High-Speed Digital Design: A Handbook of Black Magic

Rating:
Summary: NEVER GET TIRED OF READING THIS BOOK OVER AND OVER.
Review: One of the best books that explain the Signal Integrity concept for the High Speed Digital Circuit. I enjoyed reading all of the chapters of this book. By the end of each section, the authors summarized the important concept of each chapter in a little section called "POINTS TO REMEMBER", I found it very useful.

Rating:
Summary: A disservice to an important and beautiful subject
Review: Superficially this books appears to be a treatise on the subject. But a more careful look reveals a lack of motivation, justification, or proof for any idea presented. Indeed there are no direct references to other authorities so as to allow the reader to either confirm a purported fact or take a deeper look at a particular point. What results, instead, is a compilation of alleged "facts" (none are proved and all are stated without proof) in cook book style. Even more, some of the formulas are not verifiable by spice simulation, and even conflict with other formulas bearing on the same issue. The overall impression of this book is that it is a hodge podge of ideas drawn from other sources, without any coherent theme. In the end Johnson performs a disservice to his readers by promoting the idea that the subject is nothing more than "black magic." It is not, and there are many other books and application notes stretching back decades which bear testimony to this fact. The very essence of the subject is a study of the effects of parasitic elements which become dominant in the presence of high edge rate signals. In this regard high speed digital design draws more heavily upon analog design, with a particular emphasis on rf design techniques. By leaving out all derivations, the reader is never truly educated on this subject and is thereby left in the dark.

Rating:
Summary: Useful book if you need a cook book, however ........
Review: This book is useful if you want to have a long series of equations available in one place to jog your memory. But if you want to learn something useful and practical- and real-world - then perhaps you would be better off doing a web search for application notes, tutorial papers, and articles, particularly from semiconductor manufacturers, and vendors of high-performance test equipment such as Agilent, Tektronix, and others.

To take one example (page 134,) Johnson purports to describe problems associated with a wire-wrapped prototype processor board containing TTL devices operating at high edge rates ( 2 ns.) According to Johnson, the design engineers failed to realize that the circuits would ring excessively, making the board unusable. To "prove" this he posits a model consisting of a 30 ohm TTL driver, with a 2 ns rise time, a 4" length of wire with 89 nH of self inductance, and a 15pf load - a series LRC circuit. Yes, this circuit will ring wildly, but the model is totally incorrect. The TTL input is not considered, which has a relatively low input impedance in the low state since it is current operated. This circuit -effectively a parallel LRC - does not ring nearly as much, as any experienced engineer knows. As a reality check, remember that wire wrap was successfully used for years by thousand of engineers. To listen to Johnson, though, this technology is almost unusable. Wire wrap circuits do ring, but under his example, the real amount of overshoot/undershoot is well within the limits of TTL. Further, no real circuit produces textbook looking traces, so the role of experience is to learn what worst-case design means, and what is acceptable for good manufacturing yield. Lesson: real experience teaches you how to produce correct, functional models. An incorrect model will cause you grief.

Much could have been done here, to be useful, by way of analysis and of recommendation. The wire should have been modeled as part of a transmission line, not as a lumped element, which any high speed digital design engineer would know, and the idea of terminating a transmission line should have been introduced. This is standard fare. Even with the series LRC, instead of deriving the formula for critical damping, he instead says: "This approximation (reduce Q to .5) is derived from the solution to a second order linear differential equation describing an RLC low pass filter. First find the point at which the derivative of the solution passes through zero (a maximum point) and then evaluate the solution at that point."

Got that? Take the derivative of a solution you want to find? Any book on circuits will reduce this to the solution of a quadratic equation. Obfuscating something that's really elementary does not help produce genuine insight. But this is what Johnson does throughout the book.

Isn't it simpler to say that if you have fast rise time signals, treat most connections as transmission lines, and add termination resistors? As for a series RLC, use the formula for critical damping: R = 1/2 (sqrt(L/C))

Rating:
Summary: Useful book if you need a cook book, however ........
Review: This book is useful if you want to have a long series of equations available in one place to jog your memory. But if you want to learn something useful and practical- and real-world - then perhaps you would be better off doing a web search for application notes, tutorial papers, and articles, particularly from semiconductor manufacturers, and vendors of high-performance test equipment such as Agilent, Tektronix, and others.

