Useful Tools

contact us contact tutor/group email to a friend accessibility options report a fault
Concepts of PCB Design

Concepts of Printed Circuit Design

Unit 1: Introduction to PCB Technology

Unit Contents

1.1 Introduction

Although many categories of electronic product exist, this unit focuses on products that involve Printed Circuit/Wiring Boards or Assemblies (PCBs/PWBs or PCAs). The terms PCB, PWB and PCA are used extensively throughout the electronic industry and in academia. Strictly speaking, a PCB or PWB refers to the bare unpopulated board (i.e. without components), whilst PCA is used properly to describe an interconnection platform or board that has completed all stages of manufacture (i.e. components installed and all tests and build operations complete). However, although the terms are often used synonymously, they will, as far as possible, be applied as intended within the context of this text.

The aim of the electronic manufacturing industry has long been to achieve a reliable circuit design with repeatable electrical characteristics, good mechanical properties and to be of an acceptable appearance. Up until the 1950s, electrical/electronic circuits and systems were assembled using individual wires to connect each of the components. The components were then mounted on what were known as tag strips and sockets.

In response to the needs of the consumer for repeatable performance, smaller sizes and above all lower costs, it was necessary to develop assembly schemes that would allow for greater manufacturing efficiency. One method that proved very successful was the use of printed circuit boards to provide the contact between components. These were made from a laminate of an insulating material and were typically about 1.6 mm thick. One side had a layer of copper foil fixed onto it.

The foil was then selectively removed to leave a pattern that interconnected the components in the desired manner. Holes were then drilled through the laminate material to enable components to be fixed to the non-copper side. The components had flexible leads as their connection points and these were passed through the laminate. Electrical (and mechanical) connection was achieved by soldering these to the remaining foil. The foil provided the required electrical connection between the components.

The process met the needs of volume manufacture in that it could be relatively easily automated and created a final product which gave repeatable electrical performance and had sound mechanical strength.

Fig 1. Early Printed Circuit Board: Top with Components/Bottom showing Copper Foil

Fig 1. Early Printed Circuit Board: Top with Components/Bottom showing Copper Foil

Early printed circuit boards were simple designs comprising a small number of components and limited interconnections. Layout level design took place by manually constructing the artworks (or interconnection patterns) for each layer using tape on transparent sheets. Due to only the one layer of connection available to the circuit designer, no connections could be permitted to cross, otherwise a short circuit would occur. These patterns were then photographed to produce the masks1 for fabrication. As circuit densities began to increase it was necessary to allow for more and more layers of interconnect to enable the complexity of design. This resulted in a more and more intricate design problem and it became apparent that some degree of automation would be needed to manage the increasing difficulty inherent in the design process.

1 A mask is the picture that determines where the foil was either to be removed OR left. Both methods can be used.

 

The design and manufacture of modern electronic equipment has taken these principles to the extreme. There are now up to 32 layers of interconnect possible between any two points on a PCB, and, depending on the manufacturing process, even more may be possible. The numbers of layers of interconnect means that the surface area on the board can be much reduced and more components fitted onto the smaller area. The type and shape of components being used in the design are also continually reducing and new features are being added. The board design process must itself be adaptable and able to keep up-to-date with these changes. Finally, the level of complexity of PCBs now means that they can only be effectively designed by using computer-based design tools (CAD).

[Back to Top]

1.2 Early History of the Industry

Dr. Paul Eisler, an Austrian scientist, is given credit for developing the first printed wiring board. After World War II, he was working in England on a concept to replace radio tube wiring, as shown in Figure 2, with something less bulky. What he developed is similar to a single-sided printed wiring board.

Figure 2: Five-Valve Radio Internals

Figure 2: Five-Valve Radio Internals

Based on Eisler's early work, single-sided boards were commercialised during the 1950s and 1960s, primarily in the United States. As the term suggests, a single-sided board has a conductive pattern on only one side. During the 1960s and 1970s, the technology was developed for plating copper on the walls of drilled holes in circuit boards. This advancement allowed manufacturers to produce double-sided boards with top and bottom circuitry interconnections through the holes.

From the mid-1970s through the 1980s there was tremendous growth in the industry. In the same period, PWBs became more complex and dense, and multilayer boards were developed and commercialised. Today, about 70 percent of the global market is in multilayer boards.

The United States dominated the world PWB market in the early 1980s. However, Japan steadily gained market share from the United States, and by 1987 Japan's world market share surpassed that of the United States and continued to grow until 1990. During the 1990s, the U.S. regained a portion of that market, such that the global market breakdown (1998) became as illustrated in Figure 3.

Figure 3: World PCB Market Share

Figure 3: World PCB Market Share

[Back to Top]

1.3 Printed Circuit Applications

It is important that future designers should have an appreciation for the broad variety of systems and products that employ PCAs. There is also a need to understand how these systems and products will be used and the environments in which they will function.

An effective design will result in a PCA that performs effectively and reliably throughout the product's life cycle, which encompasses fabrication, assembly, test, storage, transportation, and operation. During its life, a PCA may be required to survive exposure to a wide range of environmental conditions, such as temperature extremes, humidity, mechanical shock and vibration, atmospheric variations, harmful chemicals, and electromagnetic radiation.

Other major design influences and constraints related to a PCAs application include:

1.3.1 Types of PCAs

Printed circuit assemblies are classified in a variety of ways:

Most PCAs incorporate electronic components and interconnections that perform specific circuit functions.

Some, however, simply serve as system-level interconnections, such as connector motherboards, or flex-circuit wiring harnesses. Printed circuit technology is also being used increasingly to produce integrated electronic components such as multichip modules (MCMS) and microwave integrated circuit (MIC) modules. Figure 4 shows several examples of some of the more common types of PCAs.

