12 December 2011

App Note 70 part 1

"A monolithic switching regulator with 100μV output noise: Silence is the perfectest herald of joy..." 72 pages.

Almost two years have elapsed since App Note 65. Jim returns from his short absence with a great app note. This app note is one of the all-time classics, and it is filled with great advice, great circuits, and great quotes.

The primary topic of this note is low-noise switching regulators. The main trick is one that he previously discussed in App Note 29: limiting the slew rate of the transformer drive to limit the high-frequency harmonics at the output. Sharp edges on the switching waveforms give good efficiency (up to 95%), but cause wideband spikes in the output (see Figure 1). The LT1533 regulator, as discussed here, uses slew-rate control on the integrated switches to implement low-noise power supplies.

Of course, Jim spends about one-third of the app note talking about applications and two-thirds talking about measurements and instrumentation. Building a low-noise power supply is easy; verifying the results is the hard part!

The best parts of this note are the quotes. On page 1 he laments the usual difficultly of the task, "Unconscionable amounts of bypass capacitors, ferrite beads, shields, Mu-metal and aspirin have been expended in attempts to ameliorate noise-induced effects."

Most of the circuits (such as Figure 5) are straightforward applications. The meat of the app note begins on page 4 with "Measuring Output Noise". Jim discusses his instrumentation, and he places his usual emphasis on calibration of the measurement equipment. Footnote 7 tweaks the competition for their measurement techniques, "It is common industry practice to specify switching regulator noise in a 20MHz bandpass. There can be only one reason for this, and it is a disservice to users." Note the oscilloscope cartoons in Figures 6 and 8. Although the figures aren't labeled, these "artist's renditions" bear a striking resemblance to Jim's Tek 547.

He also extols the virtues of lab work. In footnote 10 he says, "The noise and efficiency characteristics appearing in Figures 20 to 23 were generated at the bench in about ten minutes. All you CAD modeling types out there might want to think about that."

The second half of the main text discusses a few more application circuits. A negative supply is shown in Figure 24, and isolated supplies are shown in Figures 25 and 26. Battery powered circuits are shown in Figures 34, 35, and 36.

High-voltage and high-power circuits are shown in Figures 38, 40, 42, and 44. In adapting the LT1533 for high-voltage inputs, cascode transistors must be used to protect the 30-volt integrated switches. Jim defines the cascode in footnote 14,
The term "cascode," derived from "cascade to cathode," is applied to a configuration that places active devices in series. The benefit may be higher breakdown voltage, decreased input capacitance, bandwidth improvement, etc. Cascoding has been employed in op amps, power supplies, oscilloscopes and other areas to obtain performance enhancement. The origin of the term is clouded and the author will mail a magnum of champagne to the first reader correctly identifying the original author and publication.

(For those of you who can't stand the suspense, this footnote was answered in App Note 75, footnote 12.)

I'll discuss the appendices next time.

Footnote: Along the same lines of footnote 14, Jim and I tried to find the earliest reference to a cascode topology using just transistors. I have a web page dedicated to the search. Long story short: 1960.


No comments: