asterisk users - Sep 2003 - False RING (incoming call) on Digium X101P FXO

I have a normal backup phone (and an alarm panel) sharing the POTS
line with the Digium X101P FXO:

     |
     |
Wall |>---+------X101P FXO as Zap/5
     |    |
     |    Phone & Alarm

Whenever the Phone is used, Asterisk sees a 'false ring' signal
immediately when the phone is hung up.

The Alarm panel dials out nightly at around 1AM, and each time it
completes the call, Asterisk sees another False Ring.

As you can imagine, this is rather annoying, since the bogus ring
causes Asterisk to kick off the incoming context (pots-line) and ring
the inside extensions.  It looks like this:

ERROR[294930]: File callerid.c, Line 192 (callerid_feed): fsk_serie made mylen
< 0 (-12)
WARNING[294930]: File chan_zap.c, Line 4317 (ss_thread): CallerID feed failed:
Success
WARNING[294930]: File chan_zap.c, Line 4334 (ss_thread): CallerID returned with
error on channel 'Zap/5-1'
  == Spawn extension (pots-line, s, 1) exited non-zero on 'Zap/5-1'

Clearly there is no CallerID because there was no actual ring signal.

Why is the FXO seeing a local hangup as a Ring?  I thought Ring was
90v AC.. not something easily mistaken.  The hangup is just one pulse.

I have tried both kewlstart and loopstart with the same results.  This
is a standard U.S. (Southwest Bell) line.

What else can I try?  Is there any way to 'tighten up' the ring
detection?

-Steve

Try increasing "PEGCOUNT" in wcfxo.c.  You'll have to do a
"make install"
in zaptel, unload, and reload wcfxo:

# cd /usr/src/zaptel
# vi wcfxo.c (change the #define PEGCOUNT 5 line to something like 10)
# make install
# rmmod wcfxo
# modprobe wcfxo
# asterisk -vvvgc

See if that does it.  Thanks!

Mark

On Sat, 20 Sep 2003, Steve Haehnichen wrote:
> I have a normal backup phone (and an alarm panel) sharing the POTS
> line with the Digium X101P FXO:
>
>      |
>      |
> Wall |>---+------X101P FXO as Zap/5
>      |    |
>      |    Phone & Alarm
>
> Whenever the Phone is used, Asterisk sees a 'false ring' signal
> immediately when the phone is hung up.
>
> The Alarm panel dials out nightly at around 1AM, and each time it
> completes the call, Asterisk sees another False Ring.
>
> As you can imagine, this is rather annoying, since the bogus ring
> causes Asterisk to kick off the incoming context (pots-line) and ring
> the inside extensions.  It looks like this:
>
> ERROR[294930]: File callerid.c, Line 192 (callerid_feed): fsk_serie made
mylen < 0 (-12)
> WARNING[294930]: File chan_zap.c, Line 4317 (ss_thread): CallerID feed
failed: Success
> WARNING[294930]: File chan_zap.c, Line 4334 (ss_thread): CallerID returned
with error on channel 'Zap/5-1'
>   == Spawn extension (pots-line, s, 1) exited non-zero on 'Zap/5-1'
>
> Clearly there is no CallerID because there was no actual ring signal.
>
> Why is the FXO seeing a local hangup as a Ring?  I thought Ring was
> 90v AC.. not something easily mistaken.  The hangup is just one pulse.
>
> I have tried both kewlstart and loopstart with the same results.  This
> is a standard U.S. (Southwest Bell) line.
>
> What else can I try?  Is there any way to 'tighten up' the ring
detection?
>
> -Steve
>
> _______________________________________________
> Asterisk-Users mailing list
> Asterisk-Users@lists.digium.com
> http://lists.digium.com/mailman/listinfo/asterisk-users
>

On Saturday, September 20, 2003 10:37 PM, Mark Spencer 
[SMTP:markster@digium.com] wrote:
> > I was surprised to see that it's 240 volts (peak-to-peak)!  Egad..
> > no
> > wonder it shocks fingertips.
> >
> > 20Hz (50ms cycle), 2 second long clean sine waveform.  I was just
> > surprised to see twice as much voltage as expected.
>
> 240 sounds like a lot.  Are you sure you were doing DC measurement?
> Normal is +/- 90 - 120, generally at least 48VRMS or about 70V.
The local phone company will, while the phone is on hook, provide a 48 
 to 52  dc voltage at the central office. To ring a phone a 50 v ac 
signal is placed on top of this ~50 v dc.

Don Pobanz
>
> Mark
>