This section is especially intended for any students who are unfamiliar with the mainstream style of board multilayer construction; (FR-4 epoxy-glass laminate material; solder or nickel-gold finish) and standard solder processes for through-hole and surface mount assembly.
The aim is to give you a context in which to study this module. This is important, because we have tried to use examples that relate to standard practice. But practices vary throughout the industry! If you are new to it, we would particularly suggest you spend time with the activity which comes before Points to think about.
If you have “been around for some time”, then you will have visited a number of fabricators and assembly houses, and will probably find the material pretty basic. However, even if you are familiar with the processes, it would be well worthwhile checking that you can draw flow diagrams both for the fabrication of a multilayer board and for assembly of a mixed-technology board. We should also like you to read the final section, Points to think about, because this will explain something of the module background.
For most assemblers, a PCB presents as a piece of fibre-glass type material, generally coloured green, bearing a conductor pattern finished in copper, solder or nickel-gold. It will often be part of a ‘panel’ (Figures 1 & 2) of several circuits which must be separated (‘de-panelled’ or ‘broken out’) at a later stage in the process. Once all components have been mounted on the board and soldered to it, the individual circuit can be powered up and tested.
Despite this simple appearance, the PCB should be regarded as a circuit component of considerable complexity. It has the ability to be ‘tailored’ for specific applications: the final application of the board has a significant effect on its design and layout, and the choices of materials, finishes and base material are affected by the initial circuit design. The manufacture of PCBs correspondingly employs a ‘kit-bag’ of methods, rather than a single process, although there is much commonality in procedures.
[ back to top ]
The Printed Circuit Board is the base on which the electronic circuit is assembled. It provides:
and may also:
Now almost universally adopted as a means for reliable, automated, high component density circuit manufacture, PCB technology is mature, and in volume production by specialist manufacturers, who have the necessary expertise and experience to ensure a consistent build quality of what, in many circuits, will be the only component unique to the product being made. IPC have reported that the board typically accounts for 10-20% of the total cost of the printed board assembly.
[ back to top ]
The distinction is made between ‘pin-through-hole’ components, whose terminations are inserted through holes in the interconnecting PCB, and soldered on the opposite side of the board from the component, and surface mount (SM) components, whose terminations are soldered to conductive tracks on the surface of the board. The choice and mix of components primarily determines which assembly sequence and processes are used.
The process stages used in the assembly of pin-through-hole devices are shown below:
The process stages used in pure surface mount assembly are shown below:
For a double-sided surface mount assembly, the processes are repeated on the reverse, and a typical process sequence is shown in Figure 11. During this second process, the components on the first side are:
There are substantial differences between through-hole and surface mount processes as regards both soldering and board practice:
Many Printed Circuit Assemblies are mixed, that is they contain both leaded (or ‘conventional’) and leadless components. Three of the assembly sequence variations are so common (Figure 12) that by convention they have acquired titles:
|Type I||single-sided surface mounting|
|Type II||mixed technology with single-sided surface mounting|
|Type III||mixed technology with double-sided surface mounting|
The choice of an assembly process has to be made on an individual basis, and depends most critically on the characteristics of components – their availability, cost and mechanical format and their ability to withstand processing. Some of the many possible process routes are shown in Figure 13: the option selected will also be influenced by design constraints, the capabilities of the equipment and the requirement to balance the line.
Where surface mounted components are to be placed on both sides of a mixed assembly, it is usual to separate the soldering of components on the top and bottom sides of the board. A typical process is to apply solder paste, place SM components on the top of the board, dry the paste, and reflow-solder the components. Next, leaded components are inserted, the board is inverted, adhesive is applied, SM components are placed, and the adhesive is cured. After inverting the board once more, wave soldering completes the process. A typical process sequence is given below.
1. On the top side of the PCB
2. On the bottom side of the PCB
3. Carry out manual assembly
4. Wave solder
5. Inspect and touch-up
6. In-circuit and functional test
7. Rework where necessary
[ back to top ]
The growth of the electronics industry in post-war years led to a massive increase in the use of electronics. Circuits were required to have better performance, small size became important (compare a radiogram with a modern micro hi-fi system!), and systems had to cost less.
The drive towards automation and mass production meant that new manufacturing techniques were needed, and a method had to be developed whereby components could be both secured mechanically and interconnected electrically easily, quickly, reliably and repeatably.
The PCB, first patented in its current form in 1943, did this by providing a suitable rigid base for mounting components and copper tracks for interconnecting them. The key advance was bringing all the solder joints into a single plane, which eased the soldering task. Enhanced by mass soldering techniques (dip and drag soldering, and later wave soldering) and by the automated insertion of components, the problems of labour-intensive circuit assembly were overcome and PCBs quickly became absorbed into all areas of electronic product design.
