Tweet

Posted on 01 September 2019

Digital Power

Free Bodo's Power Magazines!

 

 

 

What’s been happening?

For several years now, the interest in digital power has grown, and commercial products based on such technology have been released to market. However, if for some applications such as UPS, computers, or mobile applications digital power is nothing new, then for others such as board mounted products, the technology is taking time to deploy, offering an opportunity to review its history and consider what, in today’s world could prove to be a huge benefit to the industry and beyond.

By Patrick LeFevre – Marketing Director, Ericsson Power Modules

 

The pioneers

Before reviewing today’s situation, it is important to remember the origins of digital power, and how from early research work conducted in the mid seventies by Trey Burns, N.R. Miller, and others, digital power gradually took its place in the power industry to reach a level of maturity that makes sense for a designer to consider such a technology. In the seventies, at a time when the power industry was slowly considering the migration from linear-power to switching-power, Trey Burns researched and explored the use of the State-Trajectory Control Law in Step-up DC/DC converters and he compared two methods of realization, one employing a digital processor and the other using analogue computational circuits. The results of this research were presented at various conferences but PESC 1977 is considered as the origin of a wave of research in digital methods to drive, monitor and control DC/DC converters (e.g. Bell Labs engineers, Norman Richards Miller presented an innovative approach to digitally control a switching regulator, and Victor B. Boros presented a novel serial digital implementation of feedback control circuits for powering conditioning equipments). At that point of time it is anecdotal but interesting to note that an experimental product built by Trey Burs was a boost converter operating at a switching frequency of 100Hz. Given, this sounds slow, but it had to be slow because it took up to 450 microsec to execute the digital program per sample. The digital controller was a PDP 11/45 mini-computer, and the boost converter was built, using a 10mH cut-C core inductor (very big and heavy) and approximately 13,000 microF of capacitance. The research team rolled the circuit up to the computer on a cart. Considering PESC 1977, it is interesting to remember a few words from the introduction of the paper Victor Boros presented: “Today, digital controllers are economically and technically feasible. The control function is not more complicated that circuits found in hand calculators and costs are comparable for LSI circuit reduction.”

Technology impulse

If today, digital technologies are everywhere, we should remember Trey Burns using a PDP-11/45 to control and simulate his model, and that the most advanced microprocessors available in those days were 8 bits such as the 8080 created by Intel with Federico Faggin as the lead designer (Figure 1). There is no doubt that the rapid development of the microprocessor industry boosted research in digital power management and control. From PESC 1977 onwards, year-after-year, papers presented at various conferences confirmed the growing interest from the research community for digital techniques applicable to power systems. Research progressed very fast though it was only in the mid eighties that we could consider that from the huge amount of research conducted - for almost ten years - that possible commercial applications would emerge.

8 bit microprocessor 8080 in 1977

As we consider PESC 1977 to be the ignition-point for research into digital-power, then the years 1984 and 1985 are the second cornerstone in the evolution of digital power technology. One example is when Chris Henze was working on his PhD at the University of Minnesota under the direction of Ned Mohan. Chris published some interesting parts of his work at PESC in Toulouse in 1985. In this work Chris was using a microprocessor and was switch- ing at a reasonable frequency for a non-isolated dc/dc converter of that era. In his paper he identified issues like quantization issues and the need to dither to get adequate PWM resolution. The digital control work behind the 85 PESC paper was actually built as a hardware circuit using the 7400 F-series TTL. This was done by hand, using wire-wrap and the controller was 8 inches on each side. The parts were all in sockets and were mostly 14 or 16 pin DIPs. The system ran on a clock from a crystal oscillator, which has a frequency of 20Mhz.

Following promising results of his research, Chris Henze built a semiindustrial version of the digital controller that put all of the digital logic into a single ‘gate array’, and later designed a power supply with 270Vdc input and with 5V as the main output and +/-12V auxiliary for an avionic application. The transformation by Chris Henze of early research to potential commercial applications is one among many that the late eighties and early nineties witnessed when designers started to consider using a micro-processor in power supplies.

1990 - The first wave of commercial applications

In the late 1990’s based on the Digital Signal Processor (DSP) C2000, Texas Instruments contributed to develop the first Uninterruptible Power Supply (UPS) that was digitally controlled. The introduction of a DSP to control the switching and power management of a UPS was the first practical application for digital power. This real-life application was the first of a long series of experiments aiming to optimize digital control in power supplies, expanding the scope of opportunities for the DSP. At the same time that the UPS industry considered the benefits of digital control in power management, Telecom Power Manufacturers (e.g. Ericsson Energy Systems) introduced communication features based on serial communication Bus technology (e.g. I2C developed by Philips) making it possible for operators to monitor and manage energy at site level.

Millennium milestone

Since PESC 1977 many years have passed, and with the millennium digital power reached a new cornerstone. Fundamental and important research conducted by those pioneers, and the rapid development of new components based on microprocessor technologies (e.g. Dedicated DSP for power management and control) have made it possible for power supply designers to access suitable controllers and topologies that simplify the migration from analogue to digital. At that time, apart from concerns about the millennium bug, the computer industry started to realize that power required by advanced power hungry processors would not be possible to manage without changing the way energy was distributed, and allocated to core sub systems. In response to the computer industry concerns, in October 2001 Intersil and Primarion formed the Digital Power Management Alliance to co-develop digital power management solutions for high-end desktop PCs, servers and notebook PCs.