Figure 4a: Simple through-hole, double-sided PCA

Figure 4a: Simple through-hole, double-sided PCA

Figure 4b: Analogue Power Supply PCA

Figure 4b: Analogue Power Supply PCA

Figure 4c: Mixed Analogue/Digital PCA with daughter board

Figure 4c: Mixed Analogue/Digital PCA with daughter board

Figure 4d: High Density Digital Surface-Mount PCA

Figure 4d: High Density Digital Surface-Mount PCA

1.3.2 Where and How PCAs are Used

Standards developed by the Institute for Interconnecting and Packaging Electronic Circuits (IPC) identifies the following three classes of electronic products in which PCAs are used:

The basic intended functions of a printed circuit board are to support and interconnect electronic components. For most applications, these functions are required to be performed reliably, consistently, and as cost-effectively as possible. As indicated previously, circuit boards have evolved from being simple, passive parts of an electronic assembly to becoming an active circuit element whose design has a major effect on the performance of a product.

PCAs serve as key hardware building blocks for products found in all segments of the electronics market. The two largest areas of application (in terms of hardware cost/sale value) are communications equipment and computer systems. Other significant PGA utilization product areas include industrial controls, instrumentation, automotive, consumer, military/space and business/retail etc.

A substantial increase in the capabilities and application of digital circuitry [e.g. microprocessors, analogue-to-digital (A/D) converters, and digital signal processors] has resulted in an explosion of new electronic products using this technology. This has considerably expanded the need for complex, high-performance PCAs for use in games, mobile phones, cameras, video recorders, pagers, CD/DVD players, GPS devices and many other products.

In addition, functions previously performed by mechanical or electro- mechanical methods are rapidly being converted to pure electronic implementations. Two examples are appliance controllers and under-bonnet automotive sensors and controllers.

1.3.3 Basic PCA Elements

PCAs are commonly composed of a similar set of basic elements. These are:

  1. A printed circuit board that supports the components and provides interconnections between them
  2. Mechanical parts for mounting parts and hardware to the board, attaching the PCA to the system/product, providing thermal paths, and supporting and stiffening the assembly
  3. Passive and active electronic components that perform the PCA's intended circuit functions
  4. One or more connectors that are the electrical interface between the PCA and the rest of the system/product

The key elements of PCA manufacture, namely printed board creation (fabrication) and component insertion (assembly), may now be explored in some detail.

[Back to Top]

1.4 The Interconnection Platform

1.4.1 Forming the Base Platform

The platform most commonly used in the electronics industry to assemble components upon is the printed circuit board [PCB] (or printed wiring board [PWB]). It is a synthetic, laminated, insulating material to which copper tracks have been added. The steps to create the base platform are illustrated in Figure 5.

Figure 5: Interconnection Platform Formation

Figure 5: Interconnection Platform Formation

There are three core materials used in the formation of the bare board:

  1. resin
  2. reinforcement
  3. copper foil.

The resins are normally thermosetting types such as phenol formaldehyde or an epoxy and are used to hold the reinforcements together to form the laminate. Epoxy is the most popular resin type in current use.

Reinforcements can be made from a range of materials, such as paper (low cost), cotton or glass cloth or mat (high strength and low moisture absorption). Woven glass is the dominant material used in industry. Normally of a continuous web or woven fabric form, these reinforcements are used to provide reinforcing for the board. Additives may be used to adjust the board's coefficient of thermal expansion (CTE).

Copper is the dominant metal for interconnection use. The copper foil is normally specified by weight i.e. one half oz per square ft (152.5 g/m2), oz per sq ft (305 g/m2). These correspond to foil thicknesses of 17.5 and 35 microns respectively. It is usually produced by electrolytic deposition on a flat mandril, and is about 99.8% pure and has a tolerance on thickness of about ± 10%.

The laminating process to form the board is as follows. The reinforcement in web or cloth form is first impregnated with the thermosetting resin by dipping it into a solution of the resin and then squeeze rolling. The laminate is partially cured by heating to drive off solvents so that it is tack-free. These laminates are then cut into sheets. Next they are stacked between copper sheets (for double-sided boards), the number of laminate sheets and thickness of the copper sheet determining the final board thickness and copper weight. The sheets are placed between steel plates in a press where they are heated to 120-170ºC under a pressure of 20-110 kg/cm2.

The resin flows and cures permanently. After cooling the sheets are trimmed, inspected for adequate quality, cut into smaller sheets about 3 feet square and vacuum sandwiched between clear plastic sheeting.

This provides protection for the copper surface, particularly from oxidization by air. Boards are rejected at inspection stage if they have warp and twist, imperfection in the copper surface, or poor bonding between the copper and the laminate.

The final properties of the laminate depend upon the materials used and process control during manufacture. In addition to electrical properties such as dielectric strength and constant, dissipation factor, insulation resistance and resistivity (both surface and volume), there are physical characteristics. These include flexural strength, punching and drilling qualities, flame resistance, and water absorption.

Various standards define how tests are to be performed on copper-clad laminates. These include the AES, BES4584, MILL-P-13949D and IEC249.

One important non-electrical characteristic is the maximum temperature at which boards can be operated. This is also relevant for the curing of printed epoxies and thick film pastes on the boards. This is illustrated in Table 1.

Table 1: Thermal Properties of Laminates
Resin
Reinforcement
Maximum temperature ºC
Phenolic
Paper
100
Glass
250
Epoxy
Glass
120
Polyester
Glass
120
Silicone
Glass
250
Teflon
Glass
200

1.4.2 Creating the Bare Board (Fabrication)

  1. Essentially there are two elements to this task, the design of the board layout and the bare board manufacture itself. It is important to realise that, in addition to the actual copper trace (or track) layout, the designer must supply a considerable amount of other information.

