Powering down

It's an all too familiar scenario. You're on the phone to an important customer and, far from being able to hear whether they're about to place the biggest order of the year, all you hear is an irritating crackling on the line. Worse still, your plea to repeat the last sentence for the third time just doesn't seem to be getting through.

If it's any comfort, you are not alone. Every year, thousands of users report that their critical business calls have suffered from intermittent hissing, buzzing, crackling and general interference. And when the phone companies investigate they find nothing untoward. It's a phenomenon that frustrates IT departments, telecoms engineers, sales managers, directors – in fact anyone that has to use the phone on a regular basis for business.

But the annoyance that many industries suffer as a result of poor quality voice communications is nothing compared to the potential loss of revenue that can arise if their data systems suffer the same fate. And this is exactly what all too many businesses are experiencing at this moment. The worst thing is that most organisations are unable to attribute their voice and data processing errors to anything other than the occasional glitch.

Make no mistake, every click and crackle on the phone, as well as every pc crash or data error, has a cause somewhere. I'm not suggesting that spikes and surges in the power supply are at the root of every problem, but there is little doubt that they contribute to a significant number of these aberrations. In fact, IBM's Systems Development Division comments that: "More than 80% of mains power problems are transient and noise related".

On top of the annoyance of crackling and other interference on voice calls, many organisations suffer from more sinister problems and damage.

Many of us have used modems that, inexplicably, have refused to work one morning. Usually we just shrug our shoulders and attribute their failure to wear and tear. Equally, some of us will have heard of phone systems that simply refuse to operate, even though they might have just been serviced. And there are countless stories of computer power supplies failing prematurely, network cards breaking down before their time and hard drives ceasing to work, with the loss of vital data.

All of these incidents have a cost implication. In the first instance there is the bottom line financial penalty associated with repairing or replacing damaged equipment. After all, many of these highly sensitive electrical components perform essential tasks that cannot easily be switched to other equipment. As such, business continuity will be severely compromised if the damaged equipment remains inoperable, even for a short duration.

As well as the physical cost of replacing damaged and broken components, there are wider costs associated with critical system down time. If an organisation's phone system is out of commission for even just a few minutes, who is to say which vitally important calls might have been missed during that time? Moreover, if the firm's data is damaged, erased or corrupted as the result of a power problem, there is the incalculable cost of trying to piece together what has been lost. Given that many organisations do not possess back-ups of their data, it is hardly surprising to hear that some companies never recover from such a disaster.

So, what's the real cause of these potentially business threatening problems and what remedies can we propose?

The first thing to remember is that there are many sources of electricity. Mains power, distributed via the national grid, is only one of the ways in which electricity enters our daily lives. Static electricity, generated by clothing for example, can wreak havoc with sensitive microchips and processors.

Even more destructive is the awesome power contained in a lightning strike. A typical strike can lead to a voltage surge of over 20 000 V being transmitted through the mains supply. This can cause instantaneous and catastrophic equipment failure, resulting in immediate operational shutdown and long-term disruption of business.

Research indicates that up to 80% of voltage surges come from internal sources, such as motors, fluorescent lights, photocopiers and other switching devices. This leaves the remaining 20% of transients arising from external phenomena such as lightning. This doesn't mean that one is more deleterious than the other. In fact, even though internally-generated transients will normally be of a lower peak voltage they will often cause cumulative damage, leading to premature failure over a period of time. External transients receive more publicity and interest, largely because they are more spectacular and, more often than not, the cause of immediate equipment failure.

The answer is to place a component before the critical component – be it comms or data hardware – that will successfully absorb and dissipate the transient energy from both internal and external sources. But don't think that this is a problem that only affects the mains input. There are plenty of documented instances of phones, modems, faxes and data servers being damaged by transient energy transmitted down the telephone line or communications cabling.

This is where a transient voltage surge suppressor (TVSS) comes into play. A TVSS is a passive device that limits the amount of energy arising from a transient surge, protecting electrical equipment from damage. Equally, a TVSS can also be used where certain equipment known to produce transients, such as photocopiers, needs to be isolated from more sensitive hardware.

One of the most damaging events for the highly sensitive, modern, communications equipment in offices and factories is a lightning strike actually hitting a telephone line. As when lightning surges through a mains supply, the peak voltage produced can be in excess of 20 000 V. Not surprisingly, the electronics contained in our telephones, faxes and modems cannot withstand this level of voltage spike, even for a few milliseconds. What we need, therefore, is to install a transient voltage surge suppression module upstream of the comms hardware, in exactly the same manner as we would with a mains TVSS unit.

The same principle applies equally to data comms and cabling installations. Imagine the likely consequences of several thousand volts coursing unchecked through the back of your pc, via the network connection, serial port or USB. Even those of us with a less than vivid imagination will be able to foresee the makings of an IT disaster. Fried motherboards, inaccessible hard drives and unusable memory would all be on the menu. And even though the replacement costs might just be containable, the longer-term consequences might not be surmountable.

Even if we disregard the effects of an average strength lightning strike on our data and comms hardware, there is still the more insidious effect of lower energy surges and spikes to consider. Because transient energy has many sources, highly susceptible voice and data communications cabling can be subject to wide ranging variations in voltage. In practice, this means that components that were designed to accept just a few millivolts can repeatedly see many times their rated voltage. Over a period of time this exposure to unwanted transients can lead to dramatically accelerated wear and tear, resulting in the design life of each critical component being significantly reduced.

What's the consequence of this effect? Rather like the metal fatigue that sets in when a component is stressed beyond its design parameters, the electronics will inevitably fail when least expected. The likely cost of such a failure? Again, if you value the data on your system and have some inkling of what a full IT infrastructure crash could do to your organisation, you'll be able to answer that yourself.

Market solutions
There are three basic types of surge suppressor on the market: the gas discharge tube (GDT), the metal oxide varistor (MOV) and the silicon avalanche diode (SAD). Each type has its own characteristics, although the way in which the individual components are assembled also affects the device's efficiency.

Tests have indicated that a combination of SADs and MOVs provides the best overall solution. As the SADs react very quickly, they prevent the clamping voltage of the device rising as more current is dissipated by the MOVs.

The ideal structure for a TVSS is a matrix configuration. This is a passive solution where the individual SADs and MOVs are arranged in separate assemblies, allowing individual components to self-sacrifice without compromising the overall performance of the TVSS. Because of this, the matrix solution offers all of the advantages of SADs and MOVs as individual components, but provides additional reliability and crucial performance where they are most needed. Even if one of the SADs or MOVs should fail, the remaining components will continue to protect sensitive equipment from possible damage. In fact, TVSS devices using matrix technology have an anticipated survival of 100 000 lightning strikes.

There can be little doubt that the occasional replacement of a passive component such as a voltage surge suppressor is a far better option than risking your organisation's entire business in the event of critical component failure.