Early boards consisted of a base insulating laminate 1/16 inch (1.6 mm) thick with one side covered by a layer of copper foil, which was patterned by etching to create interconnection tracks or ‘traces’. Components were placed on the non-coated side and ‘through-hole’ leads soldered onto the copper foil, the point of connection to the board being called the ‘pad’.
On practical circuits, some connections need to cross each other. With a single-sided board, this ‘cross-over’ function is provided by running tracks between component leads and by bridges formed by the components themselves (or by wire links). Both design freedom and component packing density are constrained by this, and further layers of connection are desirable.
The first small step which allowed denser component placement was to make a ‘double-sided’ board, with copper tracks on both sides of the laminate. Adding the extra layer allowed an increase in component density without the need to reduce the width of the copper connections (the trace size).
Early double-sided boards used component leads, link wires or small rivets to make the electrical connections between layers, but this was expensive. A major advance was to metallise the walls of the holes to provide side-to-side connections between the layers without wires. Note that electroplating will only work once the hole walls have been made conductive, which has to be carried out by methods such as electroless plating.
‘Plated Through-Hole’ (PTH) technology is now almost universal for double-sided boards, normal practice being to plate all holes except any tooling holes used for board alignment during manufacture.
Further increases in component density can be achieved if more than two layers of track are used. Board technology was extended to give this facility by laminating layers together; two layers of connections remain on the outside of the board (as with double-sided boards) and the others are internal, with connections between adjacent layers where required. Most boards in high technology commercial use nowadays are ‘multilayer’ in construction. Of course, you can only see the outer layers unless you take an X-ray image.
Holes specifically designed only to provide connection between different layers (rather than for component or wire insertion) are called ‘vias’ (Figure 18). Holes which pass completely through the board are known as ‘through vias’. In some types of boards, internal layers can be interconnected by vias which do not pass completely through the board, known descriptively as ‘buried’ or ‘blind’ vias.
A ‘through via’ is similar in construction to a component hole, but is frequently much smaller; size and packing density depend on the limitations of the drilling and patterning techniques.
[ back to top ]
The starting point for a printed circuit board is a large sheet of laminate covered on both sides with copper foil. Depending on the size of the final board, this ‘process blank’ may eventually make more than one board, and small circuits may be arranged in groups in ‘panels’, for the convenience of the assembler. Panels allow faster operations, and more consistent handling, jigging and conveyor setting.
Merlin Circuit Technology
There are a number of variations, but the usual process steps for a double-sided through-hole plated board are given in Table 2:
Cut, edge and clean panels
NC drill all holes
De-smear and electroless copper plate all exposed surfaces
Clean board surfaces and laminate primary dry film resist
Expose and develop dry film
Electroplate approx 25µm of copper and 4–10µm of tin
Strip primary resist
Etch exposed areas of copper
Strip tin/tin-lead deposit
Brush clean copper pattern
AOI scan if high density
Apply photoimageable liquid solder mask and dry
Expose, develop and cure solder mask
Screen print component identification
Apply solderable finish (solder level, electroless nickel-gold, etc)
Trim board to final size required
Final visual inspection
PTH technology led to the development of multilayer boards; four layers is the simplest structure, and most computer boards have at least six layers of interconnect. Multilayer PCBs are made by ‘laminating’ a number of boards together. The internal boards which create the final multilayer are produced separately, usually on thin laminate with foil on both sides. The patterns are etched and inspected, and then chemically treated to promote adhesion. The outer surfaces are of plain foil or single-sided unpatterned laminate.
Merlin Circuit Technology
The bonding process uses partially cured ‘prepreg’ or bonding sheets which are interleaved between boards to provide adhesion and insulation between layers, and involves both heat to cure the prepreg and pressure to ensure a void-free structure. Some prepreg is normally squeezed out, and is removed by cropping. After ‘laying up’ and lamination, the resulting composite boards are drilled, and the outer surface patterned using the PTH process described above to connect all the layers together.
Photoresist application Volume production of simple non-PTH boards usually involves the use of wet etch resists or screen printing of wet etch resists. For higher technology requirements, dry film photo resists are used almost without exception.
The most commonly used photoresists are negative-acting, which means they polymerise on exposure to ultraviolet light and hence become insoluble in a direct developer. Non-polymerised resist is removed by ‘developing’ to expose copper areas ready for electro-plating.