A variety of board types are available, most notably:

Essentially, Multilayer boards start with a double sided board, and new laminate is then added on top and below. Either double sided or single sided boards can be added. The board is thus built in layers from the inside out.

The cross-section of a typical single-sided PCB is shown in Figure 6.

Figure 6: Single-Sided PCB Cross-section

Figure 6: Single-Sided PCB Cross-section

The design documentation must also include the following minimum information:

  1. Board identification number/title.
  2. A statement on the board type: single layer, double-sided plated through hole, multilayer, etc.
  3. Specification for the bare board material to be used, insulation and conductor material, quality and thickness.
  4. Track details for each conductor layer. If there are several layers, alignment marks must be included. The scale is to be given.
  5. Drill sizes and hole positions.
  6. Board finish: final dimensions, solder mask, screened component details, etc.

This fabrication process is illustrated in Figure 7 for a double-sided board.

Figure 7: Major Steps in the Bare Board Fabrication Process

Figure 7: Major Steps in the Bare Board Fabrication Process

A brief explanation of the stages in Figure 7 may now be given:

Drilling

This stage is normally conducted by supplying a NC drill with the information for drilling holes for component mounting, through-hole interconnections (vias), test points etc. A popular industry format for this data is EXCELLON and the usage of the fewest possible hole sizes necessary is advantageous.

Plating

Copper is deposited by both electroplating and electroless techniques, on all the exposed surfaces of the PCB. This stage seeks, in particular, to deposit copper in the barrels of the drilled holes, thus making a solid copper connection between the top and bottom layers on the PCB.

Photolithography

This stage patterns the copper on the board surfaces by a photo-reactive method (i.e. using UV light), in order to form the circuit connectivity. This is accomplished by first applying a coating of photo resist to the surfaces in question, prior to exposure to UV light through an artwork mask. The mask is formed by a photoplotter and provides a connection pattern for one copper layer. The data for this operation is normally in GERBER format.

Tin-Lead Plate

The desired copper pattern must be protected from the subsequent etch stage by coating it with tin-lead. Unhardened photoresist is removed chemically.

Etch

This stage removes copper from unwanted areas and so finally forms the circuit pattern on the board surfaces. This is a chemical process, with the usage of ferric chloride as the etchant being common.

Hot Air Level

An ‘air knife’ or similar device flattens the tin-lead (solder) plate until a uniform, smooth surface finish is achieved.

Solder Mask

Application of solder mask protects the bare copper track from oxidation effects and allows solder only to adhere to the desired points e.g. the component mounting points (or pads). This is achieved by applying the solder resist selectively through a mask in a manner similar to photolithography.

Final stages

The PCB is silkscreened with component identification lettering (usually white) and the silkscreen legend is dried or cured. Any final drilling is done of holes that are not to be plated through and the laminate panel is cut into individual printed circuit boards. The board is then inspected and tested.

The typical end result of these process stages is shown in Figure 8.

Figure 8: Cross-section of Double-Sided Board

Figure 8: Cross-section of Double-Sided Board

Figure 9 shows the cross-section of a more complex type of PCB, known as a multiplayer. Whilst this type of construction is more complex and costly to manufacture, it offer much higher density of routing (copper) interconnection and greatly improved high-frequency performance. As many as 32 layers are possible with this approach.

Figure 9: Cross-section of Multilayer Board

Figure 9: Cross-section of Multilayer Board

It should be noted that boards are rarely made singly. It is normal practice to fabricate bare boards in a panel of laminate material, upon which a number of individual PCBs are created simultaneously.

1.4.3 Populating the Bare Board (Assembly)

The second key operation in creating a PCA, is the addition of the components that make up the electronic circuit on the bare PCB.

The definition of what is classified as an electronic component is a contentious issue. In the limited sense they are the resistors, capacitors, inductors, and semiconductor devices that are assembled onto the card. However, in many instances, this is too narrow a definition.

In reality, the card onto which all other components are assembled is a component in its own right. For multilayer boards, it is probably the most expensive component in the module. The broader definition must include any element that is assembled with others to make up the card or module. Therefore, in addition to the resistors, capacitors, inductors, and semiconductors already mentioned, it includes connectors, test pins, heat sinks, switches, etc. It is not possible to cover every possible component so only with those that perform the primary function (i.e. circuit implementation) will be considered here.

Conventional passive components (resistors, inductors, and capacitors) and active components (diodes, transistors and integrated circuits) are normally produced in a range of standard packages that may be leaded or unleaded. The leaded varieties frequently come in two varieties, through-hole technology (THT) or surface mount technology (SMT).

Passive Components

Passive components are parts that exhibit a fixed or controlled value and perform an elementary function in a circuit. Examples are resistors, capacitors, inductors, and conductors. They are available packaged as discrete devices with leads for through-hole mounting or in chip packages for surface mounting. Multiple passive parts (mainly resistors) may also be enclosed in a single package. Figure 10 shows some typical passive components.

Figure 10: Passive Components

Resistor

Resistor

Capacitor

Capacitor

Active Components

Active components, such as diodes, transistors, silicon controlled rectifiers (SCRs), and integrated circuits, provide variable parametric values to a circuit and can perform specific, high-level activities by functioning, for example, as an amplifier, a switch, a rectifier, a signal detector, or a one-way conductor. Some examples of active components are shown in Figure 11.

Figure 11: Active Components

Transistor

Transistor

Integrated Circuit

Integrated Circuit

Active devices are available as discrete leaded or chip part but the vast majority are provided as integrated circuits (ICs), which contain a large (mostly very large!) number of active components.

These ICs may be packaged for either through-hole or surface mounting and are able to perform a variety of complex circuit functions. ICs are designed for either general-purpose or application-specific (ASIC) use. General-purpose components such as digital logic circuit memory arrays, and operational amplifiers, can be used as building blocks to construct a variety of functional circuits. Application-specific ICs such as microprocessors, A/D converters, or digital signal processors are usually designed for a single purpose or to perform an explicit complex function.

