The course team are very grateful to Harwin plc for creating this material especially for one of our modules, and for their kind permission to reproduce the figures it contains. We would encourage you to visit the company web site at http://www.harwin.com/. Associated web sites cover Component Assembly Systems and Application Specific Connectors, and there is a wealth of information on connector products of all kinds.
Designers of electronic equipment (OEMs) use connectors for the following reasons:
There are two basic forms of PCB connector – those that connect board to board and those that connect board to cable. Connectors consist of male and female (plug and socket) types.
Board-to-board connectors can plug together perpendicular to each other, or in-line, or can be stacked. The main, or base board is called a mother board and the board which plugs into it is called a daughter board. Any subsequent add-on boards are called child or baby-boards. Very large mother board PCBs having multiple daughter boards are also called backplanes.
Board-to-board connectors can be single part (called edge connectors) or two part. Two part are generally more reliable than edge connectors, but are more expensive. Board mounted connectors can have through hole or surface mount terminations.
Board-to-cable connectors can be mounted at 90° or parallel to the board surface. They can again be single part where the connector is soldered permanently to the PCB, or more commonly, two part. These usually consist of a male header connector, which is soldered to the PCB. The female cable connector can then be plugged and unplugged as and when required. Board-to-cable connectors can also be Input/Output (I/O) connectors, such as ‘D’ connectors found on the back of personal computers (Figure 1).
Another type of PCB connector is the IC socket (Figure 2), into which integrated circuits can be plugged and unplugged. These allow for ease of repair or upgrade, without the need for any desoldering operations.
[ back to top ]
When choosing connectors for a PCB design, there are several important characteristics that should be considered. To a large extent, these characteristics will depend on the type of circuit, that is analogue, digital, power, RF, etc.
There are some requirements that apply to the connectors on all types of circuit, such as:
When designing PCB’s for digital or high-speed digital circuits, the choice and performance of the connector becomes critical. The additional characteristics, which are of most importance, are:
Contact impedance – This must be known and consistent in order to allow the impedance of the PCB tracks to be matched to that of the connector.
Skew – Need to know the total signal path length through the contact.
VSWR – Especially important to know in high-speed circuit design.
Grounding/shielding – Grounding between each track and each connector contact maintains the correct impedance matching and eliminates crosstalk. Other ways to shield circuits are by using ferrites in the signal path (to eliminate edge ringing), or by using metal shrouds or cans to continue the shielding across from PCB connectors on to cable connectors.
[ back to top ]
There are three basic types of applications in connectors, which use different types of polymer.
RF/high frequency connectors – using polyethylene, polypropylene and PTFE.
Standard board-to-board and board-to-cable – using nylon, polyester, polycarbonate and polyacetal.
Micro-miniature, surface mount and special applications – using liquid crystal polymer, polyphenylene sulphide, diallyl phthalate and high temperature nylon.
Plastic materials have lightweight, are easily moulded into complex shapes, have good resistance to corrosion and excellent dielectric properties.
Most connectors (except RF) have a percentage of glass fibres mixed in with the basic polymer. This can vary between 15% and 40% glass content. The more glass, the stronger the moulding and the better it withstands heat.
In connectors, it is mandatory to use, and declare, an Underwriters Laboratories (UL) approved material, which will self-extinguish. That is to say that if the material catches fire, the moment the flame is taken away, the moulding will extinguish itself virtually immediately. This is achieved by including flame retardant additives in the moulding powder.
The norm for connectors is the UL94V-0 UL rating. This relates to instant self-extinguishing. Many terminal blocks and other electrical components are made from nylon having a UL94V-2 rating, which will take longer to self-extinguish. Underwriters issue a ‘Yellow Card’ for each material that they have approved.
In normal wave soldering processes, the PCB protects the connector moulding from the solder wave: normal materials like nylon, polyester, etc, can therefore be used.
In surface mount soldering processes, the connector moulding has to withstand the full soldering temperature, which requires high temperature materials such as Liquid Crystal Polymer, Nylon 46, etc. Normal nylon/polyester materials would melt under these temperatures.
Electrical contacts in connectors are usually made from a non-ferrous material so as to give good electrical conductivity (that is low resistance, which allows electrical current to flow more easily).
Male contacts are usually of a solid, non-flexing design and use brass or phosphor bronze material.
