. 

Figure 2: Reduced parasitic inductance and ringing improves EMI performance

There is also another hidden benefit of the new power package; all of the performance and efficiency comes in a small footprint by maximizing the ratio of silicon to package area. Figure 3 maps the performance and footprint of the new packages against the traditional plastic power packages. The small can offers superior Rds(on) compared with the Micro and TSOP 8 in the same footprint, the medium can houses the same die size as the DPak but occupying no more space than a 5x6 PQFN. The new large can however achieves a 60% area reduction compared with a D2Pak but holds a die that is 30% larger.

Figure 3: Package comparisons

Equally important is the manufacturability and reliability of the device. The simplicity of the new device fits in well with the high reliability requirements of the automotive industry and it is designed for use on standard SMD manufacturing lines. With the foundation of dedicated R&D and design and a zero ppm goal manufacturing initiative the new products are AEC-Q101 qualified and survive the autoclave test, enabling designers to comfortably meet the demanding specifications of next generation automotive systems.

With a range of voltages from 40 to 250V, three can sizes and silicon optimized for either Low Rds(on), low Qg or logic level operation the product family can address a broad range of applications from class D audio, to electric power steering, inverters, DC-DC converters, HID, battery switches, braking and injection applications. When compared to traditional solutions, the new DirectFET2 addresses the key issues faced by design engineers, enabling them to meet the key benchmarks of Performance, Power Density and Efficiency required to make the electrification of the automobile a lasting success.