As the preceding figures show, components are packaged in many different physical configurations. The two main packaging families are through-hole leaded devices and surface-mounted parts. Input-output lines (110s) for surface-mounted packages may be provided as leads, plated tabs, or pin or ball grid arrays. Devices are also classified by the type of package material used (i.e., plastic, ceramic, metal), allowable operational environment (i.e., temperature, humidity), critical parameter tolerances, and projected reliability (i.e., operational life).

Component Mounting

Through-Hole Mount Technology (THT)

The technology is characterised by:

Through-Hole Mount Technology is inseparably connected with printed circuit boards.

As an example of the early use of THT, one of the first commercial PCBs, the first pocket radio (1954) may be taken. The radio was to receive AM broadcasts only. It would include, ultimately, just four transistors, in a superheterodyne design. The transistors cost around $2.50 each, approximately $18.00 in today's (2003) dollars - leaving just a small profit margin in the radio. The radio sold for $49.95, or about $490.00 of today's (2003) dollars.

The parts were inserted in the PCB manually. Later semiautomatic insertion equipment and solderwave soldering equipment were developed. Nowadays THT remains in use in some limited applications in the form of mixed technology (THT + SMT). It is used in some products that could not be miniaturised (TV, videos, home audio, etc.). The reason is cost consideration. Typical semi-automatic THT equipment has typically an insertion rate of 750 cph (components per hour).The method of soldering in THT is solderwave technology.

Surface Mount Technology (SMT)

The technology is characterised by:

In the early 1980s the industry began to replace the traditional through-hole mounting technique (THT) with surface mount technology (SMT). Special surface mounted devices (SMDs) replaced the traditional wire-leaded components. The historical roots of SMDs can be seen in the hybrid circuits based on ceramic substrates (middle of 70s) and microwave circuits with "leadless components".

SMD components are placed by a Pick-and-Place machines (Chip Shooter) with a placement rate of up to 40,000 cph. This is a wholly-automated technique. Placement precision is crucial due to the small size and complexity of the components. The outline dimensions of the smallest two-terminal components (resistors or capacitors) are 0.3 x 0.6 mm. Fine pitch multi-terminal components (ICs) have a pitch down to 0.3 mm. The prevailing method of PCB soldering is reflow.

Figure 12 shows the two mounting methods in cross-section.

Figure 12: THT and SMT Mounting

Figure 12: THT and SMT Mounting

Component Fixing

Soldering is a process by which the parts are fixed onto the board surface using a substance (solder) that has a lower melting point than materials of the parts concerned. Soldering differs from (a) gluing that is processed with no melting, and (b) from welding when the joined parts are melted themselves.

All of these three techniques are based on adhesion - a phenomenon of holding together two bodies by intimate interfacial contact based on physical effects (intermolecular forces), which may include also chemical bonding across the interface. “The intimate interfacial contact” is reached by introduction of liquid phase between the parts to be connected (molten base metal in welding, molten solder in soldering, glue in gluing). The liquid wets the parts and solidifies, after that keeping the parts connected, as shown in Figure 13.

Figure 13: Solder Action

Figure 13: Solder Action

 

The solder types under consideration are grouped under the generic heading ‘low temperature metallic solders’. They are based on tin and wet many metals.

The soldering process normally involve two kinds of materials: (a) solder, (b) flux, and generally two kinds of equipment: (a) oven, (b) cleaning machine.

The oven heats flux, solder and the parts to be jointed. The heated flux removes oxides and contaminations from the joining surfaces, following which molten solder wets them. The cleaning operation removes flux residues after soldering.

Lead-free soldering. Only approximately 0.5% of lead is used for electronic soldering (storage batteries 81%; paint, glass, ceramics 4.8%; ammunition 4.7%).

The EU Directive on Waste from Electrical and Electronic Equipment (WEEE) was issued on 10th May 2000. Its purpose is to direct the industry towards a ban on lead, mercury, cadmium, hexavalent chromium, PBB (polybrominated biphenyls) and PBBEs (polybrominated biphenylethers) by the 1st January 2008.

Wave Soldering employs a ‘wave’ of molten solder, over which the pre-heated underside of the PCA is passed. This process coats the component pins and pads with solder, thus fixing them onto the board surface. It is used predominately with THT.

Reflow soldering uses solder in a paste form containing both solder flux and solder. The solder paste is usually stencil printed onto a circuit board at appropriate points so that surface mount component leads rest on the paste. The board is then run through an oven, which heats the board and paste enough for the solder to melt (known as "reflow"). Oven reflow methods are typically used for surface mounted components.

Inspection and Test

Following the assembly operation, PCAs are cleaned by chemicals and de-ionised water. They are subjected to extensive visual and computer-aided inspection. This embodies a full test of the completed board, using either functional or in-circuit methods to establish that each manufactured unit meets the specification. Burn-in techniques, or similar, are also used to expose early-life failures of product.

Completion of this stage allows the PCA to be packaged and shipped to customers.

[Back to Top]

1.5 PCB Computer-Aided Design (CAD)

One of the main driving forces behind the creation of a PCA is 'time-to-market'. The modern emulation techniques mean that there is no need to actually purchase devices and build the circuit on ‘the bench’.

It is a methodology where changes can be accepted until very late in the design process. Time from concept to final device production is thus much reduced. Often the most profit made from a device is in its first six months on the market. This delay can be minimised using an approach where as many aspects of the design as possible are progressed in parallel.

Three distinct groups of CAD tools emerged to assist in the management of this complexity.