This alliance marks the beginning of a new era of innovations, which would be too long to cover under this article, though from 2000 to 2004, we could consider some of the more important steps that contributed to the digital power migration from research to real-life applications (see table 1).

Digital power migration from research to real-life applications

Millennium forward

From the millennium, the number of papers presented at various conferences demonstrating the benefits of digital control and energy management increased tremendously, and in 2006 Ericsson Power Modules started to share the results of advanced research conducted in its laboratories aiming to develop tomorrow’s power solutions that will contribute to reduce energy consumption (see table 2).

Digital Power, the last few years of progress

At component level, the semiconductor industry started to announce products expected to make the development of digital control and digital power management as simple as it was for analogue, and some end-user-ready products started to appear on the market. Unfortunately, the lack of standardization, and the multiplication of different communication protocols added a level of complexity when designers considered using this technology.

Do you speak Easy-Bus?

As for other industries, the development of a new technology always generates new demands, requiring new ways of working and standardization. For example, among new technologies introduced over the last decade, there is the short-range radio Bluetooth, which is an illustration of a new technology that in 10 years has moved from laboratory research to commercial success. Everything started in 1994 when mobile equipment makers initiated a study to investigate the feasibility of a low-power low-cost radio interface between mobile phones and their accessories. In February 1998, five companies, Ericsson, Nokia, IBM, Toshiba and Intel formed a Special Interest Group (SIG). 10 years later, 1.5 billion Bluetooth devices are in operation worldwide. Bluetooth is an illustration of a new technology born from the willingness of Industry leaders to share knowledge, working together to develop new solutions that make life easier, and more efficient. Bluetooth could seem to be an odd example though it proves the efficiency of a new way-of-working when companies are collaborating to develop new technologies, making it possible to develop interoperable units by creating a standard. The new possibilities and simplicity offered by the addition of digital control into power supplies revealed the lack of efficient communica- tion protocols dedicated to this new domain of applications emerging in the power industry.

Introduced by Philips in the early eighties, the Inter-IC-BUS (I2C) cohabited with RS-232, RS-485, SMBus, SPI Bus, CAN Bus, and many proprietary protocols and formats. In this jungle, components manufacturers, power industry leaders, and end-users began considering how to develop and standardize a common vehicle and a package of instructions to support this new technology. In the same spirit as Bluetooth, in May 2004, Artesyn Technologies, Astec Power, and a group of semiconductor suppliers (Texas Instruments, Volterra Semiconductors, Microchip Technology, Summit Microelectronics, and Zilker Labs) formed a coalition to develop an open standard for a communication vehicle and protocol dedicated to power systems. A standard named PMBus was born (Figure 2).

PMBus symbol

At the end of 2007, the PMBus Implementers Forum (PMBus-IF), comprised about 30+ adopters with the objective of providing support to, and facilitating adoption amongst users.

Digital forward

From 2004, in parallel to the development of PMBus, step by step companies started to introduce new products and solutions to facilitate the evolution from analogue to digital (see table 3).

The evolution from analogue to digital

Because equipment manufacturers are not advertising too much what’s in the box, a number of commercial applications are already using digital control and digital power management. For those using digital power, the benefits offered by such technology are obvious, and besides the possibility to manage energy distribution throughout complex systems when in operation, the benefit to profile on-board power solution through a graphic user interface during the development phase is considered as a significant benefit reducing time to market. Most of today’s ICT applications could benefit from digital power, and we could take a radio base station as an example of a product where digital control and digital power management will play a significant role in future developments. Power consumption in mobile radio applications is very much driven by traffic, and by combining traffic management and intelligent power-management it will be possible to only power the required part of the system needed at that time and to standby the rest of the time if not required. Otherwise, when traffic increases the traffic management controller will enable additional functionalities. For example traffic management could control the number of power amplifiers in operation and decide to power ON or OFF some of them when traffic increases or decreases, or to adjust the polarization voltage to the most efficient profile at the time. At cabinet level, certain boards that include mixed-functionalities that are only required at certain times during the operation will have the ability to be powered ON/OFF, or adjusted to suit vital parameters, and precisely monitored to make it possible for the traffic management controller to report in real-time to the site manager on different parameters.

Managing power management down to the vital few is also very important to reduce the fixed operational power and especially the power consumed by air-conditioning and ventilation. Adding digital control to individual boards and PMBus, it becomes simpler to better control cooling and ventilation, optimizing operational conditions to suit specific traffic conditions, and statistically profile the power demand for the requirements of the next flow of traffic. If considered at the start of the project, the same type of topology could easily be deployed to other ICT applications such as data-centres, reducing power consumption and CO² emission.

Conclusion

Driven by growing concerns about energy preservation and the reduction of CO² emissions by the Information Communication Technology (ICT) industry, power supply manufacturers have taken seriously the measure of the situation and initiated a number of projects that contribute to reduce the environmental impact. The development of efficient power conversion systems associated with active energy management, made possible by digital technologies, is the most evident way to go, contributing to the rapid development of commercial ‘digital power solutions’.

 

 

VN:F [1.9.17_1161]
Rating: 0.0/6 (0 votes cast)

This post was written by:

- who has written 791 posts on PowerGuru - Power Electronics Information Portal.


Contact the author

Leave a Response

You must be logged in to post a comment.