Pilz’s Pipanel Touch is now used at Triumph to calculate shim sizes required, specific to each engine against default sizes.

Thunderbird, Daytona and Tiger… To any discerning motorcycle enthusiast, these names conjure up a unique image of highly desirable, technologically advanced motorcycles. In a word, Triumph.

With its sophisticated engine systems, combined with attention to detail and a commitment to quality, Triumph ranks as a top name in the British motorcycle industry, and worldwide.

Based in Hinckley, employing over 600 staff and producing 100 motorcycles per day, Triumph is one of the British success stories of the 90s. Production lines are automated wherever possible and are as advanced as any in the world.

One way in which production has been streamlined is in the shim-sizing process for camshafts in the engine. Previously the size of the default shims was calculated manually and shims were measured to fit particular engine models. Shim sizes often had to be re-calculated, the shims removed, re-sized, re-measured and refitted until a perfect fit was achieved. This had always been something of a laborious process taking two people as many as four cycles per shim.

The solution to the problem lay in a high-performance industrial computer which could be used for telemetry purposes and would be compatible with the software designed by Triumph. Specifications would also have to include a touchscreen which could be easily navigated when wearing gloves, and would offer a clear bright screen together with a housing resistant to the chemicals and minerals found in engine oil.

Pilz’s Pipanel Touch is now used to calculate shim sizes required, specific to each engine against default sizes. Default shims are fitted into the engine as standard at the start of the process. The operator measures any differential that may occur. Results are input into the computer and the correct size required to make up the differential is calculated automatically. This method has eliminated the need for the re-checking and re-calculation process which previously was performed manually. Pipanel Touch operates using in-house, DOS-based software. And since Triumph has installed the industrial PC, the program has been further amended.

• Pilz

June 2000

Put these guidelines into practice when using enclosures and cases, and you’ll be OK with OKW!

Cases have quite an influence on the electromagnetic compatibility of a product. The main reasons for EMC interference are:

• Conductive – within low frequency range below 100MHz
• Radiation from Cables – within medium frequency range (30-300MHz)
• Penetration through material and openings – high frequency range > 300MHz.

Hence, in order to reduce interference radiation, shielding of cases for ranges exceeding 100MHz is of vital importance. Shielding is defined as the ratio of field intensity in front of and behind the shielding wall. The effect of electromagnetic shields is based on a combination of several effects: if an electromagnetic wave hits metal, part of the wave is reflected. Another part of the wave penetrates the metal layer and is transformed into heat. The remaining part penetrates the metal layer and finally represents the source of interference.

As almost every case is provided with openings and cable connections which reduce the shielding effect, it is important to observe the following recommendations. The interference energy which is released through the opening largely depends on the following factors:

• Maximum linear size
• Frequency of the interference energy
• Polarisation of the wave
• Distance from radiation
• Field type.

Experience suggests that an opening should have an attenuation performance of at least 20dB. This means:

In order to achieve a shielding effect of approximately 40dB, OKW applies an aluminium layer which is exclusively attached to the inside of the case. Gaskets made of silver and silicone are also recommended, though these are unsuitable for humid conditions. Display windows should be provided with a conductive foil and should have a terminal lugs for connection to the case shield. To reduce the radiation between cable, plug and case, good design practice is to:

• Reduce the number of contacts
• Remove lacquer, colours and oxides
• Ensure direct and complete contact between cable shield and case
• Use shielded cables.

In order to maintain the protection class of the case and to improve the cable radiation, a suitable screw fitting for the cables is necessary. The use of external accessories may influence the protection class and can cause corrosion. For this reason, do consult the supplier before use.

• OKW

June 2000

A new light emitting technology based electroluminescence (EL) is a thin flexible light source that can be produced in sizes from about 600mm down to a few millimetres. Where should they be used?

Innovel lamps can be made from a single element for backlighting LCD displays for example. Light is emitted evenly over the whole area of each defined segment. Equally multi-segment lamps can be produced for eye-catching graphical displays or for selectively illuminating keypad buttons. Multi-segment lamps can be selectively sequenced on and off or even gently faded up or down. These techniques allow sophisticated eye-catching standalone displays. Low power consumption allows the use of batteries in many applications.