To take one example (page 134,) Johnson purports to describe problems associated with a wire-wrapped prototype processor board containing TTL devices operating at high edge rates ( 2 ns.) According to Johnson, the design engineers failed to realize that the circuits would ring excessively, making the board unusable. To "prove" this he posits a model consisting of a 30 ohm TTL driver, with a 2 ns rise time, a 4" length of wire with 89 nH of self inductance, and a 15pf load - a series LRC circuit. Yes, this circuit will ring wildly, but the model is totally incorrect. The TTL input is not considered, which has a relatively low input impedance in the low state since it is current operated. This circuit -effectively a parallel LRC - does not ring nearly as much, as any experienced engineer knows. As a reality check, remember that wire wrap was successfully used for years by thousand of engineers. To listen to Johnson, though, this technology is almost unusable. Wire wrap circuits do ring, but under his example, the real amount of overshoot/undershoot is well within the limits of TTL. Further, no real circuit produces textbook looking traces, so the role of experience is to learn what worst-case design means, and what is acceptable for good manufacturing yield. Lesson: real experience teaches you how to produce correct, functional models. An incorrect model will cause you grief.

Much could have been done here, to be useful, by way of analysis and of recommendation. The wire should have been modeled as part of a transmission line, not as a lumped element, which any high speed digital design engineer would know, and the idea of terminating a transmission line should have been introduced. This is standard fare. Even with the series LRC, instead of deriving the formula for critical damping, he instead says: "This approximation (reduce Q to .5) is derived from the solution to a second order linear differential equation describing an RLC low pass filter. First find the point at which the derivative of the solution passes through zero (a maximum point) and then evaluate the solution at that point."

Got that? Take the derivative of a solution you want to find? Any book on circuits will reduce this to the solution of a quadratic equation. Obfuscating something that's really elementary does not help produce genuine insight. But this is what Johnson does throughout the book.

Isn't it simpler to say that if you have fast rise time signals, treat most connections as transmission lines, and add termination resistors? As for a series RLC, use the formula for critical damping: R = 1/2 (sqrt(L/C))

Rating:
Summary: One of the best books I have read on High Speed Design
Review: This book spells out (using math and English) what I had learned the hard way over years of hardware design and takes the black magic out of why high speed systems work the way they do. My favorite practice was to go running for this book as a second opinion whenever someone didn't grasp what I was saying about intricate signal or grounding related issues. From PCB layout, to circuit design practices to how to get your oscilloscope to accurately display the real signal, this book tackles it all..

-Old Commodore Engineer, Designer of the C128, Plus4

Rating:
Summary: Easy to read, not too theoretical.
Review: This book will be a delight for digital designers who don't know much about signal integrity. It is very basic, doesn't use too much math or theory and covers a lot of fundamental useful concepts. There are original discussions that are not seen in many other books such as measurement techniques showing the effects of the ground clip of a probe on the measurement accuracy. Discussions on how to measure inductance and capacitance of traces are also very nice. There are easy to read sections on crosstalk. However, there are also quite a few sections that could have benefited from a more standard text book approach building the subject gradually.

The books biggest weakness is also perhaps its greatest strength: It doesn't go into more rigorous derivation or theoretical discussions. Although this makes the book easier to read in general, sometimes actually it makes it harder to understand what is going on! For example, right in the very first pages in Figure 1.1 the author shows the frequency spectrum of a random digital waveform. There is no explanation of where it came from. I can understand no mathematical derivation existing, but at least the author should try to explain what is going on. It is also unsatisfying how some of the discussions are so intuitive but there is no simulation or theoretical proof of them. For example the author states "Use extra ground planes, not power planes, to isolate routing layers." Although there is some discussion of why this is recommended as usual there is very little to "prove" it by an example.

If you want to learn about very advanced coverage of crosstalk and many other EMI/EMC ideas you should buy Clayton R. Paul's book "Introduction to Electromagnetic Compatibility." By the way that book has the most impressive coverage of how the spectrum of a digital waveform shown in figure 1.1 in "Black Magic" can be calculated.

Rating:
Summary: The best hardware book I read in years.
Review: Very good balance of theory and practice, nice examples from the real world, must-read for anybody designing high speed digital logic and having PCB board design in mind.