High Speed Signal Propagation: Advanced Black Magic
by: Howard W. Johnson
AMAZON multi-meters discounts AMAZON oscilloscope discounts
Topics include:
CLICK HERE for more info and price
"....one of the finest efforts to come along in the field of applied high-speed digital design because of its focus on providing tools for the whole design team bringing a high-speed product to life. For all the PCB designers and circuit designers out there, buy it; read it; keep it." -- Dan Baumgartner, Printed Circuit Design Faster and farther: State-of-the-art signal transmission techniques In High-Speed Signal Propagation, Howard Johnson and Martin Graham bring together state-of-the-art techniques for building digital interconnections that can transmit faster, farther, and more efficiently than ever before. Packed with new examples and never-before-published high-speed design guidance, this book offers a complete and unified theory of signal propagation for all metallic media, from cables to pcb traces to chips. Coverage includes: Managing tradeoffs between speed and distance Physical theory of signal impairments: skin and proximity effects, dielectric loss, surface roughness, and non-TEM mode propagation Generalized frequency- and step-response models Calculation of time-domain waveforms from frequency-domain transfer functions Differential signaling: Edge-coupled and broadside-coupled differential pairs, bends, intra-pair skew, differential trace geometry impedance, crosstalk, and radiation Inter-cabinet connections: Coax, twisted-pair, fiber, equalizers, and LAN building wiring Clock distribution: Special requirements, repeaters, multi-drop clock distribution, jitter, and power filtering Simulation: Frequency domain simulation methods, Spice, and IBIS
Reviews:
The book covers extremely useful material, but in an uneven way. It will especially appeal to readers who want simplified discussions of the ideas rather than deeper more theoretical discussions. For example, crosstalk --one of the most important topics, is covered in a very light manner, recommending use of simulators to obtain accurate results. Compare this with the coverage of the same subject in the remarkable book "Electromagnetic Compatibility" by Clayton Paul. Now, that book is "advanced" and "magical". This book doesn't get into any deep discussions of the fundamentals. It says that current flows in a loop, and inductance is related to loop area. Simple enough. Yet, how do we explain the inductance of the pin of an IC package, which is definitely not a loop? How is the energy-based definition of inductance used by simulators related to this simple definition? Even more fundamental, given that voltage can't be defined in a region where the magnetic field is changing, how come we can use the concept of voltage? (Read Paul Clayton for discussing all these concepts in great detail.) Many previous articles are included in the book. Although some of them are entertaining (e.g. the pot hole story) most are just disturbances, and even worse, outdated and contradictory to authors later approach. For example, Dr. Johnson spends 3 pages! discussing how close a termination has to be, and drives an equation for the amount of reflection when the ideal termination is slightly away from the source. Later in a witty article the author declares that there is no more any excuse for not using simulation tools. He says "If you have to ask, simulate it." Definitely anybody who is wondering if their termination is close enough can download a free demo copy of PSpice and simulate this quickly and accurately.Nobody is going to use the equation the author drived and there is no new insight or theory developed either. The author being the editor of the IEEE Ethernet specs, it is quite surprising how many sloppy mistakes there are. (KHz, kHz, Ghz, HZ,Hertz, hertz, pS, ps, ohm, Ohm, V, v, ; with and without spaces after the value are used with no consistency, "it's" is confused for "its" in more than one place, wrong equation numbers, irregular figures, mixed fonts and fonts sizes within the same equation, etc. are common occurrence.) Despite its problems, this book will be a very useful reference book to people working in the field. Many sections have excellent detailed discussions. For example, performance regions of the transmission line breaking it into lumped, RC, LC, skin effect, dielectric loss, and waveguide dispersion regions is the most comprehensive I have seen. A useful book to add to one's library, but not a book lover's delight.