Innovel is constructed using a thin flat polymer substrate with a clear conductive coating. This is used as a base and the clear conductor allows the light generated in the phosphor layer to be seen. This sheet is usually 0.175mm thick. A very thin layer of phosphor is then applied followed by a very thin opaque conductor layer. The whole lamp is therefore very thin flexible and light.

When an electric field is applied across a thin layer of phosphor particles light is emitted. The phosphor layer is a good insulator so very little power is consumed. The electric field is generated by applying an AC voltage across the phosphor layer. Voltages between tens and hundreds of volts can be used at frequencies between tens of Hz to many kHz. Typically though somewhere between 50V and 120V is used at frequencies between 400 and 2000Hz.

Current flow is very small so driving voltages can be generated in a very cost-effective manner by small electronic circuits running from batteries.

Light output rises as voltage and frequency are increased . For a fixed driving voltage and frequency. For a fixed driving voltage and frequency – light output lowers with time – the rate of change will vary according to the level of driving. Acceptable variations will vary considerably according to particular applications. Some driving schemes use known ageing characteristics to alter driving voltages to compensate for loss of lamp brightness (up to certain limits).

As with other lamps such as incandescent and fluorescent light output lowers with use but unlike these other sources Innovel does not suddenly fail. Light output continues to fall gradually .

The initial brightness of the lamp can be within a wide range. However the higher this initial brightness the faster the drop in brightness will be. High temperature and high humidity are often used for accelerated life testing.

• RH Technical Industries
• 01264 363451

October 1998

The SD250 battery disconnection switch combines a 250A DC manual disconnect switch and a coil-operated contactor in a single unit. Alan Quinn takes a closer look.

Modern battery-powered electric vehicles are inherently very reliable and safe, but even when sophisticated electronic controllers are used, you want to have a means of disconnecting the battery in the event of an emergency, such as a vehicle failing to stop or an electrical short circuit.

In many countries, it is mandatory to fit one or more devices to achieve an emergency disconnection of the battery. The options are as follows:

• Manual disconnecting switches, such as the Albright SD and ED device.
• A minimum of two contactors with their contacts in series. This usually means a main line contactor and an additional contactor for each major electrical circuit.
• A battery plug, providing this is within easy reach of the vehicle driver.

Other practical considerations are the need to make the vehicle electrically “Dead” with a keyswitch when not in use and the ability to isolate the battery during routine maintenance.

A highly cost-effective means of achieving all the safety, legal and practical requirements is the Albright SD combined manual and electrical disconnecting switches. Alternatively, a manual disconnecting switch such as an Albright ED, together with a separate line contactor is worth considering.

The SD and ED ranges of disconnecting switches have been designed to provide a rapid means of disconnecting batteries or other power supplies in the event of serious electrical faults. Though primarily intended for use with battery-powered vehicles, they are also suitable for use with static power systems. All types are capable of safely rupturing full load battery currents in the event of an emergency.

SD switches combine the dual function of a manual disconnect and coil-operated line contactor. The benefits of this design include compact size and reduced installation costs, combined with an electrical capacity sufficient for most small and medium-sized electrical vehicles.

With the operating knob depressed (OFF), no electrical functions can take place, However, if the knob is ON, the option of energising the coil and thus closing the main contacts becomes available. The coil energisation can be carried out either through the vehicle keyswitch or as a result of a signal from the vehicle electronic controller.

When an SD switch is used as an emergency battery disconnecting switch, manually depressing the operating knob will override the energised coil such that the main contact and the auxiliary contact will open, until such time as the knob is again moved to the ON position.

Operating coils are normally continuously rated, although intermittently rated versions are also available.

An integral auxiliary contact is fitted. This has one normally open contact which closes when the knob is put into the ON position. The status of the auxiliary contact is not affected by the energisation or de-energisation of the coil. An additional double circuit normally open/normally closed changeover auxiliary contact can be fitted.

