[clug] Telephone line voltage

[Robert Edwards]

On 06/09/10 10:56, Mike Carden wrote:
> Curioser and curioser (said Alice).
>
> I'm inclined to agree that something peculiar is happening in
> Measurement Land. That a signal is present imposed on a DC offset is
> what's to be expected for ADSL, but the oddness of a meter seeing an
> 'AC' present in one polarity but not the other suggests that a meter
> is the wrong tool for the job.
>
> To really figure out what's going on, a decent CRO and an isolation
> transformer are needed. The tranny is essential because your CRO will
> happily try to ground the phone line via its power cord otherwise and
> that would be Bad.
>
> Aside from a smidgen of curiosity, I'd tend to side with Telstra here.
> If it works as it's meant to, there isn't really an issue.
>

I'd agree with Michael - if it ain't broke, don't fix it.

The following wasn't written by me, as suggesting attaching anything
to the phone line other that Austel approved equipment is a big no-no...
So DON'T DO IT! Really!

If you are keen to continue investigating, the CRO is a recommended
step - we need to know what frequency this 110VAC is at. The multimeter
may have a frequency reading, but there may be multiple frequencies at
different voltages present, and the multimeter can only tell you the
frequency of one "signal" and it may not be the same one that it tells 
you the amplitude of... A CRO will solve that (even better - a power
spectrum analyser...)

Not all CROs in the 21st century require isolation transformers (but
using an isolation transformer is a good safety policy in any case).

Failing using a CRO, another diagnostic step would be to hang a known
impedance across the line and get the multimeter to read the voltage
again. If, as most would suspect, the 110VAC is induced from the ADSL,
then there will be very little current behind it. A 22kohm resistor
across 110VAC would require 5mA to keep the voltage up. If you are
getting that much power (1/2W) at 110VAC, then I would say that you
have a problem. If a 22kohm resistor "snubs" the 110VAC (according to
the multimeter) then the 110VAC is stray noise and can probably be
safely ignored.

Remember that most multimeters have an input impedance of around
20Mohm, so only a very small amount of power would be required to be
induced to read such a high voltage (110/20,000,000 = 5.5uA, or 0.6mW -
quite reasonable for a long phone line running in parallel with lots
of other lines).

Cheers,

Bob Edwards.