Female contacts are usually designed to flex and therefore have to withstand stresses. These normally require stronger materials such as phosphor bronze or beryllium copper (the strongest available). The normal rule is that the smaller the contact, the higher the stress, the greater the need to use beryllium copper material.
An electro-plated deposit normally of gold or tin covers most contacts in connectors. Without this covering, the base material would oxidise and corrode, and therefore give high contact resistances.
The best material for contact resistance is gold. This is a ‘noble’ metal and remains clean under severe environments. However, it is expensive and can cause problems of solderability. Gold plated contacts are therefore often selectively gold/tin plated, having tin on the areas that are to be soldered.
Lower cost, lower performance connectors have tin-plating all over.
On brass components it is essential to have an absolute minimum of 1 micron of nickel as an undercoat. This is to form a barrier to prevent zinc from migrating through into the tin plating. Zinc will cause de-wetting.
The standard minimum thickness for tin plating is 3 microns. Plating adhesion is important because if the plating lifts-off, de-wetting will occur. It is therefore very important to thoroughly clean the components before plating, and to ensure that the nickel undercoat does not become ‘passive’ during the plating process.
Tin plating (actually 98% tin, 2% lead) is now the preferred finish for most connector applications. This gives better wear properties than 90/10 or 60/40 tin lead, and solderability is acceptable, except in surface mount applications. Pure tin can grow crystals at low temperatures (-40°), 10% of lead eliminates this phenomenon. Environmental issues means that it is desirable to remove as much lead as possible from the process which also explains the move towards 98/2 tin/lead solutions. Some tin solutions also contain fluoroborates and formaldehyde and these are being phased out.
In order to be cost competitive, most connectors are assembled using automatic machinery. The fastest method of contact assembly is called gang insertion, where an entire row of contacts is assembled into the connector housing in one operation. As many as 100 contacts can be gang inserted in 4 seconds.
Another automatic assembly method is called terminal stitching, which, as the name implies, fits one contact at a time into the housing, at high speed (Figure 3). This technique is particularly useful when contact patterns change within a housing. In some instances, a secondary operation is needed after insertion of the contacts into the housing. This can be some form of locking in of the contacts or bending the terminations through 90°, etc. On gang insertion machines, this is normally carried out on the same machine, followed by automatic testing and inspection, and even packing.
[ back to top ]
There are many ways of attaching electrical cables to connector contacts:
Electrical cables can have their ends stripped of insulation, and then the internal conductor can be soldered into the back of the connector contact. This is normally called a solder cup. In most applications, this is a one at a time operation. However, in sub-miniature connectors, all cables can sometimes be soldered to the contacts simultaneously by use of a hot bar soldering technique. After soldering, it is usual to fit an insulating rubber sleeve over the soldered joint. This sleeve can be a heat-shrink sleeve, which is fitted over the cable and shrinks tightly around it when heat is applied.
Often called ‘hoods’: can be fitted to many cable connectors and this has the advantage of tidying up all the cables and clamping them as a strain relief (Figure 4). In this way, if someone pulls on the cables, it does not strain the soldered joints. Some of these hoods are made of metal, or metallised plastic. This is for users who need EMI shielding. Cables and connectors emit stray electrical fields, which can affect nearby equipment, or be affected by other incoming electrical fields. Covering the cable/connector with a metal case collects these stray fields. Legislation on EMI shielding is now quite severe.
Crimp connections offer a more reliable cable joint than soldering, which is difficult to inspect. Once a crimping tool is set, it reproduces a consistent joint, which can be easily measured and inspected.
There are two forms of crimp connection, circular and F crimp. Circular crimps are used with high reliability precision turned circular contacts. F crimps are used with stamped and formed flat strip contacts. In both instances, the electrical cable has to have its insulation stripped off at the cable end. In circular crimps, the crimping tool has four indenters positioned at 90° to each other. The crimp barrel on the contact is placed in a positioner, which is fitted to the crimping tool. Operation of the tool brings in the four indenters which crush the crimp barrel onto the electrical conductor and welds it in a gas tight joint. (Gas tight means that after crimping; the electrical joint will not deteriorate in a severe industrial environment). The completed contact/cable joint is then removed from the tool and can be fitted into the appropriate position in the cable connector.