ED switches are manually operated devices with a simple over-centre spring mechanism which provides a snap action for both opening and closing the main contacts.

Simply pulling the operating knob upwards closes the main contacts and depressing the operating knob opens them again.

A double circuit, normally open, normally closed microswitch auxiliary can be fitted which has a DC resistive rating of 15A at 24V.

The auxiliary contact can be set to operate either before (default) or after the main contacts open, according to the circuit requirements.

• Albright International

January 2000

by Hans-Jurgen Wochian, Product Manager, Non-Portable Systems, Kontron Elektronik

While the development of software and electronics made enormous progress, the mechanics of the PC remained largely unchanged, even up to the present day. The interior of a PC still contains screws and metal plates. A fan blows air through a more or less empty case, with guiding plates influencing thermal currents.

This is where the new E-PAC (Electronic Packaging and Assembly Concept) internal design comes into play. In this system, all parts of the PC such as hard disks, power supplies etc. are no longer screwed into the case, but embedded in foam parts which hold them in place with positive locking. In order to fulfil industry requirements for robustness, temperature resistance and ease of servicing, Kontron Elektronik is the first company in the industrial computer market to produce a new 19in rack computer using E-PAC.

One effect of production using E-PAC is the increased lifespan of the equipment. This stems from the excellent cooling of components. Air flow within the case can now be controlled, because ventilation channels are incorporated in the foam, which directs cool air to the hot parts. In Kontron Elektronik’s IR KPR III rack-mounted computer, for example, about 30% of the ventilation airflow is channelled directly to the disk drives, which give off a great deal of heat when operating normally. The system not only provides cooling for heat-producing parts, it also protects heat-sensitive components.

The air flows through the foam from the hard disk to the CD ROM and floppy disk drives, protecting these components from overheating. And this is the point: the cooler the interior, the longer the life of all components. This allows the lifespan of the IR KPR III to be improved with higher MTBF (Mean Time Between Failures).

A weak point in classic construction methods is processor fans, which are not available on the market in an industry-compatible form, and therefore represent a particular risk factor. Kontron Elektronik has successfully done away with these fans, since even here, temperatures can be kept constant using directed air flows. E-PAC will allow today’s processors, and those of the future with even higher clock speeds, to be kept at non-critical temperatures without CPU fans.

A further significant consequence of production using E-PAC is the reduction in purchasing costs for customers. When high industry specifications are required, the mechanical effort involved in the old Oplate and screwsO method is very high, and correspondingly expensive. E-PAC allows the number and complexity of parts required for the construction of a computer to be reduced. The assembly time for the IR KPR III has been halved.

The EPP (expanded polypropylene) foam used for E-PAC brings great advantages in terms of the shock and vibration resistance of equipment. The sensitive parts of the computer are well protected, without increased expenditure. This property of the foam makes E-PAC design far superior to the conventional process.

A further aspect which reduces costs for users is the equipment development time. If E-PAC is used for internal design, a prototype can be assembled very quickly. The foam parts are cut from a plain block of foam and can quickly be made available for prototype tests and equipment certification. The tools for series production of the foam parts will then be produced only when all the foam parts meet the requirements fully. Instead of 12 months development time for a single case, the IR KPR III was ready for mass production after only 5 months.

In the long term, we can expect to see a changeover from the old mechanical principles used in all industrial computers, and even in office and home computers. Processor clock speeds are continually increasing, and heating problems are becoming ever greater. In addition to temperature problems, EMC problems also occur. Radiation will become more and more difficult to deal with. Here too the the fewer metal parts used, the less perturbing radiation occurs.

• Kontron Electronik
• Marilyn Comrie
• 01923 412345

September 1998

Ian McLeod of Papst, explains why doubling the number of fans doesn’t always produce the desired results

The idea of mounting two fans in series or parallel often arises when one fan isn’t up to the task of cooling and a larger one just won’t fit. The only time two fans of equal size can provide double the airflow, however, is when they are operating in free air, ie with no back pressure to restrict the airflow. Usually, engineers are working with an enclosure, and need to know how to optimise the performance of multiple fans.

