One should be careful about the type of load presented to a
telephone line pair.  The equipment at the switch to which your phone is
connected may range in complexity from rotary steppers and mechanical relays
which are extremely forgiving of marginal conditions to solid-state switches
that do lots of line quality testing on an automated basis.  If you have
ever heard your telephone bell ding once or twice in the middle of the night,
that was an automated line test.  The system looks for there to be essentially
no current flow between the tip and ring wires and no connection between
either and other things such as Earth.  Those conditions usually mean damage
to the cabling somewhere between the telephone office and your house or
business.  If one of these tests detects leakage and they send a crew out
to check things out and it turns out to be a poorly designed line interface,
they may not be very friendly about it.

        In the little bit of research I did on this, I found out that the
DC voltage can vary quite a bit depending upon the design of the telephone
switching equipment.  While it is usually a nominal 48 volts, it can be
several volts lower or higher with no ill effects.  Modern switches are
supposed to send out .1 A of ringing current or enough to ring five standard
20-mil ringers.  This may dwindle down to something less if there is a long
run of wire, of course.  The ringers are usually isolated from the DC with
a suitable blocking capacitor.

        Telephones usually drop the line voltage to about 8 or 10 volts when
off-hook, but they should not present any DC load when on-hook.

        Party lines are a whole different story.  They use not only the
pair of wires to carry the conversation and dialing signals, but they also
use the path from tip to Earth and ring to Earth to separately ring each
phone or group of phones on the party line.  They have even designed schemes
for sending various frequencies between 20 and 50 HZ on multi-party lines
and each telephone must have a tuned reed that resonates with one of the
frequencies.  I hope this answers some of the phone interface questions
that ahve come up on the list.  Another excellent resource is
the comp.dcom.telecom news group and its archive site whose URL escapes me
at the moment.

Martin McCormick