These types are called crimp, poke-home contacts. They are always poked-home from the rear face of the connector, and are pushed fully in until they latch into position. The latching feature varies from connector to connector. Also, these poke-home contacts can be removed from the connector should replacement be necessary. These are called replaceable contacts and the de-latching can be front releasable, side releasable or rear releasable (less usual). A small special tool or miniature screwdriver is the normal way to release contacts.
In the case of F crimps, contacts are normally supplied for volume use on reels, where the contacts are joined together by a metal bandolier. Crimping machines, typically called applicators, feed the reel of contacts into position between a crimping die set. The stripped cable is inserted into the crimp section of the contact and the top and bottom die is closed to again wrap the contact around the conductor in a cold, gas tight weld. Usually, another section of the contact is crimped around the cable insulation, to act as a strain relief. The F crimp operation is much quicker and lower cost than circular crimping, and is therefore much more widely used. It can be fully automated as a production process.
Another popular cable connection system uses insulation displacement connectors (IDC), sometimes called IDT (insulation displacement terminations). These use ribbon cables, which have normal circular conductors bonded into a flat cable insulation. The cables are positioned at the same pitch (centre to centre distance between conductors and contacts) as the connector contacts. The contacts are shaped like a tuning fork, with a precise gap. The cable is positioned above the contacts and forced down through the contacts, which pierce (displace) the insulation and make permanent electrical contact between the cable and the connector contacts. This can happen at all positions simultaneously (mass termination). The connection system is very easy to use, requiring only simple bench presses with location nests.
There is another type of ribbon cable called flat flexible cable (FFC). This has flat, rather than circular conductors inside the insulation. Connection to this cable is very difficult and very few connector manufacturers have products suitable for its use, except with cable pre-stripped of insulation on one side.
Wire-wrap connectors are mainly used on large backplanes, and are therefore a relatively limited market. To make a wire-wrap joint, approximately 25 mm of insulation is stripped off a 30 AWG cable (AWG = American Wire Gauge) and relates to 0.3 mm diameter. The uninsulated cable, and a short length of insulated cable are positioned into a wire-wrap tool (these can be electrically, pneumatically or hand operated). The tool is then used to tightly wrap the cable around a square post (usually of 0.635 mm size). During wrapping, the cable is stretched tightly around each of the four corners of the post, over a length corresponding to at least 7 complete turns, plus 1½ turns insulated. This results in 4 x 7 = 28 gas tight joints per complete wire-wrap. This method is therefore very reliable, but uses space behind the connector. One other advantage is that wire wrapping can be automated using a CNC (computer controlled) machine.
[ back to top ]
Coding keys: Normally made of plastic, and which fit into or around the mating faces of connectors, to prevent the wrong connectors being plugged together, or to the wrong PCB. (Similar to polarising keys).
Earthing tabs or strips: Metal components, which surround contacts or connectors and which take any short circuit or stray electrical fields to earth on the PC board.
Hold-downs: Can be integral in a connector housing, or additional parts, which hold a connector firmly down onto the surface of a pc board. This prevents lifting of the connector during the soldering operation, and acts as a strain relief on the soldered joints during service.
Hoods (covers): Used on cable connectors to insulate and protect the cables as they leave the back of the connector housing. Will usually include some form of clamp to prevent any movement of the cables from affecting the termination joints. (that is crimp, solder, etc.)
Jackscrews: Devices which, when rotated by hand or hand tool, plug and unplug board to cable connectors in a controlled way. When fully tightened, they prevent the cable connector from ‘walking out’ of the board connector under vibration conditions.
Latches: A simple way to hold board to cable connectors together to prevent them ‘walking’ apart during service, usually comprising of a metal or plastic spring which latches over some feature moulded into the connector housings as they are plugged together. Deflection of the latches by hand or with a small hand tool, allows the connectors to be unplugged.
Pick-and-place-pads: Simple plastic or metal ‘throw-away’ components which are initially fitted to surface mount PCB connectors for the sole purpose of allowing them to be picked from some form of packaging, and placed onto a PCB automatically (Figure 5). After the subsequent soldering operation, the pick-and-place pads are discarded.
Polarising features: Can be features moulded into a connector housing, or additional add-on components, which prevent connectors being plugged together 180 degrees from their intended orientation. (See also Polarising keys and Coding keys).
Polarising keys: Add-on components fitted to connectors to prevent two or more similar connectors being plugged together into the wrong mating half.