Fans produce turbulent air currents which, when forced through equipment enclosures, remove heat from internal components. Obstructions to this airflow not only provide a reverse pressure, which the fan must overcome, but can also mask components from the cooling air stream. To design the layout of the box, the designer must consider the cooling paths. Densely packed enclosures exhibit airflow resistance, manifested as pressure loss in the direction of airflow. It’s analogous to an electrical generator forcing current through a resistor.

In theory, weighting factors can be applied to determine the flow/pressure characteristics of systems. In practice, the variety of designs used in enclosures and the presence of internal components that interfere with airflow, mean that it is impossible to calculate weighting factors using general formulae. Designers must rely on measurements or rough approximations. For practical purposes, the pressure loss ‘p’ of an enclosure is approximated by the formula:

p = Rv x Q/2 x V2

where Rv is a weighting factor for pressure loss, Q is the density of the displacement medium and V is the velocity of air flow through the system. It can be seen that pressure loss increases as the square of flow rate.

In the characteristic curve based on this formula where pressure loss is plotted as a function of flow rate, we see the air flow characteristics of a given enclosure. When fitted within an enclosure, the fan has to overcome the inherent airflow resistance. To achieve this, the fan needs to produce a pressure increase which will in turn decrease the flow rate. A characteristic fan curve expresses the relationship between flow rate and pressure.

The operating point of the fan is determined by the point at which the characteristic enclosure curve and characteristic fan curve intersect. At this point, the pressure loss of the enclosure is just compensated by the pressure increase of the fan and this point determines the flow rate available within that enclosure.

With parallel mounting, the flow rate is multiplied by the number of fans – but the results must be plotted over the entire characteristic fan curve. If fans are placed too close together, other interference effects come into play and reduce the overall flow. This is largely because the flow of air into a fan is usually laminar and smooth, while the exhausted air is much more turbulent. Even in an ideal environment, where interference effects could be ignored, a pressure increase of four times would be needed to produce a doubling of air flow, as pressure loss increases with the square of flow rate. Here, the airflow only increases by approximately 20 to 25% over that achieved with a single fan.

When fans are mounted in series, the pressure increase, in theory, is doubled. But if the fans are close together, results will again fall short of the theoretical performance due to the angular component of airflow introduced in the exhaust of the rear fan. This limits the suction effectiveness of the front fan. One solution is to direct the angular component back into the main air current using guide vanes, but this eats space. A better approach is to use one fan on the intake and one on the exhaust side of the enclosure. The presence of internal components and the large cross-sectional area between the individual fans will mean that airflow is essentially unidirectional. This provides effective airflow and relatively low noise levels.

The choice of series or parallel fan combinations depends on the individual application; some applications may require a combination of both. They key point is that two fans never mean twice the air flow.

• Papst

April 2000

With the amount of data being carried over the internet doubling every 100 days, what are the implications for networking enclosures, reflects APW’s MARCUS EDWARDS.

Appropriate cable management is the key to any high performance network. Ever-higher bandwidths are being achieved in copper cabling with the development of Categories 5, 5e, 6 and 7; fibre is being increasingly used in the horizontal as well as the vertical, with FTTD now a realistic, cost-effective option.

From its humble beginnings as a passive carrier with no effect on the network’s bandwidth, the structured cabling LAN has become an increasingly important element of the total system as bandwidths and frequencies have increased. The cabinets and enclosures – the areas of highest concentration of discontinuities within the network – have themselves become essential components because improper cable management will degrade the performance of a network using copper media. Performance degradation is significant in faster systems, such as fast Ethernet and ATM, where the bandwidth is 100Mb/s or higher. And with Gigabit Ethernet now a reality, even 100Mb/s is beginning to appear pedestrian.

The new IMnet and IMserv networking cabinets are the latest developments in a long line of rack and cabinets to be introduced by APW Electronics. As well as the demonstrable technical benefits, there are direct cost savings of 25% in the cost of the enclosure itself, further direct cost savings during installation and indirect savings to be gained through improved network reliability and decreased outages caused by physical layer faults.