Design Entry (or Schematic Capture) Tools

Intel's first microprocessor was drawn on paper sheets laid out on the floor of a rented aircraft hanger! Photoreduction techniques reduced them to the size of a silicon chip - an approach that was acceptable for a relatively modest design, by today’s standards. The increase in the level of complexity of modern complex circuits meant that a more automated mechanism was required. Design entry tools enable a design to be partitioned into a number of manageable blocks that define a hierarchy from system level down to component level.

Design Verification (or Simulation) Tools

Implicit in all circuits of whatever complexity is the potential for design errors. Design verification tools enable the functionality, performance and testability of a design to be evaluated prior to fabrication. This is essential in ensuring that the design will work first time and will remain robust and resilient in service under all operating conditions.

Design Layout Tools

Defining circuit structures at the layout level is particularly complex. A detailed knowledge of device topologies and associated process technologies is required. Design layout tools incorporate a set of process related design rules to ensure that the required structures can be successfully manufactured within the limits of the intended process technology.

The development of Computer Aided Design (CAD) tools for Printed Circuit Board design has evolved alongside the development of CAD tools for Application Specific Integrated Circuit (ASIC). Both design techniques face similar problems in that great complexity must be fitted onto as small a place as possible. There are considerable restrictions on the way in which the designs can be completed. The result is that there are considerable similarities between both families of CAD tools.

[Back to Top]

1.6 Component Symbols and Standards

Before a design can be entered into a computer there must be some form of decision. How are the various components of the system to be represented graphically? Circuit symbols are a simple representation of electronic components. They simplify the task of drawing circuit diagrams. The symbols do not necessarily give an example of the physical make up of the device being represented. A symbol can best be thought of as a block diagram device where the important part of the representation is the inputs and the outputs. Each of the connection points of the block represents a connection point on the ‘real’ device.

Fig 14: A ‘Real’ Resistor and its Symbol

Fig 14: A ‘Real’ Resistor and its Symbol

The more complex the component the more connections will be required. For logic circuits another problem with using symbols is posed. Often more than one gate of the same type is packaged in a single device when the packaging costs of a single gate would be prohibitive. All the devices share the same power and ground connections. It would be very complicated if every time a gate of this type was used in the circuit all the connections had to be shown. It is normal at the stage of drawing the circuit to enter a single symbol and to take account of the number of gates in a package at a later stage in the design process. The example below shows how a 2 Input NAND gate is represented by a symbol but also how it appears in a package, a QUAD NAND gate package. There are four NAND gates available in a single package.

Figure 15: Quad NAND Gate (7400) Implementation and as Symbol

Figure 15: Quad NAND Gate (7400) Implementation and as Symbol

Figure 15: Quad NAND Gate (7400) Implementation and as Symbol

Figure 15: Quad NAND Gate (7400) Implementation and as Symbol

It must be noted at this stage that the symbols themselves are purely diagrammatic representations. To make them appear as if they were real, it is necessary to attach extra data to the symbol. The extra data is normally invisible to the user and is only called upon as and when required in the design process. The extra data to be attached is key to the design process and will be considered in much more detail throughout the course.

The symbols are then ‘interconnected’ using the design capture tool. Modern electronic packages and screens make this very easy to do in real time. The connections are between the ‘nodes’ of the symbols and are used to represent the electrical connections required in the functional circuit.

Fig 16: Physical Representation and Schematic Diagram

Fig 16: Physical Representation and Schematic Diagram

The diagram above shows how the physical components are represented by symbols and the symbols are connected in the schematic representation. The symbols provide a means of enabling the physical components that can be of a wide variety of different shapes. Most devices can appear in different formats but it is not convenient to have a symbol for each.

There is obviously a need to standardise on the symbols being used, otherwise there could be as many symbols as there are shapes. The most common of these in the UK are called British Standard BS3939, which was released in 1975 and updated in 1986. This particular standard contains details of circuit symbols used by industry. The diagram below illustrates some of the more common circuits used in electrical circuit diagrams.

Figure 17: Some Typical Circuit Components from BS3939

Figure 17: Some Typical Circuit Components from BS3939

These tend to be used in most circuits in Europe. There are, however, other standards that are in use throughout the world. The use of these is often both country and customer dependent. There are some differences with those used by the American National Standards Institute [ANSI] system, but most circuit designers can easily follow the differences particularly with regard to analogue components.

It is in the area of logic gate symbols where the biggest variations in standards are found. The most common standards used for logic gates are as given below:

BS3939-1:1986 Graphical symbols for electrical power, telecommunications and electronics diagrams.

IEC 617-12: 1991 Graphical symbols for diagrams, Part 12: Binary logic elements.

ANSI/IEEE Std 91 IEEE Standard Graphic Symbols for Logic Functions (Including and incorporating IEEE Std 91a-1991, Supplement to IEEE Standard Graphic Symbols for Logic Functions).

The suggestions made by the initial version of the BS3939 specification were almost universally ignored. They can still be found in some industrial drawings but the most common symbols for logic gates are still the ANSI forms. The table below illustrates some of the differences between the two standards. For completeness the table includes the IEC form of the logic symbols.

Table 2: Comparison of Logic Symbols
Logic Function ANSI IEC
Buffer Table 2: Comparison of Logic Symbols Table 2: Comparison of Logic Symbols
Inverter/NOT gate Table 2: Comparison of Logic Symbols Table 2: Comparison of Logic Symbols
2 input AND gate Table 2: Comparison of Logic Symbols Table 2: Comparison of Logic Symbols
2 input NAND gate Table 2: Comparison of Logic Symbols Table 2: Comparison of Logic Symbols
2 input OR gate Table 2: Comparison of Logic Symbols Table 2: Comparison of Logic Symbols
2 input NOR gate Table 2: Comparison of Logic Symbols Table 2: Comparison of Logic Symbols

If there is an option the drawing of the circuit representation should be done using the choice of symbols of the customer of the company. At all times, however, standard symbols should be used unless for some very specialist purpose. Like any ‘language’ it is very difficult for everyone to follow non-standard structures.

