It’s convenient when we can assume that the voltage on a particular pin will always be more or less what we expect. This is often the case in the low-voltage and digital circuits that are so common these days. Sure, there are always imperfections—a little overvoltage here, a small noise spike there. But often we can simply ignore these things; no protection circuitry is required.
All images courtesy of Diodes Incorporated (PDF)
But not all electronic environments are so benign. In some cases, a low-voltage signal can be subject to relatively large transients that will (at best) interfere with normal functionality or (at worst) destroy your device. Actually, device destruction is not the worst-case scenario if you consider the possibility of a component catching on fire and eventually reducing the entire system to ashes and charred metal.
This is where transient-suppression products come into play. A common term here is TVS (transient voltage suppressor), though I’ve also heard the rather creative portmanteau “transzorb” (which is apparently a registered trademark of Vishay). These parts are essentially diodes that are placed in parallel with the circuitry that needs to be protected.
Actually, in the case of a unidirectional TVS, it would be more correct to say that it is in antiparallel, because the cathode of the TVS is connected to the positive voltage. As with a Zener diode, the relevant voltage rating here is the reverse breakdown voltage (VBR), not the forward voltage.
The purpose of the TVS is simply to start conducting when the voltage exceeds a certain threshold—and to hopefully not burn up before the voltage returns to normal levels.
This illustrates one of the difficulties in using a TVS. When the voltage goes above VBR, reverse breakdown occurs and the TVS conducts. This clamps the voltage at VBR and diverts the transient current to ground. The circuit is safe now, but what about the TVS? If there is little or no series resistance for limiting the current, the TVS has to fend for itself with whatever (potentially high) currents result from the transient.
This is where manufacturers try to make their components stand out from the crowd. Diodes are not exactly new; I think that most semiconductor companies have the diode thing more or less figured out. The innovation comes into play when these diodes are made capable of surviving large currents and dissipating large amounts of heat.
Peak Power and the 10/1000
The D28V0H1U2P5Q TVS from Diodes Inc. has a “peak pulse power dissipation” of 1800 W; if you look in the datasheet, you’ll see that this spec has “10/1000 µs” in the “Conditions” column. You might also notice that this cryptic 10/1000 (aka 10 × 1000) thing appears frequently in discussions of TVS components.
It turns out that 10/1000 µs refers to a standardized transient waveform:
With a little bit of deductive reasoning, you can conclude from this diagram that the “10” in 10/1000 refers to the length of time from the beginning of the transient to the peak, and that the “1000” refers to the length of time from the beginning of the transient to the “half value.”
TVS manufacturers can’t just say that a particular product can safely handle 1800 W of power dissipation because that implies 1800 joules per second—indefinitely. That’s a lot of power for a little IC; you can boil a pot of water pretty quickly on a 1300 W electric burner. So the 1800 W spec actually tells you that the D28V0H1U2P5Q can safely dissipate peak power of 1800 W in the context of the 10/1000 µs transient waveform.
The marketing information that I saw indicates that this 1800 W spec is 20% higher than the peak power dissipation offered by “comparable solutions.” That’s a bit vague, but if you really need that extra 20%, the D28V0H1U2P5Q is worth a look.
Remember to derate! The peak power dissipation spec is only applicable at room temperature and below.
Another major selling point for the D28V0H1U2P5Q is its status as an “automotive” part. It is compatible with Automotive Electronics Council reliability requirements (i.e., AEC-Q101), and Diodes Inc. describes it as an appropriate part for suppressing transients in automotive applications—more specifically, transients associated with inductive loads. However, I’m sure that the D28V0H1U2P5Q could do quite well in industrial environments, and perhaps in military and aerospace systems as well.
Have you worked with any new TVS components? Feel free to share your experiences in the comments.