Intuitive IC OP Amps

Many observers note that new digital designs are being used in many systems which, in the past, were realized entirely with linear circuits. This apparent extinction of linear products is not showing up as a decrease in linear sales. What is happening is an increase in the perva- siveness of all semiconductor products. This increase in the total application of semiconduc- tors has also greatly increased the demand for operational amplifiers (op amps). A few years ago, if a particular IC op amp from a major supplier sold 100 thousand per month, it was con- sidered to be very successful. This number now approaches 1 to 3 million per month. If we con- sider the total number of individual op amps (duals counted as 2 and quads as 4) the total unit sales of op amps, by just one major supplier, is approximately one-quarter billion per year! This large consumption of op amps and the fact that op amps are basic to many of the more complex linear IC products suggests a greater need for information about op amps today than at any time in the past. Two additional factors are adding to this problem: (1) many times, system engineers have to design both the linear and digital sections on a project and (2) Univer- sities generally have had to reduce the number of linear courses to fit in the new digital (micro- processor) courses. So it is not surprising that books on op amps are still in demand. One goal of this book is to develop an understanding and appreciation for the reasons that have caused such a vast number of different IC op amp part numbers to exist. As will be seen, various applications demand that specific parameter specifications of an op amp be improved. The difficulty of simultaneously improving all the specs - especially where many times the most important one is low cost - has created a large number of op amp products. The idea for this book came as a result of a 1980 nationwide linear seminar where the author gave a presentation entitled "Op Amp Primer." The favorable reception and the many resulting requests for a book that made use of this intuitive approach provided the encourage- ment to create this book. Although many books have been written about op amps, the focus of these books has either been on the detailed design of the internal circuitry of op amps or the rigors of obtaining high precision in linear circuit design. An intuitive groundwork in the basic functioning con- cepts of the op amp has been missing. Intuition involves thinking about physical systems and circuits in an almost personal way. The emphasis on only a mathematical description, that is given during the formal education process, tends to block this physical intuition. This is why experience must be used to acquire the feelings a circuit designer must have. This background is needed before the reader can fully appreciate the way application circuits are really developed by the endangered species of linear circuit designers. Overheard conversations between op amp users reveal that most design is done with cre- ative imaginations and discussions that produce statements like: "When this input is jerked up, this guy up here is kicked ON and dumps a gob of current into this small cap. . . . " These comments sound like the planning of an electronic Rube Goldberg contraption - which is a valid description for most of the really neat application circuits. Th conceive and create circuits in this component-personal way is what the majority of lin- ear circuit design is all about. This type of thinking requires an intimate understanding of and a feeling for the op amp and the passive and active components that are added to provide the complete application circuit. The purpose of this book is to pass along this feeling for op amps, passive components, and op amp application circuits. Toward this end, only relatively simple mathematics will be used so as not to unnecessarily obscure the main issues involved. The major emphasis will therefore be on first order effects and the high volume, popular op amps. Information is fea- tured that will benefit the designer who may have little or no time for research or study and is under pressure to rapidly produce functioning circuits. Interested designers will then be ready to read the extensive literature on this subject to add further details. Chapter 1 traces the op amp to the early analog computers. The specifications of a high quality vacuum tube op amp circuit of bygone days are compared with those of a particular IC op amp that is often used as a reference point, or benchmark op amp, the 741. A simple model is then introduced for the IC op amp and this is used to explain the reasons for both op amp specs and limitations, because understanding the limitations of real-world op amps can save valuable design time for the reader. The evolution of the monolithic op amp is quickly traced to help appreciate the technical problems that were sequentially solved to arrive at the low cost, high performance op amps of today. The Bi-FET (bipolar combined with field effect transistors) op amps are then described. Insights into the design improvements that were made possible with a new process that allowed adding JFETs (junction field effect transistors) to the bipolar op amp process, are given. Most users of Bi-FET op amps are more interested in the increased slew rate and fre- quency response - relatively few users want only the dc benefits. The story of the popular Quads is next. The large volumes that are shipped each month make the quads the industry's most popular linear products. How and why they came into being historically ties to the requirements of the electronic control systems for automobiles. The low prices that have resulted force many designers to use a quad: sometimes it's because they simply can't afford (or perhaps have forgotten how to bias) a transistor! Inner workings are described so the reader can appreciate some unusually good performance specs. For example, the split-collector gm reduction trick, first used on one of these quads, is shown to solve the fabrication problems of the unsuccessful, early, dual 741. This circuit trick has been the key to the modern, small die-size, low cost op amps. Chapter 2 opens with an intuitive approach to explain how feedback is used to control the performance of an op amp circuit. This shows that feedback can most easily be understood as going UP an attenuator. This novel concept has been found to be an interesting and easily grasped idea that also explains why op amp circuits are less precise than a simple resistive attenuator. Feedback control theory is then introduced to describe the op amp application circuits. The requirement for large open-loop voltage gain and the effects of changes in open-loop volt- age gain on the closed-loop voltage gain of application circuits are quantitatively presented. The major op amp error sources, from the imperfection of the feedback components to the nonideal nature of a real op amp, are discussed in Chapter 3. The benefits of the Bi-FET input stage are described and both the large-signal, high frequency, and rise-time limits of both bipo- lar and Bi-FET op amps are covered. Much of this chapter is devoted to a discussion of the undesired noise sources that contaminate the output voltage of the op amp. The similarity between the analysis of the effects of the dc noise sources (that affect V OUT dc) and the ac noise sources is stressed to aid understanding. The performance predictions that can be made by using these noise sources is illustrated in a number of numerical examples. Many nonexperts shy away from considerations of ac noise because of the confusion that results from the statistical nature of noise. The purpose of the material in this chapter is to show how to predict the amount of ac noise that can be expected in the output voltage of an application circuit. Many common misconceptions are pointed out and a novel way to visually display the effects of the individual ac noise sources is given graphically. Surprising results can be obtained: for some applications the 709 may be the best op amp, for others the lowest noise voltage, high cost op amp may not help. The concepts of noise bandwidth and noise gain are introduced and some very practical examples of their usefulness are given. This chapter ends with a discussion of flicker noise (lff) and popcorn noise. Stability, or freedom from undesired oscillations, is the subject of Chapter 4. The numeri- cal measures of how stable (or unstable) a circuit is, the stability margins (gain margin and phase margin) are defined. Ways of testing the stability of an application circuit are given and the effects of insufficient stability margins are covered. A basic introduction to poles, zeros, and root locus is included to explain these terms and to indicate why they are of interest to the linear system designer. This chapter ends with a practical way to guide the op amp user to the cause of an undesired oscillation. Many basic op amp application circuits are presented in Chapter 5. This is not meant to be a complete listing, but will provide the operating concepts on which more complex op amp application circuits are based. Useful application circuits result from combining a few known circuit tricks in an unusual way to accomplish a desired overall function. Practical user problems are the subject of Chapter 6. These are the generally unpub- lished facts that are provided by experience (and many blown-out op amps or application cir- cuits that "hung up"). It is true that data sheets don't list everything. This can be demonstrated by having a linear system designer read a digital product data sheet or having a digital designer read a linear systems product data sheet. Today, this engineer cross-over is happening more frequently, as one designer is often doing both jobs. (If you find that this list is not complete, please send your "gotcha" to the author - in care of the publisher). Finally, Chapter 7 discusses some of the newest bipolar, Bi-FET, and MOS op