[Back to Top]

1.7 Component Value Representation

One of the main areas of success of BS3939 was in its recommendation of how to represent the specific component values on an electrical diagram. It is a streamlined version of the SI unit system. A simple easily printed code indicates the electronic component value clearly with the minimum of characters and without the possibility of misunderstanding if poorly reproduced.

It was this later restriction that led to the removal of the decimal point as a marker of magnitude and the compression of the required data to minimise the length by removing where possible unnecessary or obvious information. The aim was achieved in two ways:

  1. Where it was obvious from the symbol what the device was, and hence the units of its value, they could be ignored. This meant that 800Ω would simply be given as 800 on a diagram when attached to a symbol for a resistor.

  2. All values would be qualified by a multiplier in terms of powers of ten. e.g. x103, x106 or x109

The table below lists the symbols used within the range of typical components.

Table 3: International Multiplier Symbols
Multiplier
R
C
L
f
Time
109
G
G
106
M
M
103
k
k
100
R
F
H
Hz
S
10-3
m
m
m
10-6
µ
µ
µ
10-9
n
n
10-12
p
p

This can best be further understood by considering how it is applied to the three most common values of circuit component; resistance, capacitance and inductance.

1.7.1 Resistance

The standard unit of resistance is the ohm, Ω. In 1975 when the BS3939 specification was first published there were few typewriters with the symbol Ω. It took considerable time to add to a drawing by hand. BS3939 recommended that the symbol Ω should not be used on drawing as a postscript when denoting resistance values but that units of resistance should be denoted by an upper case R. When the value related to thousands of ohms the lower case letter, k indicated the value and so on as per the table above.

e.g. in this representation method 56Ω, 56 kΩ and 56MΩ become 56R, 56k and 56M

1.7.2 Capacitance

The standard unit of capacitance is the Farad, F. The value of 1 Farad is much too large for most electronic circuits. It was the norm to use two multipliers, 10-6 (µ) and 10-12 (p) when dealing with values of capacitance. BS3939 called for the capacitance value to be represented using one of the decade multipliers µ, n and p.

In circuit diagrams, when indicating the value of a capacitor, it is self-evident that the component’s value is expressed in Farad, and it is therefore unnecessary to show this on the diagram.

e.g. in this representation method 4700µF, 47µF, 0.047µF and 47pF became 4700µ, 47µ, 47n and 47p.

1.7.3 Inductance

The standard unit of inductance is the Henry, H. It was the norm to use two multipliers, 10-3 (m) and 10-6(µ) when dealing with values of inductance. BS3939 called for the capacitance value to be represented using one of decade multipliers m and µ.

In circuit diagrams, when indicating the value of an inductor, it is self-evident that the component’s value is expressed in Henry and it is therefore unnecessary to show this on the diagram.

e.g. in this method 3.5H, 43.5 mH and 3.5 µH becomes 3H5, 43.5 m, 3.5 µ

[Back to Top]

1.8 Capturing the Circuit Diagram

The user starts to create the integrated design project by capturing the circuit in the form of a diagram. Symbols are used to represent the components that make up the circuit. They are interconnected by the addition of ‘wires’ that connect between the nodes of the symbols and are used to represent a zero Ω electrical connection between the two parts.

1.8.1 Interconnection

As has been mentioned above some of the symbols will have different effects from others. It is difficult to represent a voltage source on a diagram and signals that flow onto and off a circuit. It is usual for the circuit to have some form of connection to these. These are not true components but must be shown on the diagram never less. It will be shown that some symbols are purely to aid the simulation of the program operation at the next stage of the process.

Figure 18: Sample of Circuit Capture showing Interconnections

Figure 18: Sample of Circuit Capture showing Interconnections

The parts available to the user are stored in the ‘parts/symbols libraries’. These are data files where all the information relating to the specific components being chosen by the designer are stored. The information here is usually provided by the software supplier but can also be customised by the designer for components that are unique or not used commonly.

1.8.2 Part and Symbol Library Information

Libraries also hold the information that links the symbolic representation to the ‘real’ device. It has information about its physical packaging as well as its electrical performance. The diagram below illustrates some of the typical information held about a device. The particular gate that this data is held for is the 7400 NAND gate.

Figure 19: Library Information for 7400 NAND Gate

Figure 19: Library Information for 7400 NAND Gate

At this stage not all of the data above will be explained - the most important parts only will be considered.

Note also the format as shown above is relevant to one specific package. Other CAD packages will present the information in a different format.

Symbol

The most used part of the library data is the symbol. It is this part that is used to produce the schematic diagram and enables all concerned with the circuit to be made aware of the purpose of that part.

Fig 20: Symbol for NAND Gate

Fig 20: Symbol for NAND Gate

Packaging

The packaging information determines what physical representation the device will take. It is this data that joins the symbolic and the real world. It is not used at the schematic capture stage when the designer is simply interested in the circuit functionality but is key to the practical implementation of the circuit. Similar packages are used for many similar products, particularly digital devices.

Figure 21: ‘Raw’ Information for DIP 14/300 Package

Figure 21: ‘Raw’ Information for DIP 14/300 Package

 

The user can modify a label depending on the requirements of their own specific circuit information. The information in uppercase is fixed for every usage of this device in the design. The information in lower case is changed depending on how many of the particular packages are in the design; each package will have a unique identifier. Pin numbers are also shown as optional information.

Simulation

When testing circuit performance it is ideal if the information regarding the circuit interconnection and hence its functionality is only entered once. The library should thus include any data that will be used in the simulation of the operation.

