Schematics and Components

Schematic diagrams

In an earlier section we started with a logic diagram of an S-R flip-flop, then assigned gates from a chip and filled in the pin numbers. That "marked-up" version of the logic diagram was almost a complete schematic diagram.

A schematic diagram is a drawing (however complex) which fully describes an electrical circuit. All the electrical components and their interconnections are represented. A schematic diagram tells you how to build a circuit and how to check it. In order to do much with digitial electronics, you will need to be able to draw schematic diagrams and to read schematic diagrams created by others.

Every electrical component within the circuit is represented as a symbol on the schematic diagram, or sometimes more than one symbol. Each unique instance of a component is given a circuit-specific name, called a "designator" so you have a way of referring to that component. For example, imagine that your circuit contains two 74LS193 binary counter chips. The chip's part number ("74LS193") is not a sufficient term for identifying the component because there are two of them. By assigning one a designator of "U1" and assigning the other a designator of "U2," you have a way of referring to the specific component. The schematic diagram includes the designator and the part number, both located close to the component symbol.

When a component contains several distinct functional elements, like our 74LS00, it is often divided into a separate symbol for each element. For example, the gate using pins 1 through 3 of the 74LS00 (a single NAND gate) could appear as its own symbol in the schematic, just as it did when we built our flip-flop. All the elements of a chip share the same designator, but generally have a suffix attached to describe the element. If our 74LS00 had a designator of U1, the gate using pins 1 through 3 might be called "U1A" or "U1-1."

But whether one symbol or several, every pin of a component is shown, with its pin name (if any) shown inside the symbol and its pin number (if any) shown outside the symbol. As with logic diagrams, straight lines connect the "pins" where a connection is intended to exist. (When building a circuit, these straight lines become the wires.)

Connection lines may go up, down, left or right regardless of the "flow" of the signal; a convenient convention is to have signals flow from left to right across the page but this is not a requirement (and sometimes makes the diagram complicated; for parts which can flow signals in two directions, such as the 74LS245, it is simply impossible!). Lines are free to cross; this does not indicate a connection. Rather, where the lines are connected, there is a closed dot to represent the join. (You may see a schematic, particularly older analog schematics, where connection lines make a horseshoe-shaped "jump" over other lines to indicate that no connection is intended. However, this is an older convention which in most cases makes the schematic harder to read. Simply be sure in your schematics to make the connection dots heavy and obvious so there is no question of the intent.)

Within a schematic, power connections are a recurring element. Rather than show lines connecting all of these pin, you can use symbols to indicate a connection to the "power plane" (usually a triangle pointing up) or the "ground plane" (usually a set of horizontal lines of decreasing size, looking something like a triangle pointing down). When working with TTL logic circuits, the power plane is assumed to be a 5V supply and does not require a label. (In other, more complex systems where there are a variety of power supply voltages, a label might appear next to the power symbol indicating just which power is being used.) These symbols for power and ground may be used anywhere a signal needs to be connected to that level, not just for chip power supply pins. Remember, it is not a good idea to let inputs "float;" every unused input should be "tied" to a logic level.

An unconnected output, on the other hand, is perfectly all right. However, in order to avoid confusing a deliberately unconnected output with a forgotten pin, schematics sometimes use the convention of drawing a small "X" at the end of an unused pin to indicate a deliberate no-connect.

Sometimes a signal (connection between pins) is important enough that it warrants its own label. This is a simple name placed immediately above a signal line. Naming a signal has no electrical significance; it is simply a convenience for human beings to describe the meaning of that signal. There is also no reason that signal names have to agree with any of the pin names to which they might be attached. If you are familiar with computer programming, you might think of these signal names as being comparable to local variable names within a function.

The sample schematic shown above also includes a connector. There are a lot of different schematic symbols for different types of connectors, but one common convention shows a series of open circles arranged in the same general layout as the pins on the physical connector, each given a pin number. In the sample schematic, a twelve-pin straight line "header" connector is the way signals get into and out of the circuit (a simple eight-bit binary counter).