The designs of the IMserv and IMnet top and bottom frames are based on a curved ‘X’ shape, giving good strength and rigidity and, most importantly, providing three places where cables can be dropped into the enclosure without having to thread them through apertures. Space for cable runs is maximised and, in bayed suites, cables can run vertically between cabinets, breaking out into the enclosure as required. The front verticals are recessed, enabling simple front patching between cabinets.

Segmented cable channel

All verticals accept both ETSI and 19in panel mountings; the 19in versions provide a 50mm wide segmented cable channel on one side of the 19in area in a 600mm wide unit. In an 800mm wide rack, there are 125mm wide cable management areas on both sides of the 19in rack mounting. Additional external 100mm cable housings can be mounted on one or both sides of the cabinet for heavy cabling density applications, and all accessories and cross members feature cable management facilities, rounded tops for cable guidance and clip into place without tools. In bayed suites, cables can be run between racks in front of the recessed verticals, and again, the inter-rack cables can be laid into the enclosures.

In both products, cables can be laid into the cabinet through three separate apertures in the top, doing away with the need to thread cables through frames. Cable runs can be positioned first, with the cabinet then being offered up and the breakouts into the cabinet connected, saving installation time.

Looking ahead, in the immediate future the differences between the needs of the patching/cabling enclosure and the server housing will continue to diverge. Already there are a number of 1U high, multi-processor servers on the market; the trend towards far greater distribution of computing power within the network is already evident. Instead of one central server, the trend is towards multiple workgroup servers and their associated hubs and routers, located at strategic points throughout the organisation. Already, application-specific servers are a reality: web servers, e-mail servers, CAD servers and so on will all co-exist and inter-operate across the network. The server housings will have to cope with physically larger equipment, so increased depths are required; higher packaging densities require more effective thermal management, and as most equipment has rear power and cable entry, the enclosure must provide access to the rear of the units for reconfiguration and upgrading. More equipment inevitably leads to more cabling, so cable management, even in server racks, becomes more significant. The server housing is becoming increasingly integrated into the LAN, with environmental and security monitoring already an established feature: this trend is bound to continue.

• APW Electronics

June 2000

Are you having problems with condensation? MARK JONES from OEM Automatic is your guide through the various types of condensation and how to deal with them

Careful thermal management of the internal environmental air has become an essential factor, because of increasing demands for extending the life of enclosed control systems.

One important area to consider is the elimination of condensed water vapour (dew) contained within the air molecules inside an enclosure. If unaddressed, moisture, when combined with aggressive gases or dust, can cause failures of components, such as pneumatic valves, relays, switchgear, PCBs and electromechanical assemblies.

• Dew problems manifest themselves in the following symptoms:
• Reduced electrical insulation at cable/wire joints
• Electrical creepage
• Flashovers
• Changes in contact resistance
• Corrosion of solder pads/components
• Valve gasket and seal deterioration.

There are two distinct forms of condensation: film condensation and dropwise condensation

In the former case, continuous condensate wets the surface (panel sides or component faces) and forms a continuous wet film across the entire surface, which then slides downwards under the influence of gravity. As the liquid film gets thicker, flow direction increases as mor vapour condenses on the surface. The liquid film provides a thermal barrier, so thermal energy cannot pass easily from the components/enclosure into the ambient air.

With dropwise condensation, the condensed vapour forms sporadic droplets on the surface instead of a continuous film. On inspection, many droplets of varying diameter are seen. The droplets roll down the vertical faces and clear the surface for vapour to escape, thereby aiding thermal emission by ten times that of film condensation. However, neither forms are conducive to good thermal management and should be viewed as a hazard.

As a rule of thumb, relative air humidity of 65% and upwards is considered to be problematic and therefore the risk of corrosion and other faults are more likely. Typically, if the air temperature in the cabinet is allowed to fall, the air molecules can no longer hold the water vapour in a gaseous state and hence dew is formed.

