From: Floyd Davidson <floyd@ptialaska.net>
Newsgroups: comp.dcom.modems
Subject: Re: Excessive Line Current
Date: 22 Jun 2000 09:46:02 -0800

Bob Starnes <rastarnesNOSPAM@hotmail.com> wrote:
>> lklk wrote:
>>
>> My husband just got a new computer at work and he can't get the modem to
>> work. I know more about computers and hardware than he does so I am just
>> going to see if I can get some help here.
>>
>> Like I said he just got a new computer. His old one had a modem and worked
>> fine. His new one has a modem but when he tries to connect to his isp he
>> gets the following error:
>>
>> Excessive Line current. Please unplug the phone line, otherwise your modem
>> may be damaged.
>>
>> I understand what this is saying, I am just a little confused about why the
>> current was fine for the other modem. Do the jumpers just need to be moved
>> around on this new modem? I am just guessing.
>
>
> Perhaps the other modem didn't have the capability of reporting the excessive
>line current, might be worthwhile to have your telco check your phone line.

Because this is describes ad "at work", it is likely the line is
from a PBX and not directly from the telco.  If that is true,
then whoever maintains the PBX should be called instead of the
phone company (it might be maintained by the phone company
though).

Another alternative is to just measure the line current.  Find a
reasonable Volt-Ohm-Meter (it doesn't have to be a particularly
good one) and set it to a 500 milliamp or greater scale, and put
it directly across the phone line.  The main thing to note first
is polarity, does the needle jump up, or down.  Switch the leads
if it is wrong.  Then perhaps also switch to a more sensitive
scale to get a reading in roughly the middle of the scale if
possble.  (With a digital meter, ignore polarity and set it to
read milliamps.)

The maximum loop current should be less than 120 ma.  The
minimum should be greater than 23 milliamps.  It is exceedingly
unlikely that one would see more than 60 ma and in fact the
modem might find that "excessive", even if the telco specs say
it is acceptable.

Typical figures range from 25 ma to 45 ma.

If the modem is complaining about less than 60 ma, replace the
modem with a properly designed one.  If the current is in excess
of 60 ma it is probably possible to re-configure the line card
in the PBX to lower the current.  I do not know of any telco
line switching systems that provide more than 60 ma these days,
so if it comes from the telco and is that high it might be an
indication of an equipment problem.  It is also possible that it
is normal, and they will refuse to change anything.

  Floyd

--
Floyd L. Davidson                          floyd@barrow.com
Ukpeagvik (Barrow, Alaska)

From: Floyd Davidson <floyd@ptialaska.net>
Newsgroups: comp.dcom.telecom.tech
Subject: Re: POTS line max current draw
Date: 13 Feb 2001 20:49:09 -0900

"Don Russell" <drussel2@san.rr.com> wrote:
>What is the max current draw allowed on a POTS line:
>    A - before the telco sees that as "off hook"
>    B - if "off hook", still within FCC rules
>
>Thanks,

At 20 mA or greater, it is offhook.  At 18 mA or lower, it is
onhook.

The switch should supply a minimum of 23 mA.

The absolute maximum allowed is 120 mA.  The max recommended is
60 mA.

Typical values range from 23 mA to 55 mA, with 28 mA to 45 mA
being the most common.

Perhaps the most notable thing about the above is that any
resistance in series with a loop should be capable of
dissipating the heat generated when 120 mA of current flows (and
the assumption should be made that enough voltage can exist to
cause that much current).  That can be a significant amount of
heat!

--
Floyd L. Davidson         <http://www.ptialaska.net/~floyd>
Ukpeagvik (Barrow, Alaska)                 floyd@barrow.com

From: Floyd Davidson <floyd@ptialaska.net>
Newsgroups: comp.dcom.telecom.tech
Subject: Re: POTS line max current draw
Date: 14 Feb 2001 14:25:33 -0900

Bill Krause <krause@azte_CUT-TO-REPLY_k-eng.com> wrote:
>Floyd Davidson wrote:
>> >What is the max current draw allowed on a POTS line:
>> >    A - before the telco sees that as "off hook"
>> >    B - if "off hook", still within FCC rules
>>
>> At 20 mA or greater, it is offhook.  At 18 mA or lower, it is
>> onhook.
>
>I must disagree with the statement "at 18mA or lower, it is on-hook".

Well, I sure didn't describe that number quite correctly.  Once
the line is offhook, it must stay offhook with a current as low
as 18 mA.  It may revert to onhook at a value below 18 mA.

What I can't find right now is the spec for the minimum current
that it can stay in the offhook status.  I think it is 10 mA,
but I'm not sure.

>In my experience (I've designed a few line circuits), the line
>circuit is designed with the loop closure (i.e. off-hook)
>detection circuit threshold set somewhere between 8 and 15
>milliAmps (mA).  The default for many SLICs is near 11mA.  The
>requirement for the CO-side DLC equipment, per TR-57, is found
>in section 5.3.7, (R-13) "a resistance of >= 10,000 ohms
>applied between the tip and ring conductors of the RT line unit
>shall be recognized as loop open.". At a -48VDC on-hook feed
>voltage, this implies that current draw of less than 4.8mA must
>be seen as on-hook.  Greater current draw than this may be
>detected as a loop-closure (off-hook).

It does not imply that any current greater than 4.8mA causes
offhook.  Also note that using -48VDC to calculate anything is
guaranteed to cause invalid results.  A "48V" battery plant
provides from 42 to 56 volts under normally experienced
operating conditions.  And loop current may also be supplied
with a voltage as low as 24 volts and as high as 105 volts.