Given the schematic, you would be able to identify all of the components (this list is called a "Bill of Materials"), and once you had these components (and construction materials such as a breadboard and wires), you would be able to assemble the intended circuit. Further, given an assembled circuit and its schematic diagram, you would be able to understand the function of the circuit and probe its behavior.

Integrated circuits

An integrated circuit is generally represented in a schematic as a box, with pins arranged around the sides in any order that is convenient. The name of the pin is inside the box; the line for the pin extends straight out from the box opposite the name. The pin number is outside the box, close enough to the line for the pin to make clear which pin is indicated.

When an integrated circuit contains more than one functional block, it may make the schematic clearer to split these blocks into separate boxes. Alternately, if a circuit or section of the circuit corresponds to a logic symbol, that logic symbol may be used instead of a box (since it is a clearer representation of the function). However, the power and ground connections for the chip need to be shown somewhere; generally just one of the blocks shows the power pins and their connections. (In the earlier section where we built the S-R flip-flop circuit, none of the NAND gates showed the power and ground connections; in a proper schematic, these would have been indicated on one gate.)

Part designators for integrated circuits sometimes use "IC" as the prefix (IC1, IC2, IC3...), but more often the letter "U" is used (U1, U2, U3...).

Switches

So far in our discussions of digital logic circuits I have skirted around the question of where input signals come from. In our sample S-R flip-flop circuit the inputs were simply wires which we connected to the power or ground bus as needed. But a more practical signal source in a circuit is a switch.

Actually a switch is nothing more, electrically, than the wire we moved. The simplest "on-off" switch makes an electrical path when "closed" ("on") and breaks the path when "open" ("off"). This on-off switch is called a "single-pole, single-throw" switch (SPST) because there is one electrical path (single pole) and only one setting in which the path is closed (single throw). Another useful switch is the SPDT (single-pole, double-throw), which has two different closed paths (from a common pole) depending upon the way the switch is set. We might have used the SPDT switch in our S-R flip-flop circuit. However, it would have needed to be a "break-before-make" switch, meaning (for example) that when sliding the switch, the path to power would have to be broken before the path to ground was made. Using a "make-before-break" switch would mean that sliding the switch would cause a temporary short circuit (where both power and ground were connected to the input, and hence to each other).

The SPST and SPDT switches are modeled on a slide or toggle switch (like a standard light switch). But a more common type of switch for digital electronic circuits is the push-button. A key on a computer keyboard is a good example of a push-button. When you press down the key an electrical path is closed, but as soon as you releasd the key a spring causes the button to move back to its normal position and break the circuit. This is called a "normally open" (N.O.) push-button; a button which is "normally closed" (N.C.) is passing a signal in its normal position and breaks the circuit when pressed. All the push-buttons I'm talking about are called "momentary" push-buttons, meaning they return to their normal state the moment you release them; you may also see "push-on, push-off" buttons in a catalog, but these are really just toggle SPST switches that look like the momentary push-buttons.

Part designators for switches generally use a prefix of "S" (S1, S2, S3...). The standard schematic symbols for these different switches are shown below.

SPST	SPDT	N.O.	N.C.

Resistors

In electronics, any component or even wire has a property called "resistance," which describes the reduction of current experienced by electricity flowing through that component. A component built specifically to present a known amount of resistance is called a resistor. The schematic symbol for a resistor is a zig-zag line, generally having three "points" on either side; the part designator prefix for a resistor is generally "R" (R1, R2, R3...).

Resistor

One use of a resistor in digital electronics is as a "pull-up." Consider the use of an SPST switch, or a push-button, as an input to a digital logic chip. It's no good. When the switch is open, the input is effectively unconnected, and you already know that an unconnected input is an input you can't count on to be in any particular state. The solution is commonly to use the switch to create a path to ground (or not, when the switch is open), and also have a resistor connecting the input to power. The resistor is said to "pull up" the signal when the switch is open; however the current is weakened by the resistor to the point that when the switch is closed, the path to ground is stronger than the path to power and the input "sees" a "0."