Figure 22: Typical Simulation Information for Product

Figure 22: Typical Simulation Information for Product

 

The information above shows the simulation data to allow two particular simulations to take place; Electrical and Thermal.

The upper section relates to the electrical performance simulation and includes details of two models; a Mixed Mode Function (MM) and EDSpice Simulation Model. It is unusual to have two models but these relate to different simulation packages.

The lower section is for carrying out an analysis of the thermal performance of the circuit and lists the necessary information for this particular DIL package. The thermal parameters are obviously tied in with the packager shape being used.

[Back to Top]

1.9 Simulating the Circuit

Traditionally the verification of a circuit’s operational characteristics is by building the circuit on a breadboard or equivalent and subjecting it to a series of tests using a range of inputs and equipment. This can be a very expensive and slow process. When the designer has completed the tests and is happy that the operation will be within specification he then commits the design to paper for the layout designer to produce a PCB.

Considerable time can be saved if the designer can enter the circuit schematic and from this test the circuit operation. Not only can the circuit be tested in less time but with more accuracy and performance can be assured over a range of component values and timing scenarios.

After completing a schematic diagram, simulation may be performed on the circuit. In simulation a mathematical approximation to the component's operation is used to predict how it will operate. Most circuits will consists of several components and the operation of each must be predicted. Given that the input signal to one will be the output of the other the number of calculations that must be completed are soon extremely large. It is only recently that it has been possible to complete these calculations on a PC in a reasonable timescale. Even now it is still not unknown for designers to leave a circuit to simulate ‘over the weekend’.

The diagram below shows the result of the analysis of the operation of a digital logic circuit. It illustrates the predicted voltage levels and the time when they occur.

Fig 23: Simulation Results for Digital Circuit

Fig 23: Simulation Results for Digital Circuit

Several benefits occur because of the use of mathematical models both for the circuit designer and the PCB designer. One of the simplest and often most useful checks is that the diagram entered at the schematic stage has all the connections that the designer expects!

Different types of analysis may be performed to find out how component placement, routing connections, changes in temperatures, etc. may influence the functionality of the circuit. Not only can the electrical performance be predicted but also thermal analysis can predict any problems and EMI performance can be predicted before design is committed to manufacture. The diagram below shows the result of an electromagnetic analysis.

Fig 24: Electromagnetic Analysis of Circuit Design

Fig 24: Electromagnetic Analysis of Circuit Design

[Back to Top]

1.10 Designing the Layout

When the schematic has been fully entered and the simulation has been completed, the next step is to place the components onto the designed PCB and then determinine the pattern of the interconnections between the components.

The physical size and shape of the printed circuit board has to be defined first at this stage. There are several standard board shapes that have been developed. Popular board shapes are ISA (Industry Standard Architecture) for the PC; other standards include Eurocards and VME.

The board shape may be fixed by any number of constraints. It need not be a simple rectangular shape, it may be flexible, and it may have a complex shape to fit into a complex package. The importance of the board is that it determines the maximum number of components that can be mounted on the board, and it must agree with the number in the schematic. It there is not enough room then changes must be made in the product.

The physical shape and the final package define the position of some components. Connectors, displays and switches have their position fixed and there is no variation in this. When working with a predetermined card size, the choices are narrowed from what will be an optimum layout to what is the best that can be done with the space available.

1.10.1 Component Placement

Placing components on a PCB is not a straightforward task. There are several points that must be taken into consideration when placing the components on the board space; signal path length, clock skewing and unconnectible traces are typical problems. Placement can either be done manually or by use of an automatic placement algorithm.

Manual Placement

Manual Placement is used when specific design features are to be implemented. Key components can then be grouped together and placed in close proximity to one another. There are various features that can be used to help with placement. A ratsnest1 of connection is a very useful aid to enable a neater placement of components. The diagram below shows a ratsnest for a simple circuit.

1 A ratsnest is simply a jumble of skewed lines that show how each part is wired to every other part of the circuit

 


Figure 25: Simple Placement showing Ratsnest Connections

Figure 25: Simple Placement showing Ratsnest Connections

Actual placement of the component depends on several factors, many of which are interactive and rely on where the previous part has been placed. Most important, as has already been mentioned, is the placement of connectors, switches, displays etc. Exact positional placement is usually critical for this type of device. A connector is one of the least flexible components and may even consist of simply a pattern etched into the copper surface (e.g. an edge connector).

 

After the fixed components are in place the design must be considered as a single piece of work. There are often associations between specific components; e.g. integrated circuits and decoupling capacitors that must be adjacent to have any benefit. Where such an association exists, the parts should be placed close together.

Automatic Placement

Automatic placement programs are normally based on an algorithm that calculates the routability of the pattern for a given interconnection pattern. It will examine several different options and present a "maximally" good layout. It should be noted at this stage that the placement program is only as good as the algorithm and that there will be wide variations in package performance. The package will not necessarily be very acceptable in terms of manufacturability and presentation, as is shown in the diagrams below.

Figure 26: Component Layout using Autoplacement Algorithm

Figure 26: Component Layout using Autoplacement Algorithm

Figure 27: Component Layout Preferred

Figure 27: Component Layout Preferred

As can be seen from the above, routability does not correlate well with interconnect length. It has the advantage that it is easy to calculate and therefore, runs very fast. It does tend to bunch the components closely together in the middle of the board. Unless the designer has added specific instructions about spacing manual separation of the components will be required.

Many packages will give a guide as to whether a design is routable with some form of routability analysis. The analysis will highlight any areas of the board where it may be difficult to complete the required interconnections as per example below.

Figure 28: Connection Density Predictions

Figure 28: Connection Density Predictions

Generally it is a mix of both methods that is preferred. The autoplacement means that a first pass can be quickly achieved which can then be passed through as many manual iterations as are required to achieve a potentially routable design.