Solving the problem

By using an anti-condensation heater this problem can be avoided. Heating is initiated only when required – in daytime or peak time operation, it is usually reckoned that there is enough waste heat emitted from the processing circuits to maintain the required difference in temperature.

DBK’s range of ABC cabin heaters, which are available through the thermal division of UK and European business partner OEM Automatic, have been designed to meet market requirements of physical size, output power and method of fixing – base or DIN clip mount. This series of profiles is available in packages of 30, 45, 50, 60, 80 and 100W, all carrying UL, CSA and VDE approval.

The HP series is designed to be affixed to cabinet panels or thermally conductive surfaces. Their low profile and compact surface area makes them useful for small enclosures such as CCTV camera housings and telecoms forced convection air channels, used for pre-warming prior to PCB installation. For assemblies rated above 100W, OEM Automatic has introduced a new concept in forced convection heating called the Watt Box and Spider series.

• .OEM automatic

February 2000

What are forms of separation and why are they important? Paul Markham product manager for low-voltage devices at Merlin Gerin provides the answers.

National Annex NC to BS EN 60439-1 the new British Standard for switchboards is all about forms of separation.

But why separate the components making up a switchboard? Why not just mount the incoming and outgoing devices the busbars and the terminals in one big enclosure and shut the door? There is of course no fundamental reason why this can’t be done and indeed this type of all-in-one construction is sometimes adopted. It is perhaps worth noting that merely putting the equipment in an enclosure provides a basic degree of separation by preventing accidental contact with live parts at least while the enclosure door is shut!

In most applications however there are substantial benefits to be gained by providing additional separation between components. For example dividing the enclosure into separate compartments may allow fuses in one compartment to be replaced safely without the need to shut down the whole; and if the outgoing terminals are separated from other components it may be possible to connect new circuits while the board remains in service.

These are not the only reasons for providing separation in switchboards. Another feature is useful protection against the development of arc faults. Separation also protects against the passage of foreign objects from one part of the switchboard to another preventing for example a loose screw from finding its way into the magnet assembly of a contactor.

From these simple examples it might seem that the more separation the greater the benefits – but beware! More separation often means a larger and more expensive switchboard and it is quite likely that some of the additional “benefits” may not be useful in the intended application. Specifying maximum separation is not therefore a default or safe option. In every case the requirements of the application should be looked at carefully and the level of separation chosen to suit.

Clearly however if every switchboard manufacturer supplied products with slightly different levels of separation or even if they used the same types of separation but described them differently from their competitors switchboard purchasers would face severe difficulties in drawing up specifications and evaluating tenders. Fortunately BS EN 60439-1 provides the answer. The standard itself defines four basic forms of separation while the National Annex NC further splits these forms into a total of fourteen subdivisions.

Without explaining the subdivisions in detail they are largely concerned with separation between terminals and other elements and with the way that separation is achieved. They make a distinction for example between separation provided by metallic barriers such as the enclosure’s internal walls and that provided by insulating barriers.

But there are two important points which need to be made concerning forms of separation. The first is the popular belief that the higher forms of separation particularly Form 4 specify that the assembly is safe for live working. This dangerous belief is entirely incorrect. In fact the standard states categorically that no assembly which it covers is ever considered safe for live working!

The next point is that the forms of separation and more particularly the subdivisions defined in BS EN 60439-1 and National Annex NC are not to be confused with those relating to an earlier standard BS 5486 and the EIEMA guide relating to this standard. Unfortunately the designations are very similar but the detailed requirements differ. It pays to ensure that suppliers’ literature and documentation specifically relates to the current standard.

Switchboard users cannot fail to benefit from matching the forms of separation which they specify to the requirements of their applications. The uniformity imposed on manufacturers by BS EN 60439-1 makes this matching easier but specifiers still need to have a clear understanding of the standard and its implications. Those needing more detailed information are recommended to attend one of the seminars currently being presented by Merlin Gerin which provide a complete and straightforward explanation of BS EN 60439-1 and the National Annex. The seminars also cover type-testing another important switchboard-related topic about which there is considerable confusion and even misinformation. They are hosted by senior engineer Tom Mennell who is a member of the committee which drafted BS EN 60439-1. Full details of seminar dates and venues are available from Dan Varnam on (01952) 290536.