When the switch is open, the input sees a "1" because there is a weak path to the power supply. When the switch is closed, the direct connection to ground "wins." Of course, the path through the resistor to power still exists, and that is now connected to the ground, so some current flows through the reistor between power and ground is wasted. However, the pull-up resistor has a high enough resistance that this wasted current is small. In TTL logic, a pull-up resistor value of 4.7K is common; the wasted current is around 1mA.

The pull-up technique can also be used when an input is fed by one or more "open-collector" devices; like the switch above, an open-collector output with a logic "0" provides a path to ground, but if no outputs are "0," the pull-up resistor causes the result to be "1."

Capacitors

A capacitor is a component which can hold a small amount of electric current, like a very short-term battery. When a current is applied to a capacitor the capacitor first "fills" with current before allowing the current to pass through. When current is removed, the charge stored within the capacitor discharges.

The schematic symbol for a capacitor is two parallel lines, or a line and a curve, interrupting the line of a circuit. Some capacitors are polarized, meaning they may only be used in a circuit where one side is known to always be more positive (higher voltage) than the other; in this case a plus sign marks the more positive side. The designator for a capacitor generally begins with a "C" (C1, C2, C3...).

Capacitor		Polarized capacitor

One important use of a capacitor in digital electronics is in smoothing out glitches in a power supply. When a logic gate changes, the current consumed by the integrated circuit also changes by some small amount. When a number of gates change at roughly the same time, the cumulative effect of these changes can be a noticeable difference in the total circuit power consumption; this in turn can have a short-term effect on the power supply, causing the voltage to go above or below the intended voltage. These effects also tend to be localized near the components where the changes are occurring. By placing a small capacitor near the power supply pin of each integrated circuit the localized effects can be reduced. This improves the reliability of the chip and decreases the chances that the switching effects will cause problems for other chips in the circuit.

These "decoupling" capacitors are usually small values, such as 0.1µF or smaller, and are often drawn on the schematic close to the power supply pin. Sometimes the capacitors are shown in their own area of the schematic to not further clutter the drawing, but this can obscure the intent, which is to place the capacitors close to the parts they protect.

Semiconductor components

There are a number of components made of semiconductor materials, just like integrated circuits are, for doing basic tasks. The best-known is, of course, the transistor, a component where a signal at one pin switches a signal between other pins. An NPN transistor takes a small current of a positive voltage at a pin called the "base" in order to let a different current (possibly larger and/or a higher voltage) pass between a pin called the "collector" (more positive) and the "emitter" (more negative). The base current also flows through the emitter, so the emitter is generally a path to the signal ground shared by both signals. A PNP transistor takes a negative signal at the base to let a different current pass between the collector (this time more negative) and the emitter (this time more positive, but still common between the two signals). There are other types of transistors with different symbols, but the NPN and PNP types are the most common in digital logic. One application for transistors in digital logic is to switch a current too high for a logic chip to generate on its own, such as the coil current for a relay.

Transistors		Diodes
NPN	PNP	Signal	LED

Another useful semiconductor component is the diode. A signal diode is the electrical equivalent of a one-way street: current can flow through a diode if the "anode" side of the circuit is more positive than the "cathode" side, but if the cathode is more positive, no current flows.

An interesting variant on the diode is the light-emitting diode (LED), which works electrically like a signal diode, but has the useful side effect that when current is flowing through the diode it emits light. The symbol for the LED is the same as a diode, except with one or two zig-zag arrows moving away from the diode to indicate giving off light.

The standard prefix for a designator for a transistor on a schematic is "Q" (Q1, Q2, Q3...) while the standard prefix for a diode or LED is "D" (D1, D2, D3...).

Oscillators

One other component worth mentioning is an oscillator. This is effectively a module like an integrated circuit which produces a signal that changes state at a predictable rate. The rate is generally quite high in human terms; a "slow" oscillator would be 32,768 cycles per second (32,768Hz or 32.768kHz). Many oscillators run at speeds above one million cycles per second (1MHz).