Shrouds: Plastic or metal shields which fit over connectors and protrude out from the mating face such that any exposed contacts are protected from mechanical damage. Can also incorporate a pre-alignment feature to assist in guiding connector together during plug-in.
Tape and reel packaging: Continuous reels of pre-formed plastic tape containing pockets into which connectors or components are housed (Figure 6). These reels are used on automatic pick and place assembly machines, which feed the tape under a tool head. This head picks the connector out of the pocket and places it on the surface of the PC board. Tape and reel is the most popular form of packaging for surface mount connectors and components.
[ back to top ]
Often electronic designer engineers find that a ‘standard’ connector is not exactly what they need for a particular application. When that happens, connector suppliers who have in-house design and manufacturing capabilities are able to offer a customised connector.
Typical ‘custom’ connectors (Figure 7) can be:
For the more simple changes from standard product, a relatively small tooling charge or minimum sales order is usually applied. For new custom designs requiring production tooling, tool charges are payable by the customer. The connector is then produced exclusively for that customer.
[ back to top ]
IC sockets (Figure 8) are useful when the designer wishes to allow easy upgrade, reprogramming or repair, without the need to desolder, and are available for many of the larger standard package formats, in particular DIPs and PGAs.
For parts with through-hole leads, such as DIPs, t here are two distinct types of socket on the market – turned pin or stamped and formed contacts.
The turned pin type is for use in severe environments, or where utmost life and performance are required. Each contact is manufactured in two parts, an outer precision-turned brass shell (usually tin plated) and a four-finger beryllium copper inner contact (usually gold plated). This gives genuine selective plating, and no chance of solder wicking up into the contact area. The four-point contact is obviously very reliable (four chances not to fail) and the socket can be re-used many times.
Stamped and formed contacts can have single or double-sided contacts, and are used where lowest cost is required, and probably only two plug-ins will ever take place.
The vast majority of these sockets are either the precision-turned type (as in the DIL sockets) or have zero-insertion force stamped and formed contacts. Pin Grid Array packages may have up to 320 contacts, which means that precision-turned contacts must have low forces. For this reason, PGA socket contacts usually have six-finger inner contacts (Figure 9).
This need for low insertion force also fuels a growing trend towards ZIF-PGA sockets. In their normal state, the contacts in a ZIF-PGA socket are wide open, allowing the PGA package to be placed inside without resistance. The contacts can then be operated by a lever, which closes them onto each leg of the package. Reversing the lever opens the contacts, allowing the package to be easily removed and replaced. This is particularly important in computer applications, where upgrades are common.
Sockets for PLCCs normally have beryllium-copper or perhaps phosphor-bronze spring contacts, positioned vertically in rows around each side of a square. The PLCC package is plugged into the socket from the top side and goes just ‘over-centre’ on the contacts, which helps to retain the package. Packages can be removed for replacement by levering them out, using a screwdriver, in a slot provided for the purpose. Legs from these sockets (sometimes called pin-outs) can be in-line, staggered or in surface mount format. Socket mouldings can be high or low profile (height above the PCB), and surface mount types will use high temperature plastic. Similar sockets are available for the smaller SOJs.
Individual PCB sockets are of similar construction to turned-pin IC sockets. Some have four-finger and some have six-finger internal contacts, and they will accept mating pin diameters of nominally 0.5 mm, 0.8 mm, 1.0 mm and 2.0 mm. The outer shells on some types have a knurl that acts as an interference fit in the PCB hole, and are referred to as ‘self-retaining’. Other types have the top aperture closed by a thick film of silicone rubber, which is called an RTV seal (RTV = Room Temperature Vulcanising). In production, the silicone rubber is dispensed onto the socket aperture, where it quickly forms a skin and then fully hardens, at room temperature.
Surface mount PCB sockets are basically square in section, and sit down onto pads on the board for reflow soldering. Some sizes of PCB sockets can be supplied to customers fitted onto metal combs, or in plastic (Mylar polyester) film tape. This enables them to be fitted in multiples, and so reduce assembly time and cost.
[ back to top ]
Not to be forgotten within the connector family are a number of specialised pin-based products:
Power pins: Usually spade or circular format, used to take power onto boards. Various standard sizes and current/power ratings can be fitted into housings or directly into a PCB, and this operation can be automated.
Fuse holders: Stamped and formed from flat strip, and solderable into PCBs, these accommodate standard circular fuses.
