Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!kitty!larry
From: larry@kitty.UUCP (Larry Lippman)
Newsgroups: sci.electronics
Subject: Re: I need a telephone ring detector chip
Message-ID: <1725@kitty.UUCP>
Date: Sun, 26-Apr-87 17:27:59 EDT
Article-I.D.: kitty.1725
Posted: Sun Apr 26 17:27:59 1987
Date-Received: Sun, 26-Apr-87 23:48:36 EDT
References: <228@phoenix.PRINCETON.EDU> <7940@utzoo.UUCP> <441@haddock.UUCP> <565@inuxh.UUCP>
Organization: Recognition Research Corp., Clarence, NY
Lines: 73
Summary: Telephone line loading and ALIT testing

In article <565@inuxh.UUCP>, rmrin@inuxh.UUCP (D Rickert) writes:
> > >You have two wires coming in from the phone line.  First, each goes through
> > >a 100k resistor.  Then there is a 10M resistor between them.
> > ...
> > One minor point: The 10 meg 'bleeder' resistor could cause problems.
> > Some telephone line integrity tests involve measuring minute leakage
> > currants. Leakages in that range (<100 meg) are usually indicative of
> > brittle insulation on the line pairs. This could cause some repair
> > folks to investigate the source.
> 
> Ten megs will not cause any problems.

	A few facts:

1.	# 22 AWG telephone cable insulated with polyethylene (PIC), with
	"perfect splices", has a conductance of about 0.05 mho/per mile. 
	A 2 mile loop would have a "theoretical" leakage resistance of
	about 10 megohms.

2.	# 22 AWG telephone cable insulated with pulp (i.e., paper), with
	"perfect splices", has a conductance of about 0.1 mho/per mile. 
	A 2 mile loop would have a "theoretical" leakage resistance of
	about 5 megohms.

	In reality, outside cable plant will never achieve such high leakage
resistance due to moisture present in cable splices and outside cable
terminals.  While pulp-insulated cable is almost always pressurized with
dry air or nitrogen, there is still enough moisture present to halve the
above value.  The same concept holds true for polyethylene cable.  A
gel-filled cable (usually called "icky-pic") does somewhat better, but
we are still in the same ballpark.
	Just _opening_ a pulp-insulated cable slice on a warm, humid day,
with consequent loss of pressurization can reduce leakage resistance to
well under 1 megohm.
	An actual cable fault, caused by presence of gross amounts of
water which wets a splice, can drive insulation resistance to << 100,000
ohms.
	So, the point it: most telephone companies set Automatic Line
Insulation Test (ALIT) apparatus to alarm at well under 100,000 ohms
leakage resistance, since: (1) there is such a large leakage variation
in subscriber loops; (2) no serious problem develops until leakage
resistance becomes << 100,000 ohms; and (3) to minimize false alarms.
Threshhold values used by one operating telephone company are 67,000
ohms leakage for "minor" alarm, and 43,000 ohms leakage for "major"
alarm.
	A "truly" wet splice may measure < 10,000 ohms leakage resistance,
so these ALIT threshholds are set to minimize false alarms.
	There are basically two types of ALIT tests: (1) leakage from
TIP to RING, and (2) leakage from either conductor-to-ground.  Most ALIT
apparatus can perform both types of tests.  Since there are many weird
telephones on the market which draw a small amount of current in an on-hook
state to power dialer memory, clocks, etc., most ALIT testing is done
from conductor-to-ground only to mimimize false alarms.
	
	So, to get back to the original issue, 200,000 ohms across TIP and
RING has always been considered a "safe" subscriber instrument leakage
value (historical reference: "old" HZM voice-connecting arrangement).
The two 100,000 ohm resistors suggested by the TI app note is an ideal
arrangement.
	While ALIT apparatus may use a higher threshhold for ground leakage,
we are not talking about ground leakage in a ring detector circuit (except
in some key telephone units, but that's another story...).

> I would suggest builiding a circuit
> such as this in a metal box to guard against component overheating in the
> event funny voltages get on the phone lines.

	I wouldn't worry about overheating.

<>  Larry Lippman @ Recognition Research Corp., Clarence, New York
<>  UUCP:  {allegra|ames|boulder|decvax|rocksanne|watmath}!sunybcs!kitty!larry
<>  VOICE: 716/688-1231        {hplabs|ihnp4|mtune|seismo|utzoo}!/
<>  FAX:   716/741-9635 {G1,G2,G3 modes}    "Have you hugged your cat today?"