A fundamental problem with flexible photographic films is that their size will increase with temperature, typically by 25ppm/°C. There is also a more complex non-linear response of the film base to humidity.
Merlin Circuit Technology
Merlin Circuit Technology
Copper plating Electroplating is a cheap and effective way of building up a layer of copper with almost bulk metal properties, but needs contact with a fully conductive surface. More complex and costly ‘electroless’ and ‘direct’ plating processes have, however, been developed which will plate copper (or nickel, or gold) onto any surface.
Etching is a process of chemically dissolving unprotected parts of the copper foil to create the desired pattern. Unlike silicon etching, the etching of copper takes place in all directions at similar rates, so that there is a tendency to ‘undercutting’, particularly with thicker layers.
Merlin Circuit Technology
The choice of finishing processes depends very much on the application. Whilst a solder finish can be achieved simply by reflowing the tin-lead plating, a somewhat flatter surface can be made by Hot Air Solder Levelling, where the board is dipped successively in flux and molten solder, and air knives are used to remove surplus material during withdrawal.
Merlin Circuit Technology
This ‘HASL’ process is cost-effective and very common, but many applications demand a flatter plated surface. One way of doing this is to strip the tin (or tin-lead) resist and plate a nickel underlay followed by a thin gold flash (referred to as ‘nickel-gold’ or ENIG).
An alternative is to protect the copper after stripping by a thin coat of Organic Surface Protection (OSP) material. Although OSP processes usually look just like the underlying metal, these ‘bare copper’ boards retain their solderability for a much longer period than would the untreated foil.
As with the PTH board, solder resist and component identification print follow, the board is trimmed to the final size required, and boards are inspected for defects and electrically tested.
The base laminate consists of resin plus reinforcement, and its properties are determined both by the materials and their ‘lay-up’. For any application, the material selected from the very wide range of laminates available, will be determined by:
as well as by the usual constraints of quality, cost and availability.
Woven glass fabric continues to be the most widely used reinforcement in rigid laminates as it has:
The most common laminating resin used is an epoxy (properly ‘epoxide’), and the resultant epoxy-woven glass laminate is referred to as ‘FR-4’. This designation indicates that it has fire retardant properties. One must bear in mind, however, that ‘FR-4’ is a generic term, and covers a wide range of materials of differing qualities.
Copper is almost always used as the base conductor material, although its appearance is usually modified by a surface finish of solder or nickel-gold. For a moderate price, copper combines excellent electrical and thermal conductivity with ease of plating, good ductility (important for through-hole plating), and stable oxides with excellent adhesion to resin systems.
Board surfaces are normally coated with solder resist (‘solder mask’), which defines the areas to be soldered. Apart from the immediate reduction in the amount of solder used and the incidence of bridging, solder resist also enhances the insulation and humidity resistance of the board. Resists need excellent solder and solvent resistance, and are frequently epoxy materials applied by one of three techniques:
A pad is the connection point at which a component joins a PCB conductor. In through-hole designs, it is conventionally a circular or rectangular area at the end of a track, with a ‘plated-through’ component mounting hole through its centre. Solder is applied to the lead and the surrounding copper, to secure the component and establish electrical connection. The diameter of the component mounting pin dictates the diameter of this hole, and the adhesion of copper to laminate determines the size of the pad.
For SMT, the pad size is a compromise between area used, solder joint strength and ease/accuracy of placement. Under-sized, the volume of the solder fillet will be affected and the joint strength reduced; sizes above the optimum will not markedly improve strength.
A via is similar in construction to a pad, but is used only to provide connection between different layers. Its size and packing density depend on the limitations of the drilling and patterning techniques, and vias are frequently much smaller than through holes.
When the interconnection requirement is complex, there is a balance of cost and performance between the use of extra layers and designing some areas with fine track and gap. The choice will depend on the capabilities of the manufacturer’s process, and any design constraints due to current/voltage restrictions. With high-frequency designs, the layout will be influenced by the need to minimise cross-talk and control impedance. With frequencies above about 300MHz, the distributed inductance and capacitance of the tracking mean that the board becomes a UHF/microwave component in its own right. This has implications for feature width, dielectric constant and board thickness, all of which demand tighter process control.
All boards are made using CAD, although the detail of the board layout is generally subject to modification to align with the requirements of the board production process. At higher frequency, where the board has become a component, simulation and design verification can now be attempted without the need to ‘cut metal’, although the programs for doing this are still at an early stage of development.
Test coupons are built into the board design, usually using spare material on the panel outside the circuit area, to allow the manufacturer to demonstrate that all processes have been completed with correct alignment, to test the electrical and mechanical characteristics of a standard pattern, and to verify the integrity of the assembly, especially of the plated through-holes.