1.10.2 Board Routing

Routing is the act of the placement of copper tracks on the PCB for electrical connection. At this stage there are several design factors that must be implemented although these should have been decided earlier.

The width of the conductors and their spacing determines the electrical characteristics of the connection. This is mainly because the width of the conductor determines its current carrying capacity. This particular factor will be investigated in much more detail later in the course. Conductor spacing is an equally important parameter, the distance between adjacent edges of conductors on the same layer. It is a measure of the insulating properties of the board. The further apart the conductors are, the higher the voltage that can be carried on adjacent tracks.

The number of layers to be utilised is another key feature of the design. The more layers present, the easier it is to route the design. However, more layers makes the circuit more expensive to build - more holes will be required and the drilling costs will increase.

Other factors must also be taken into account but these will not be considered at this stage.

Manual Routing

Manual routing is quite straightforward. The required interconnection is selected for routing and the connection between both ends made. It is usual to route key connections first, i.e. Vcc and Ground as these can be most important to the electrical performance of the final product.

Any key connections in terms of length or placement are then routed as necessary. The procedure is then done time and again until the board is fully routed and all interconnections have been made. Manual routing is very time consuming but can give a very effective and often optimal solution. The diagram below shows a partially routed circuit that has been done manually.

Figure 29: Partially Manually Routed Circuit

Figure 29: Partially Manually Routed Circuit

Auto Routing

Autorouter software uses the CAD netlist as its input as to how to generate the circuit interconnections. It compares the placement of the components with the netlist details and decides which trace should go where. . Autorouters find paths for the connections by following a prescribed set of wiring rules and following these much more speedily than a human operator.

The set of rules that can be produced is somewhat limited and the result is not always what the designer requires. The autorouter does not consider the manufacture of the board and can often have redundant connections that lead to a poor looking design. It is usually necessary for the designer to manually tidy up the design and enhance the finished product. The diagram below illustrates how an autorouter does not always produce the best result.

Figure 30: Example of Autorouter Weakness

Figure 30: Example of Autorouter Weakness

The fact that an autorouter does not always do as good work as expected must be realised and accepted by the PCB designer where time to market is limited. Some of the most modern autorouters do a much better job than some of the others; the cost of the product becomes the limiting factor. Where many thousands of connections are to be made, it can be the only economic method.

It is usually a mix of both automatic and manual methods that achieve the best results in terms of an efficient and manufacturable design. The diagram below shows a board of 202 connections autorouted for a two-sided board in 35 seconds on a PC. Try doing that by hand!!!

Figure 31: 100% Routed Two-layer PCB

Figure 31: 100% Routed Two -layer PCB

1.10.3 Design Rule Checking

Design verification is a very important step because mistakes that pass through here end up as mistakes on the PCB and these are very costly to correct. The process checks to see that tracks, holes and connection points have all been placed according to the set of rules for the design.


[Back to Top]

1.11 Fabricating the Layout

Usually fabrication of layout is the final step of circuit design. The ultimate goal is to produce artwork that can be used to make a functional printed circuit board and generate the data to drive a numerical controlled drilling machine and generate the co-ordinates and package data to drive a pick-and-place machine.

The artworks contain the trace patterns for each layer of the board. Each layer in the ‘sandwich’ of the board will have its own unique artwork. Artworks will also be produced for the Solder Resist Layers, for top and bottom and for any Silkscreen layers required.

The drill data will consist of a series of co-ordinates and drill data relative to some predefined datum. Production documentation will also come under the generic term artwork.

Data can be produced which can be used with pick-and-place machines to populate boards and generate component lists.

More sophisticated packages will be capable of producing a set of test vectors and values for test during manufacture.

The following diagram shows one of the outputs at this stage.

Figure 32: Sample Artwork

The data generated at this stage is normally straight from the package. A few options may require to be changed depending on the data required by the PCB manufacturer chosen to complete the design.

[Back to Top]

1.12 Summary

A PCB is a key element in any modern electronic circuit. It is often the only component uniquely designed for the product. As technology has progressed over the last 50 years, the PCB, or PCA, has gone from being a simple, manually laid out, connection platform to being a complex and sophisticated computer-designed product in itself.

The PCB designer must be familiar with not only the concepts of board design, but with all the ancillary elements of other related engineering disciplines as well. These embody mechanical, electrical, chemical, thermal and material engineering principles. Thus the designer should be involved with the process from the concept to its final manufacture. They should work closely with all staff and must ensure that the PCB meets all the specifications that would be expected of the product.

The layout of the board must be done with due consideration being given to many different factors, e.g. component spacing, complexity of routing, track size requirements etc.

Documentation must be produced which meets the necessary standards and can be understood by all of those involved. The final documentation sent to the specialist PCB manufacturer is the designer’s responsibility, both to ensure its correctness and the formats being used.

Complex boards must be designed in an appropriate time span, given that a new design makes most of its profit in its first six months of release (before the competitors can catch up), any reduction in the time-to-market for the product will increase profitability. CAD software is used to ease the tasks of the PCB designer.

However, it must be remembered that the usage of such CAD tools are only an aid to product realisation and considerable skill is still required by the PCB designer to maximise the efficiency of the process and produce a reliable, cost-effective and manufacturable board.

Self Assessment Questions

Question 1

What are the classes of electronic product? Give examples within each classification level.

Question 2

What are the key elements of a simple bare PCB? Give examples of the material used to realise each element.

Question 3

What is the function of photolithography in PCB fabrication?

Question 4

What are the main differences between Through-Hole Mounting (THT) and Surface-Mounting (SMT), as methods of attaching electronic components to bare PCBs?

Question 5

Define the three main sections within the PCB CAD toolset

show solution

[Back to Top]

01-02-2005 RA

Site Search

Powered by Google
Site Map