Connector pins: Often used as an alternative to pin headers, especially when non-standard pin configurations, or longer/shorter pins are specified. Pin insertion into PCBs can be easily automated.
Test points: Designed-in to some PCBs in order to connect to Automatic Test Equipment (ATE). This equipment contains spring probes, which make electrical contact with the test pins.
[ back to top ]
Coaxial connectors are for use with Radio Frequency (RF) signal circuits, usually operating in the frequency range between 3 kHz and 300 GHz. This covers the range of radio and television transmissions. The connectors are constructed to have a centre contact, surrounded by an insulator (normally PTFE for its low dielectric constant), contained in a metal body. The centre conductor of the coaxial cable is terminated to the centre contact. The outer copper braid of the cable is terminated to the connector body, and the insulation remains in place. Most coaxial connectors are required to have an impedance of 50 ohms.
Fibre optic connectors are specialist connectors that allow light transmission via fibre optic cables. Connectors can transmit light from one cable to another, from one cable to a pair of cables, or to an electro-optic component. These connectors are used primarily in the telecommunications industry, or where light or laser systems are required.
Audio connectors are often referred to as ‘jacks’ or ‘jack plugs and sockets’. These connectors work in the frequency range up to 20 kHz, and can be found on all types of recording and playing equipment.
[ back to top ]
The following parameters are important for design engineers to be able to assess whether a particular connector will meet their needs.
Contact resistance is the electrical resistance measured from the back of the male connector through to the back of the female connector (measured in milliohms, 1mΩ = 0.001Ω ). This is sometimes called the total path resistance. The actual contact resistance is the resistance of the joint between the male and female contacts. This is also the only variable resistance. Normal maximum value for contact resistance variation, from initial to after use, is 5mΩ .
Dielectric withstand (or proof) voltage
Dielectric withstand voltage, is a derated voltage, below breakdown voltage. It is specified as the minimum acceptable test voltage at which flashover must not occur, when the voltage is applied for one minute.
Refers to the range of climatic conditions and mechanical stress conditions, which the connector has been proved to be able to withstand. Ratings are normally shown as, for example, 55/105/21. This refers to -55°C, +105°C, and 21 days under standard damp heat conditions (+40°C, 95%RH)
Insertion and withdrawal force
Is the force measured in newtons required to plug or unplug a male and female connector (or pair of contacts) together. Connectors (or contacts) can be high force, low force or zero insertion force. Zero insertion force means that the connectors can be joined together without restriction, and subsequently the electrical contacts brought together by some integral mechanical mechanism.
Insulation resistance is the electrical resistance recorded between adjacent contacts (measured in M W ). This measurement checks the design and quality of the mouldings.
Number of operations
Sometimes called life or durability. Refers to the number of times a male and female connector can be plugged together and still meet electrical and mechanical specifications.
Voltage breakdown is a measurement of the minimum voltage at which flashover occurs due to the voltage jumping across from one contact to the next. This measurement checks the creepage distance across the moulded surfaces and the air gap between contacts.
Working voltage is the maximum voltage at which the connector can be used. Normally this figure is one third of the withstanding voltage.
Typical connectors will be specified to withstand the following tests/conditions:
High temperature: usually between +70°C and +125°C.
Low temperature: usually between -10°C and -65°C.
Damp heat: usually +40°C; 95% relative humidity.
Temperature cycling: usually 10 cycles of maximum to minimum temperature.
Vibration: usually 10g maximum.
Mechanical shock: usually 100g.
Mechanical bump: usually 40g.
Mechanical acceleration: usually 25g.
Industrial atmosphere: gas test.
[ back to top ]
In your researches into connectors, you will meet references to a number of common standards:
ISO 9000: An internationally accepted quality standard. It is recognised in virtually all markets across the world. It specifies how a business should control its processes in order to ensure its customers a quality product or service. Third Party Accreditation is used to monitor the adherence of businesses to the ISO 9000 standard.
UL: Underwriters Laboratories approvals are often required, especially when product safety, and use by the general public are involved. The approvals cover topics such as flammability of plastics, electrical safety, etc.
CE: European standard ensuring safety on all types of electrical and electronic equipment. CE marking is applied to show that approval has been granted.
FCC: American standard to control electrical interference from and to equipment. (EMI).
CSA: Canadian standard, similar to FCC/CE.
TUV: German standard for products, mainly invoking safety requirements.
[ back to top ]