Power system design has advanced a lot over the years, resulting in some exciting tools and technologies being made available to power engineers. let’s take a look at what some of these are.
Phil Davies, global marketing and sales vice president, Vicor Corporation speaks with Atul Goel and Dilin Anand from EFY.
Q. Could you explain what is meant by power component methodology?
A. Power component methodology is for almost all power control and regulation engineers. Any engineering being done on AC-DC or DC-AC at any power levels can be considered to be part of it. Even front-end power systems developing a voltage for use as a hub from 48 volts (or 28 volts in military) down to the point of low voltages would fit in. Power component methodology aims to make it simple, scalable, repeatable, deliver faster time to market and for a design manager to get more productivity out of his design engineers because you do more design in less time.
Q. How are modern power regulation components helping engineers?
A. Modern power conversion and regulation engines typically switch at a very high frequency. This enables us to reduce the size of the magnetics and all types of passives that go within it. Having a lot of intellectual property (IP) in planar magnetics gives our engineers the ability to put cores of transformers and inductors inside our packages. In our case, this coupled with the proprietary conversion and regulation engines and the fact that we have our own silicon team allows us to design these control IC’s ourselves. We leverage that with our packaging technology to build the densest and most efficient power conversion and regulation modules available.
Q. Could you elaborate more on these engines in the power system design methodology?
A. The two basic corner stones of these engines are the sine amplitude convertor (SAC), which is similar to a resonant converter. The benefit is that when we use it in bus converter modules and voltage multipliers, there is no energy storage. So you can pulse very high current through here, with very fast transient response and the only thing that limits these products is how you thermally manage them. These are used in bus converters and voltage transformation modules. The other engine is ZVS zero voltage switching topology, that allows us to switch at very high frequencies but when we switch or transition it is done at zero voltage and in some cases zero current. This really minimises switching losses and body recovery all through the diodes in the power MOSFET and other devices.
Q. What are the technological improvements that have helped these components in the past few years?
A. Improvements over the years were with respect to switching at higher frequencies, as well as with the integrated core planar magnetics technology. In the past, these chips were manufactured single-sided, but these have now improved and are being manufactured double-sided. This lets the designer ensure that heat is drawn out of the component from both sides of the package. By extracting the heat faster, you get better watts per cubic centimetre or watts per square inch. This further reduces the density of the product. Moulding material, printed circuit board and silicon integration are the areas where we also focus on. For example, early controllers used many separate components inside these chips but now we integrate more moving from Bipolar to BiCMOS to Bipolar-CMOS-DMOS (BCD) technology. This helps make the control integrated circuit more efficient and better.
Q. How has the process of designing an actual power system changed over the years?
A. Many a times, a company will develop a new electronic system and then go to the power engineer and tell him their requirements: we need this type of voltage here and that voltage there, this power here and you’ve only got this much space so make sure the design is good thermally. The problem here is that the power engineers are often considered right at the end, so many times these guys end up designing their own power converters over and over again from the ground up.
Q. Wouldn’t they be able to re-use a design, even if they built it from the ground up the first time?
A. It’s very difficult to leverage one design from one platform to the next. Our power component methodology suggests a different way. Use power modules from front-end to the point-of-load, and use these modules to rapidly design a power system instead of starting from scratch. This also allows design engineers to improve their productivity out of their engineering time thus enabling them to do more designs. It is in this case that you can typically leverage the same components again on the next system design. The constant reduction in the size of the chips also helps in ensuring they fit into even smaller designs.