Why envelope tracking is both impressive and scary
Article By : Bill Schweber
Envelope tracking is widely used to increase the PA efficiency, yet its premise is contrary to the established analog design best practices.
One of the first lessons learned in electronics in general and analog circuits in particular is the importance of a solid, stable, stiff DC-power rail. It’s axiomatic that a rail which is noisy, unstable, sensitive to load transients, or otherwise not pristine can cause all sorts of inexplicable problems.
It’s especially so with low-level signals and their sensors, op amps, and A/D and D/A converters. It’s not unusual to have to revisit a prototype and add components such as ferrite beads and additional bypass capacitors; in difficult cases, it may be necessary to re-do the power topology to put a point-of-load (PoL) converter right next to a sensitive load.
That’s why I was somewhat mystified when I first encountered the concept of “envelope tracking” (ET) used alongside transmitter power amplifiers (PAs). No, ET is not a shortened title for the classic movie “ET: The Extraterrestrial” (1982) nor is it a secret scheme of the post office to track your mail. It’s not new, either; it was first proposed in 1952 by Leonard Kahn to improve efficiency of SSB transmitters, except that he called it “Envelope Elimination and Restoration.”
It was originally used in higher-power transmitters pumping out hundreds of watts and above, but in recent years, it has been embedded into smaller systems, including leading-edge smartphones and even IoT devices which have outputs on the order of one watt or below. The technique was not widely adopted for many years due to difficulty of implementation, particularly for signals with wide bandwidth. However, in the last few years, the implementational issues have been overcome and ET is now widely used to improve PA efficiency in cellular handsets.
It’s a very sophisticated technique which dynamically adjusts the rail voltage on the PA to maximize efficiency while also enhancing linearity. ET is complicated, but in brief, it works the way shown in the figure below. A conventional fixed-supply PA has to use a supply voltage high enough to support peak power, but is only energy efficient at or close to the peak. Most of the time that voltage is much higher than needed, resulting in high PA heat dissipation. In ET, the PA supply voltage is dynamically adjusted to the instantaneous amplitude of the signal, resulting in high PA efficiency at all times.
Envelope tracking overturns the rule of using a tightly-fixed DC supply rail as it dynamically adjusts the rail’s voltage to maximize PA efficiency without compromising other performance characteristics.
Complete ET technology suites include control algorithms, power supplies and more, and are available in whole or part from vendors such as Nujira (acquired by Qualcomm in 2015), MediaTek, Quantance, EPC, and others.
Envelope tracking should not be confused with the Doherty amplifier, an eponymous technique first proposed in 1936 to improve efficiency of a PA when it’s handling lower-level and higher-level signals. Both ET and Doherty amplifiers are examples of ideas from long ago—in the days of vacuum tubes and basic radios—that are finding new life and providing benefits in the smartphone and 5G worlds.
So, why does ET scare me? In general, every dynamic loop in a circuit, regardless of its the objective or function, brings issues of time constants, overshoots, ringing, and all sorts of surprises as they interact with each other in unforeseen and hard-to-assess ways. Not only does ET mean that the DC rail is not fixed, but it’s another thing that will be changing “on-the-fly” in an analog circuit. In fact, as partial insurance against supply variations, some building-block analog components even have a specification for power supply rejection ratio (PSRR), a figure of merit indication how well they maintain their composure despite supply variations.
My experience has been that the fewer things “in motion” in an analog circuit, except for the signal(s) of interest, the more consistent is its performance. While processors have used dynamic voltage scaling for years to save power when less is being demanded of them, the implications are different in the analog world of amplifiers. That’s why the fact that ET works and works well is so amazing to me, as its premise is contrary to my long-standing conventional wisdom.
Have you ever had misgivings or worries when using a technique which goes against some of your basic principles? What was your level of confidence when you tried it? Were you still skeptical, or did the tangible results convince you?
This article was originally published on Planet Analog.
Bill Schweber is an electronics engineer who has written three textbooks on electronic communications systems, as well as hundreds of technical articles, opinion columns, and product features. In past roles, he worked as a technical website manager for multiple EE Times sites and as both Executive Editor and Analog Editor at EDN. At Analog Devices, he was in marketing communications; as a result, he has been on both sides of the technical PR function, presenting company products, stories, and messages to the media and also as the recipient of these. Prior to the marcom role at Analog, Bill was Associate Editor of its respected technical journal, and also worked in its product marketing and applications engineering groups. Before those roles, he was at Instron Corp., doing hands-on analog- and power-circuit design and systems integration for materials-testing machine controls. He has a BSEE from Columbia University and an MSEE from the University of Massachusetts, is a Registered Professional Engineer, and holds an Advanced Class amateur radio license. He has also planned, written, and presented online courses on a variety of engineering topics, including MOSFET basics, ADC selection, and driving LEDs.