Protecting Your Telephone Lines

By Elektor Magazine

Historically, telephones were so simple that almost nothing could go amiss from an electrical point of view, but those days have changed. In the past, phone companies installed surge protection on all telephone lines exposed to storm risks, but now that we are connecting up so many complicated electronics to lines, the protection has disappeared.

Telephone While we have lost count today of all the equipment that has been destroyed by a 'bolt of lightning', surprisingly one you only have to invest a few dollars to get a remarkably efficient protection device. If lightning strikes near a telephone pole, the line carries transient voltages--up to several thousands of volts.

To protect your lines from destructive pulses, traditional components are not powerful or fast enough. The illustration shows a (gas-filled) spark gap with three electrodes, each insulated in an airtight cylinder filled with rare gas. As long as the voltage present between the electrodes is below a certain threshold, the spark gap remains perfectly passive and presents an impedance of several hundreds of MΩ. On the other hand, when the voltage rises above this threshold, the gas is very rapidly ionized and the spark-gap suddenly becomes a full conductor. This conductor then can absorb colossal currents without being destroyed.


This illustration has the same magnitude as an ordinary one watt resistor. It can absorb a standardized 5,000 amps pulse lasting 8-20 ms! Since we are utilizing a three-electrode spark gap, the voltage between the two wires of the line or between any wire and ground, cannot exceed the sparking voltage, which is about 250 volts here. While this might offer sufficient protection, we prefer to add a second security device made with a VDR (GeMOV or SiOV depending on the manufacturer). This element limits the voltage between line wires to a maximum of 250 volts.

Since pulses generated by lightning are very brief, the ground connection of our assembly must be as low inductance as possible. It must therefore be short, and composed of heavy-duty wire (1.5 mm2 c.s.a.). If not, the coil, composed of the ground connection, blocks the high frequency signal that constitutes the pulse and reduces the assembly's effectiveness to nothing.

Finally, the beauty of this solution is that it has no effect on the low frequency signals used by telephones, fax machines and routers.


I think you do your (intelligent) readers a disservice by publishing potentially dangerous advice from Elektor on this very important subject.

There is no mention made of the 10-ohm resistors and their special characteristics. They are intended to act as quick fuses in case something like a traffic accident brings high-voltage (commonly 4,800 VAC) power lines in contact with telephone lines. Their circuit neither mentions that nor has any shunt element to ground on the user side to prevent entrance of this voltage to the user (after the gas-discharge-tube is destroyed).

Perhaps most important, it encourages DIY folks to replace the Telco-installed unit that is required by Code to be in place (commonly called the NIU or network interface unit). Stuff like this, written by folks who clearly don't have any comprehensive knowledge of the subject, is nothing short of hazardous to reader's health.

Bill Whitlock
President & Chief Engineer
Jensen Transformers, Inc.