Since an oscillator is a single component module like an integrated circuit, it is represented like an integrated circuit as a box with a few pins. Typical oscillator modules have four pins; confusingly, many of them are indicated by the manufacturer as having their pins numbered 1, 7, 8, and 14. This seems bizarre until you observe that these modules were designed to fit exactly into the corners of a 14-pin DIP socket, so these pin numbers are based on the DIP package.

There are a number of ways to build an oscillator circuit using various components such as resistors, capacitors, and logic gates or transistors, with a part known as crystal providing the timing accuracy. (In fact, the oscillator components contain just such a circuit, including crystal, inside their package.) However, it would take more conventional analog electronics than I am prepared to include in this course to explain one of these circuits; the easiest thing is to simply buy an oscillator component that does the whole job for you.

Really big schematics

While you may not have thought of it yet, sooner or later you can reach a point where a schematic is simply too large to fit on a single piece of paper. (Using larger sheets of paper only gets you so far.) When the schematic becomes too big to reasonably draw on one sheet, you need to divide the schematic across multiple pages (obviously) but still indicate how connections cross the pages.

The simplest approach to describe is the off-page connector, which is an arrow leading off the page and a name for the signal. The assumption is that on one or more pages there are corresponding arrows with the identical name. A connection is assumed to exist between all off-page connectors with the same name. (If you have a computer programming background, think of the off-page connectors as global variables. The "scope" of these connectors is across the entire project.)

Some common symbols for the off-page connectors are the double arrows shown here:

Output	Input

While it is sometimes a convention to use the arrows pointing toward the signal line to indicate an input to this sheet from another sheet, in fact the arrows pointing outward can be used for outputs, inputs, bidirectional signals or whatever.

A more hierarchical approach to the same problem is to divide your project into functional blocks, where each block fits on a single sheet. At its lowest level a sheet shows a block, where signals in and out of that function are indicated with port pins. The port pin is an arrow indicating a signal connection in or out of the sheet (or both); as with the off-sheet connector, the pin has a name at the opposite side from the signal line.

This sheet is then considered to be its own function block, which is included in the larger project. On some other sheet, there is a box like an integrated circuit whose "pins" are the ports on the function sheet. The box has a name that indicates the sheet containing the schematic details. Each port is shown by name as a pin on the box, so that a signal that connects to the pin is assumed to connect down into the port on the appropriate sheet.

Output	Input	Bidirectional

If you have a computer programming background, the sheet with the single function is comparable to a code function and the ports are comparable to the parameters of the function. The other sheet that shows the function as a box is comparable to the code that calls the function. The scope of a port is local to that sheet, and accessible only through the box representing that sheet on another sheet. Now, experience in structured programming would seem to indicate that the hierarchical approach and ports would be the preferred way to interconnect sections of a schematic. However, in practice this does not work as well as functions in programming for a number of reasons. First, in code you can call a function from several places, but in a schematic it becomes confusing to use the function sheet more than once (because the part designators can't repeat: if you used a sheet with U1 three times you would have a circuit with three "U1" parts, or would connect the U1 part three different ways, neither of which work). Second, not every signal going into a functional block necessarily comes from the same sheet, but apart from power and ground there is no provision for "global" signals. Third, not every good functional block fits nicely on a sheet; sometimes you end up making fairly arbitrary divisions of a schematic not on functional boundaries but on the basis of circuit size.

Some schematic capture tools address these issues with special syntax for repeated blocks, for example. There are projects whose size and complexity demands a better system of managing interconnects than global off-page connectors. But for hobbyist-sized projects, the off-page connectors are clearly the easier way to span multiple pages.

(You will often encounter schematics where off-sheet connectors are used to span pages, but the symbols used are the ones I showed for ports. This is because these symbols are used both as ports and as off-sheet connectors by different engineers. A project actually using hierarchical ports will have at least one sheet looking like a block diagram, showing the connection between sheets; if there is no such diagram, then the project is using simple off-sheet connectors.)

Next: ...yet to be written...

Previous: 74LS00 Family

Copyright ©2003-2006, Mark Bereit. All rights reserved.