# Why diameter of telephone cable is exactly 0.44mm and 0.63mm

I am wondering why telephone cables are having the diameter or exact 0.44mm and 0.63mm? Is that any special reason?

I have seen some of the cable have up to 2400 twisted pairs, could I know is the number of pairs will affect performance of cable?

Thank you.

• Proper telephone cables have a characteristic impedance of 600 ohm, read here: en.wikipedia.org/wiki/Characteristic_impedance what that means. Making the dimensions different would not result in 600 ohms and therefore an unsuitable cable for a telephone cable. Of course more twisted pairs affect performance but as long as the performance is good enough it is OK. – Bimpelrekkie Sep 21 '16 at 12:09
• Here is another one to read up on: en.wikipedia.org/wiki/Balanced_line – Voltage Spike Sep 21 '16 at 15:54

The diameters are almost certainly for historical reasons. The sizes you've given correspond to 1/40 inch and 1/60 inch. Probably these looked like good round numbers to someone who was making the decision a long time ago, and we've stuck with them ever since. Any new cable would have to have the same characteristic impedance as the old, fit into the same connectors, have the same or better bend radius etc.

Each twisted pair in a cable carries a signal. More pairs, more signals. More pairs can slightly reduce the performance of a cable due to crosstalk, but it's generally worth it to carry the extra signals.

The characteristic impedance of a telephone cable is very important to get right when it comes to minimizing side-tone (that's the handset microphone producing an amplified sound in your ear). This is achieved by a side-tone cancellation circuit: -

When the line impedance matches Rw in series with Cw there is no microphone signal transferred to the earpiece connected to winding A. The microphone feeds the centre tap between S and P windings.

So, for this to work effectively, the impedance of the telephone line has to "look like" Rw in series with Cw across the range of useful speech frequencies. Here's what a cable impedance might typically look like: -

As you should be able to see, the nominal impedance at 1 kHz is about 600 ohms in magnitude but is in fact similar to the complex impedance mentioned above.

The characteristic impedance of any cable at low frequencies is $\sqrt{\dfrac{R}{j\omega C}}$ where R is the series resistance of the loop per metre and C is the parallel capacitance per metre. Any change in conductor spacing or diameter means the capacitance per metre will change and the approximation of Rw in series with Cw won't be as ideal.

Capacitance for twisted pair for example: -

$C_{twistedpair}\left ( \frac{pF}{inch} \right )=\left ( \frac{.7065}{\ln \left ( \frac{2s}{d} \right )} \right )\cdot \epsilon_r$

• You state that sidetone should be minimised - from my memory of working in a telephone manufacturing campany that is not correct. The sidetone level is set to be the same as you would hear your own voice if you did not have the handset held to your ear. If there is no sidetone people tend to shout until the sidetone level is correct. – JohnM Apr 20 '18 at 11:01
• In the GPO phones there was a clever circuit that adjusted the sidetone depending on the line length. I can't recall exactly how it worked but it depended on the line current resulting from the 50V supply in the exchange divided by the combined resistance of the line and telephone. In addition to adjusting the sidetone this also increased the transmit level on long lines to compensate for the loss. Part of the circuit was 8 non-linear diodes in a pack. When using the phones on 24V PABX circuits we had to change the resistor values in the circuit to get the sidetone and transmit levels correct. – JohnM Apr 20 '18 at 11:01

In order to maintain a specific impedeance all properties of a telephone cable are specific. So the wire diameter,the wire distance to each other, the insulation thickness and type. The way of twisting and so on. This results from the past in the thickness indicated