• Merlin Gerin
• Tel: 01952 290029
• Fax: 01952 290534

April 1996

Mark Knightley of CKS maintains that keyboards must retain their reliability in harsh environments even when being hosed down!

In recent years ruggedisation and environmental protection have enabled even the most sophisticated electronic equipment to be installed on the factory floor. Today computer-based systems are routinely near such ‘dirty’ processes as steel making and continuous casting as well as in the super-clean environments of food beverage and pharmaceutical plants where they face frequent hose-downs often with aggressive cleaning chemicals.

To achieve high levels of reliability under such conditions engineers have developed a battery of dustproof hose-down-proof and gas-tight enclosure designs. But not all the equipment can be mounted inside enclosures. This is particularly true of keyboards and other operator interface input devices which must be accessible to users and are therefore exposed to the environment.

It’s no secret that the keyboard can be an Achilles’ heel in dirty environments. Even in ‘soft’ office applications a spilled cup of coffee can spell disaster. On the shop floor an otherwise ruggedised industrial system can be rendered inoperable ? and the integrity and safety of the plant compromised ? all due to the use of an unsuitable keyboard.

To prevent problems of this kind industry is adopting an increasing variety of ruggedised sealed keyboards. Generally these keyboards are designed not only to survive in harsh industrial environments but also work with specialised software and so make it as easy as possible for workers to perform specific tasks. For example a given software programme may make it possible for a factory worker to perform a variety of tasks by only hitting a few function keys. To make it clear which keys initiate which tasks customised overlays are often placed on the keyboards.

Manufacturers of sealed keyboards typically offer a variety of enhancement products ? including mouse replacements (that is trackballs and pads). Pointing devices are virtually standard auxiliary components with many mouse-type pointers being integrated into the keyboard.

To ensure the quality and reliability of their products sealed keyboard manufacturers subject them to a variety of tests. The keyboards are dropped subjected to temperature extremes and undergo keystroke-failure testing. Typical operating temperature ratings are from 0 to +55?with keyswitch life expectancies available at greater than 20 million cycles.

Rugged technologies

Keyboards are essentially a collection of switches with circuitry to convert keystrokes into digital signals. Sealed keyboards add a number of innovative technologies to this basic design. There are various membrane-style keyboards built for industrial applications where slow to medium keyboard input is not an important factor. Many membrane keyboards use a substrate that has embedded circuits protected by a polycarbonate or polyester overlay installed in a watertight metal housing. Other varieties use full-travel membrane keyswitches.

A primary benefit of membrane keyboards is their flat profile. This is especially important in clean room applications such as food and pharmaceutical industries where sterilisation is critical. The flat profile of the keys doesn’t have crevices where bacteria can grow. These keyboards are also used in computer equipment designed for hazardous-rated areas.

One of the more interesting sealed designs uses piezoelectric elements in place of flexible membrane switch contacts. An activation force applied to the stainless steel graphic top panel causes piezoelectric elements to emit an active voltage. Although there is both visual feedback and audible feedback the rigid top panel precludes any tactile response.

Some manufacturers have resorted to such devices as conductive rubber pads and snap domes to overcome the membrane keyboard’s lack of tactile feedback (ie the ‘bounce’) that is so important to keyboard users. These however are susceptible to cracking tearing or collapse. Other manufacturers use stainless steel or moulded polyester domes as tactile elements.

In addition to their lack of tactile feedback most membrane keyboards are difficult to repair. Failure of a single switch will usually require replacement of the entire keyboard.

To overcome these weaknesses many manufacturers of sealed keyboards have concentrated on developing rugged reliable keyswitches with a polyester top sheet. This top sheet is then bonded to a rigid steel or aluminium plate. Square holes cut in the plate allow the embossed key caps to protrude and contact the bottom face of the polyester front sheet. This approach offers reliability and eliminates the tactile response problem.

• CKS
• Tel: 01707 322528
• Fax: 01707 372851

October 1996