The correctness of interconnection pattern itself is assessed by a ‘bare board’ test. This term can mean only a partial test confirming the tracks on each outer surface without a check on inner layers and through-board connections but, for boards containing many vias, a test which exercises the complete PCB structure is strongly recommended because of high cost penalty of assembling a faulty board. Most boards are electrically tested by the supplier for opens and shorts against the user’s design data.
Packaging has to protect both against ionic contamination and the ingress of moisture:
FR-4 multilayer boards can be held in stock for up to six months in the manufacturer’s original packaging. ‘Exotic’ materials should be dried and sealed in metallised bags with silica gel. This is particularly important for boards made of cyanate ester or with aramid reinforcement, especially those having large unbroken areas of copper.
Given that the reflow process takes the whole assembly above the glass transition temperature of the board, consideration has to be given as to how to support the board during this process to prevent warping.
You need to have a very firm grasp of typical process sequences, both for board fabrication and mixed technology assembly. If a copy is available, you could with advantage read the appropriate parts of Marks and Caterina Printed Circuit Assembly Design (3.5 for board fabrication; 3.6 for assembly). We also recommend that you do some browsing to establish the process flows used by typical manufacturers. As an example, you can take the plant tour at Proto Engineering, which will show you the kind of equipment used. If you are looking for a starting point, there are some nice long lists of fabricators and assemblers at Surfinbox.
We recommend that you don’t be seduced by the pretty pictures, but try and generate flow diagrams showing the sequence of process steps. Hopefully there will be many similarities between different companies! You should, however, expect there to be some differences, reflecting whether the company’s primary focus is in prototyping or in volume production. There will certainly be differences in the materials used.
Comparing our process flow sequences (Table 1 and Table 2) against real life should also illustrate that most companies introduce more QA stages than our simplified outline. This helps protect yield as well as ensuring satisfactory overall quality.
[ back to top ]
Track widths of 125µm are now achievable in volume, and multilayer boards can be constructed with upwards of 30 layers. However, there are costs associated with yield, and the finer the tracks and the more layers in a board the harder it becomes to test the operation of the device. There can be a balance between feature size and number of layers in favour of a more complex board but with fewer layers.
Be aware that the printed circuit board is a concept rather than a single solution. Alternatives from the PCB ‘kit-bag’ include:
15 years ago, larger Original Equipment Manufacturers (OEMs) tended to have in-house all the means of creating their product “from soup to nuts”, and would typically be able to fabricate a board, assemble it, and configure and test the completed equipment. However, smaller companies tended to subcontract the board fabrication, mostly because the processes used involved a knowledge of chemistry and needed facilities not easily implemented within a standard factory unit. In consequence, a number of small fabricators proliferated, providing a local service. Some however, catered for larger users, and had sophisticated volume manufacturing lines. In the early 1980s, there were over 600 board fabricators in the UK alone, ranging from multi-nationals to “Fred in a shed”.
The situation is now totally different, with drives for quality, enhanced technology, and economies of scale, putting out of business almost all but very proficient manufacturers of high quality products. In consequence, the number of UK operations is down into two figures, and recent years have seen much integration of activity.
At the same time as the fabrication industry has moved towards having a smaller number of bigger players, a contract manufacturing industry has grown up to support companies who no longer want to assemble boards. As with fabrication, there are continuing pressures on suppliers to become more professional. Also, as the manufacturing services sector comes under continued profit pressure both from customers and competitors in the Third World, the range of manufacturing services has widened. Companies that were once “board stuffers” now offer box build, and sell design and test engineering support as well as procurement and assembly.
The name of the game is ‘partnership’, with companies seeking to build up relationships with key fabricators and assemblers, rather than procuring solely on the basis of price and delivery. Fabricators and assemblers are expected to produce high quality as a ‘given’, and themselves will expect to work closely with designers, to mutual benefit. Building up relationships, and making sure that communication happens are important aspects of the professional’s work. This is becoming even more important as we move further into the 21st Century, because there is a growing tendency for volume manufacturing to migrate to areas of the world with a lower cost base. Your boards may be made and populated anywhere in the world, so your information has to survive travelling!
Whether you are a designer, a layout engineer, a process engineer, a procurement professional, or a manager, you need to be familiar with the range of processes carried out by or on behalf of your company by board fabricators, assemblers and box builders, and know how to be a ‘friend’ to your partners. Key requirements for manufacturing, whether at board or equipment level, are:
These are things which we ask you to bear in mind throughout this module.
[ back to top ]