This article will help you understand rectifiers: how their static behavior differs from their dynamic behavior, and how “speed” is used as a defining characteristic.

Introduction

The concept of an ideal diode can deceive the designer into overlooking the importance of specifying the correct power rectifier to meet his/her circuit designs' performance requirements. Far from being simple “one-way valves for current,” real-world rectifiers present an array of circuit-affecting characteristic parameters.

In this two-part series we are exploring how these parameters both define the types of rectifiers and drive the development of rectifier technology, and along the way we will gain the understanding necessary to choose wisely for our design projects.

Here in this first article, we explore the key static parameters applicable to any rectifier, and introduce the dynamic behavior associated with the first three rectifier types.

 

So Many Rectifier Choices! But Which Is the Best?

At first glance, a rectifier appears to be simply a diode rated for handling larger currents and voltages than a signal diode. Ideally, it is a two-terminal device that allows conventional current to flow through it in one direction (anode-to-cathode) and blocks current from flowing through it in the other direction (cathode-to-anode). 

But this simplified ideal-diode concept can deceive the designer into overlooking the importance of specifying the correct power rectifier to meet his/her circuit designs' performance requirements.

Semiconductor physics and the realities of manufacturing prevent the creation of “ideal diodes,” and the creation of diode rectifiers capable of handling power is a balancing act of parameter trade-offs. Far from being simple “one-way valves for current,” real-world rectifiers are manufactured in several different types with different performance parameters to address different applications.

Rectifier types include standard, fast, and ultrafast silicon PN junction rectifiers, as well as Schottky barrier rectifiers. And each of these types presents a different set of circuit-affecting characteristics. To understand these differences, we will explore both the static and dynamic aspects of rectifier function.

 

I-V Curves Tell Part of the Story

So the first step in understanding these different rectifier types is to look at a representative real-world current-vs-voltage plot of their operation. A typical static (time-invariant or steady-state) plot of the current vs. applied voltage is shown here. (We will assume that this plot is for a rectifier having the desired average rectified forward current rating (IO) that you need.)

 

Representative I-V plot for a modern power rectifier

 

Diodes, including power diodes, operate in the first and third quadrants of the I-V plot. The first quadrant shows the forward-biased operating region; the third quadrant shows the reverse-biased operating region. (Note that the voltage scale in the reverse direction is much larger than the scale in the forward direction.)

In the first quadrant, you can see how the rectifier starts conducting large currents well before the forward bias reaches one volt. As we increase the forward bias to the point where the current begins increasing exponentially, we have established the parameter known as the rectifier's forward voltage drop or VF.

In the third quadrant, we see that the rectifier is blocking the flow of current (except for a small leakage value) even as the reverse bias is increased many tens of volts. The small amount of leakage current is termed the reverse current, or IR, and is a key performance parameter (the smaller the better).

Observe, however, that as we continue to increase the reverse bias we will eventually reach a point where the blocking action begins to fail and large reverse breakdown currents occur. If we note on the I-V plot's horizontal axis the absolute value of the reverse bias that immediately precedes the large breakdown currents, we have established the rectifier's DC blocking voltage or VR.

Another very important parameter—in fact, the first one the engineer usually determines for his rectifier application—is the average rectified forward current, or IO. This parameter is primarily determined by the die size and package (body) of the rectifier. These are created in a wide number of physical form-factors, including:

  • bare die
  • axial lead molded
  • button
  • tab-mounted
  • surface-mounted
  • bolt-mounted

And there are more. This photo presents just a small sample of the wide array of packages in which rectifiers are manufactured.

 

Some rectifier packages, from 1A to 600A

 

Thus, via the steady-state I-V plot, we have established three key datasheet design parameters: VF, VR, and IR.

The I-V plot, however, usually will not be very helpful in comparing the various types of rectifiers. This is because, in a plot such as this, we won't see much (if any) difference between ultrafast, fast, standard or even soft recovery silicon PN junction rectifiers of the same voltage (VR) and current (IO) rating.

Even a Schottky rectifier would look much the same. The main differences would be that the forward voltage drop would be about half that of the silicon PN junction equivalent device, and the leakage current would be perhaps five times worse (greater magnitude) to as much as 50 times worse, depending on operating temperatures.

But if your power circuitry is operating at 60Hz or slower, or your current never changes direction faster than a few hundred milliamps per millisecond, you probably don't need to consider anything more than the static I-V plot and the associated rectifier parameters derived from this plot such as forward voltage (VF), reverse blocking voltage (VR), and reverse leakage current (IR).

But if your dI/dt is in the range of amps to tens of amps per microsecond, then you need to pay attention to the recovery characteristics of your candidate diodes and look at the diode in a dynamic (time-variant) plot of the current. This dynamic recovery behavior is where we can clearly differentiate between these different types of rectifiers.