The 10,000 Ohms is a "leak test" value, not the definition of
normal loop current detection.  In essense, whatever the
conditions of the loop are, adding a 10,000 Ohm resistance in
shunt should never cause an offhook condition.

>Also, per FCC part 68.312 (b) (1) (i) which governs the
>telephone-side on-hook impedance limitations, "The dc
>resistance between tip and ring conductors, and between each of
>the tip and ring conductors and earth ground SHALL be greater
>than 5 MegaOhms for all dc voltages up to and including 100
>volts.  " This says that the maximum off-hook current draw for
>a telephone on a line with a -48VDC feed would be 0.0096 mA.

It says no such thing.  Give that one more thought.

>Also per part 68.312 (b) (1) (iii) to (iv), during ringing the
>dc current draw is limited to 3mA , and the AC impedance must
>be between 1400 ohms and 40,000 ohms. So during ringing, the
>phone could draw a minimal amount of current safely.

Yes.

>I suspect the original question is really "how much current can
>I steal from the phone company to power my phone in the on-hook
>state without causing problems?".  I would suggest that if you
>need to pass FCC regulations, the answer is virtually none.

I'm not positive, but doesn't Radio Shack sell a number of FCC
certified devices which do exactly that?  Hence the answer could
not be virtually none.

> If you are just doing a home project and not selling
>equipment, then I would say you could get away with drawing 1
>or 2 mA before anyone or anything would notice or care.
>
>However, the service provider could ask you to remove the
>equipment from the line if they detect it during Metallic Line
>Testing (which happens every night on some switches and will
>test the line impedance).

Do they test for that?

If there is too much current draw, the line goes offhook and
eventually times out and goes to some form of treatment and
eventually to a lockout state.  No alarms, and no maintenance
actions, are likely to result.  The telco will not be aware that
it is a problem unless the customer calls in a trouble report.

>  For a home project, it would be best to just provide local
>power to save the trouble of configuring a complicated line
>powered supply.  There are times when the current available on
>the loop is zero (e.g. during Loop current feed open -
>i.e. forward disconnect and during portions of Line Testing),
>so you will not have any power available and the circuit will
>need some holdover capability.
>
>The rest of the numbers provided by Floyd are typically correct.

--
Floyd L. Davidson         <http://www.ptialaska.net/~floyd>
Ukpeagvik (Barrow, Alaska)                 floyd@barrow.com

From: Floyd Davidson <floyd@ptialaska.net>
Newsgroups: comp.dcom.telecom.tech
Subject: Re: POTS line max current draw
Date: 15 Feb 2001 09:13:06 -0900

rbmccammon@mmm.com (Roy McCammon) wrote:
>Floyd Davidson wrote:
>
>> >Also, per FCC part 68.312 (b) (1) (i) which governs the
>> >telephone-side on-hook impedance limitations, "The dc
>> >resistance between tip and ring conductors, and between each of
>> >the tip and ring conductors and earth ground SHALL be greater
>> >than 5 MegaOhms for all dc voltages up to and including 100
>> >volts.  " This says that the maximum off-hook current draw for
>> >a telephone on a line with a -48VDC feed would be 0.0096 mA.
>>
>> It says no such thing.  Give that one more thought.
>
>I suspect that he meant 'on-hook current draw'

I don't pay much attention to typos, and had read that for what
he meant it to say.  But it _doesn't_ say that the line will
have a maximum onhook current of 0.0096 mA.  It specifies three
leakage paths within a telephone set in terms of resistance.

First, the 0.0096 mA would be current in only one of the three
leakage paths specified.  Second the voltage will almost never
be 48 vdc, but might be 75 or even 100 vdc (and will sometimes
be 56 vdc for a typical 48 volt battery plant).

And of course there are other leakage paths not involving the
telephone set  (I have not looked up FCC part 68 to see if that
is actually only specifying the telephone set as indicated above,
or not.)

--
Floyd L. Davidson         <http://www.ptialaska.net/~floyd>
Ukpeagvik (Barrow, Alaska)                 floyd@barrow.com

From: Floyd Davidson <floyd@ptialaska.net>
Newsgroups: comp.dcom.telecom.tech
Subject: Re: POTS line max current draw
Date: 15 Feb 2001 09:22:47 -0900

Lizard Blizzard <nospam@rsccd.org> wrote:
>In the electronics book 'The Art of Electronics' 2nd ed, there is a
>chart from Bell that shows the allowed parameters for on hook and off
>hook conditions.  I don't remember what the page is and my book is at
>home.  But in any case, the co-author Winfield Hill posts to
>sci.electronics.misc and the others on a regular basis.

I haven't seen that particular chart, so I won't comment
specifically to it.  But the problem with generalizations about
loop current standards is the combining the several standards.
There is one set of specifications for what the telephone set
must be, another for what the cable pairs must be, another for
what the line card is allowed to generate and another yet for
what the line card is required to accept.

For example, the line card should never produce more than 120mA
of loop current, yet there is a very nice chart available which
shows how a switch must react to loop conditions, and it goes
all the way up to 177 mA.  Some people have trotted out that
chart as proof that a switch is allowed to produce loop current
up to 177 mA, which is not what the chart means.

>Don Russell wrote:
>>
>> What is the max current draw allowed on a POTS line:
>>     A - before the telco sees that as "off hook"
>>     B - if "off hook", still within FCC rules
>>
>> Thanks,
>>
>> --
>> Don Russell

--
Floyd L. Davidson         <http://www.ptialaska.net/~floyd>
Ukpeagvik (Barrow, Alaska)                 floyd@barrow.com