Last year at this time, we were in the midst of our 30th anniversary celebration, which gave us an opportunity to reflect on the progress that has been made over the years in the power electronics industry. Although there is no 31st celebration in the works, we do have another chance to glance back in time as we honor Bruce Carsten with this year's Lifetime Achievement Award.
You may know Carsten from the many design seminars he has conducted over the years at PCIM, PowerSystems World and other conferences. Or you may have read some of his landmark papers on magnetics design and other topics. But I suspect (or hope) that there are some stories in this issue concerning Carsten's design experiences that will be new to many readers.
As you read some of these stories, you may be inclined to nod in agreement as you encounter familiar themes. First, you may recognize the appearance of a fundamental issue in engineering work that has as much to do with ethics as with math and physics. As illustrated by some of Carsten's experiences, the design for production environment can produce real conflicts between the designer and management over when a design should be deemed ready for production. The engineer, who so often is a perfectionist, wants the design to be fully optimized. The employer, who wants the design in production as soon as possible to start generating revenue, may put pressure on the engineer to release a design prematurely. Carsten relates how he faced this issue:
“There's always a tendency to perfect things. So, how do you know when a design is done, because you'll never have it perfect? I said to myself, ‘Put yourself in the place of the customer, knowing what you know about the product. If this product is working for you and your livelihood depends on this product, can you still sleep at night?’ If the answer is yes, the product is done. But if you'd lie awake at night being concerned about something that might fail, then it isn't done.”
Beyond these ethical issues, you may also sense a set of familiar design challenges in Carsten's experiences, challenges such as getting switchmode power converters to operate at high frequencies, making power supplies operate over wide voltage ranges and making them more robust — even immune to fault conditions. As you go from the Lifetime Achievement Award article to this month's design features, you'll see that many of the same issues are still being confronted today.
One example is found in this month's feature, “Driving Automotive Power Supplies to Higher Frequencies” (page 14). The authors, engineers at Maxim Integrated Products, tackle the issue of how to push switching frequencies higher, in this case, above the AM broadcast band to eliminate EMI to the radio. Here, the challenge of designing an efficient high-frequency regulator is complicated by the need to protect the regulator chip against high-voltage transients in a vehicle's electrical system. As the authors note, high-voltage CMOS processes can be inherently slow, making it difficult to a build a regulator that both withstands high voltages and is efficient.
Then too, you'll see that certain control methods such as peak-current-mode control, which Carsten was developing in the early 1970s, are still ripe for further innovation. A glance at the feature “Control Method Solves Low-Duty Cycle Problems” (page 22) reveals how National Semiconductor developed a variation on the peak-current-mode control method to address ongoing requirements for very short on-times for a buck regulator's high-side FET.
As you read this issue, there's no doubt you'll find other examples of design challenges that have remained with us since the dawn of the switchmode era, despite all the progress made in circuit design and integration, magnetics and passive components. Perhaps, you even know of an area where the technology is not moving ahead or even regressing because of a lack of attention. If so, please share your thoughts by writing to me at [email protected].