 

What Is “Recovery” and Why Is It Important?

Graphically, “recovery” is the dynamic (time-variant) behavior of a diode in returning (i.e., recovering) to the I-V plot's predicted operating point immediately after being subjected to high dI/dt perturbations.

There are two types of recovery important enough to be characterized on datasheets and covered by JEDEC standards: forward recovery and reverse recovery.

The speed of diode recovery is important because if it is too slow (i.e., takes too long to recover), the diode may not provide the desired rectification function in circuits operating at high frequencies. In particular, switching power supply circuits used in DC/DC converters, power factor correction circuits, motor control, and other high switching-speed PWM applications often require diodes that can quickly recover to an operating point within 10-25% of their predicted steady-state I-V curve plotted point.

As you might expect, the terms “fast” and “ultrafast” are in comparison to standard rectifiers designed for low-frequency applications such as rectifying sinusoidal current supplied from the AC mains. A “fast” rectifier typically recovers ten times faster than a standard rectifier, and an “ultrafast” designation is usually applied to rectifiers designed to beat the standard rectifier recovery by being more than fifty times faster.

It is important to note, however, that when rectifiers are categorized or typed as “ultrafast,” “fast,” or “soft,” this is in reference to their reverse recovery characteristics. “Fast” and “ultrafast” rectifiers are so termed because they cease conducting current in the reverse direction much more quickly than standard rectifiers. “Soft” is a term applied to a subset of ultrafast rectifiers that recover to the non-conducting state quickly, but in a non-abrupt manner.

 

Reverse Recovery

Your thoughts right now might be, “Wait, conducting in the reverse direction? Rectifiers aren't supposed to conduct current in the reverse direction!” And you'd be right, except for the fact that we are dealing with real-world diodes, not theoretically “ideal” diodes.

A theoretically ideal diode would cease to conduct current the very instant that the cathode becomes positive with respect to the anode. Real diodes, especially silicon PN junction power rectifiers, differ from the ideal diode in that, for a short time, they will continue to conduct a high current in the reverse direction until the charge carriers in the junction are eliminated and the depletion zone is re-established.

 

Defining the parts of the reverse recovery waveform (image derived from JEDEC Standard No. JESD282B.01, Figure 6.1)

 

JEDEC Standard No. JESD282B.01 helps us understand this dynamic behavior by defining the parts of the rectifier reverse recovery waveform. In the waveform pictured above, we are plotting the rise and fall of current through the rectifier versus time. A sinusoidal pulse of current is applied to the rectifier and falls at a circuit-determined dI/dt.

After the decreasing current crosses the x-axis (i.e., reaches zero amps), it continues in the same direction, such that the current is now flowing in the reverse direction through the rectifier. The reverse current flowing from the instant of the zero-crossing to the point at which the maximum (absolute value) of reverse current (Irm) is reached defines the parameter trrr. This is primarily supported (caused by) the injected minority carriers left over in the junction area during the forward-biased portion of the falling dI/dt. The associated charge (in coulombs) of these carriers is termed Qr and is defined by the area under the curve up to the instant of Irm.

From this point of peak overshoot, the current now reverses slope and heads back towards zero; this time period is termed trrf. The carriers supporting trrf are supplied as the depletion zone is re-established, and it is during this time period that the junction develops its reverse blocking voltage. Likewise, the associated charge is termed Qf. The total time interval for this reverse recovery is, accordingly, trr.

 

Drawing Conclusions from Static to Dynamic

We have learned that the parameters associated with the static I-V plot provide much of the information we need to predict how a rectifier would function within a circuit, as long as the frequency or switching speed of the circuit is low. We have also learned that, for circuits that subject the rectifier to rapid switching or cycling between conditions of forward- and reverse-bias operation, we will require an understanding of a rectifier's dynamic characteristic parameters—in particular, its reverse recovery time (trr).

It is the trr parameter that is used to define or classify a rectifier as “fast” or “ultrafast.” “Fast” generally denotes trr of no more than 200ns, and “ultrafast” has trr of less than 150ns. For high-frequency or fast-switching circuits, it is advisable to use rectifiers with the shortest trr times so that the rectifier can spend the vast majority of its time either conducting forward current or blocking reverse current, and not simply recovering from the former or the latter.

But know this: The shape of the reverse recovery waveform is also important, as it may have a significant effect on the circuit in which the rectifier resides. Accordingly, we will need to discuss the issue of why reverse recovery characteristics are described as either “soft” or “abrupt.” This is what we'll cover in the second article in this series, along with forward recovery and the dynamic behavior of Schottky rectifiers.

 

Next Article in Series:

 

Comments

0 Comments