cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Announcements

Get the latest Cisco news in this February issue of the Cisco Small Business Monthly Newsletter

949
Views
5
Helpful
7
Replies
Beginner

SPA232D connected to analogue phone answering machine

I have a customer with a SPA232D being used on a residential analogue phone with a built in answering machine.  Outgoing VOIP is working fine and incoming phone calls are working fine when the answering machine option is turned OFF. BUT while answer machine is turned ON, phone rings a few times then goes dead (the answer machine doesn't pick up at all). Anyone know of this issue and/or its solution?

Everyone's tags (2)
7 REPLIES 7
Highlighted
Advocate

It is almost impossible as

It is almost impossible as POTS (analogue) line have no way to reject incoming call. It can just pick up or ignore the bells.

So you need to be more specific. At the first, don't say 'few times'. Disclose the exact timing. How long the incoming call is ringing before it become lost ?

You are claiming that incoming calls are working with auto answer turned of. Did you mean that it will work even if you will wait before call pickup the same time as answering machine is waiting ?

 

What I mean ...

The incoming call will not ring forever. If the answering machine is waiting so long, the call may become lost.

Is such simple conclusion will not apply here, then full analysis will be required.

For the purpose of analysis, you need to turn on syslog&debug messages and catch them elsewhere. Those messages will help you (or us) to identify the call destiny ...

 

Highlighted
Beginner

Thanks for your response Dan.

Thanks for your response Dan. I have gathered some more detailed information from the customer as follows, however please let me know if there is any other specific info required to better resolve this case:

TEST 1 -  Incoming call WITHOUT the Cisco SPA232D and with answering function turned OFF - rings over 14 times (I hung up at that point, that is enough!)

TEST 2 - Incoming call WITHOUT the Cisco SPA232D but with answering function turned ON - rings 4 times, then answers

TEST 3 - Incoming call WITH the Cisco SPA232D turned ON and answering function turned OFF - rings 6 times at callers end, and 3 times on the phone in house **

TEST 4 - Incoming call WITH the Cisco SPA232D turned ON and answering function turned ON - rings 6 times at callers end, and 3 times on the phone in house **

TEST 5 - Incoming call WITH the Cisco SPA232D turned ON and a STANDARD plain phone only handset - rings 6 times at callers end, and 3 times on the phone in house **

**NOTE:  only when the Cisco SPA232D is turned ON and the phone rings via it, there is a delay in the telephone starting to ring in the house, and the rings are not in sync to Telstra rings that I hear when making the phone call from my mobile,  while with the SPA232D turned OFF, as in test 1 and 2, the phone rings in sync with the Telstra rings.  Also, even though the call goes to disconnected, the caller is still charged for a connected call by Telstra, as if the phone was answered.

Also, the problem of every first outgoing call made resulting in engaged signal, and second attempt gets through is still happening, so you have to dial every number twice to make a call, and we have tested this making the call to a mobile phone in the same room, so can confirm that it was not actually engaged when the first call is made.

Any update you can provide would be appreciated.

Highlighted
Advocate

TEST 1 & 2 reveal that

TEST 1 & 2 reveal that upstream switch is willing to ring (almost) forever and answering machine is set ti pick up call after 20 second of ringing. Answering machine is either not waiting for CID or expect it before first ring only or there is no CID sent by switch (at least in the form expected by answering machine).

TEST 3-5, number of rings is not the same as ringback tones: It may be related to Caller ID transmission. Answering machine may not ring while it's waiting for CID info. Voice->Regional->Miscelaneous->Caller ID * needs to be set accordingly. Also, Voice->PSTN needs to be configured according Telstra local loop requirements and your's wishes.

TEST 3-5, only 3 rings emitted: ring pattern is defined in Voice->Regional. It include maximum length of ringing. Ring patterns needs to be configured according your requirements. Also, Voice->PSTN define PSTN Timeout as well as other parameters. Be sure you have configured all parameters according your wishes and Telstra local loop requirements.

TEST 3-5, out of sync ringing: it's behavior by design. It doesn't harm.

 

According  ...

even though the call goes to disconnected, the caller is still charged for a connected call by Telstra, as if the phone was answered.

... you didn't specified who is disconnecting the call. At the first, some older phone systems doesn't allow the call become disconnected by called phone. Caller will be charged unless he will put headset on hook.

At the second, even in the case Telstra allow call disconnect from called side, we don't know the disconnect signal has not been issued by SPA232D or it has not been accepted by Telstra. SPA232D signal disconnection by putting PSTN wire on high impedance state. Open Voice -> Information page and monitor PSTN Hook state, Line Voltage and Loop current. It will disclose the signal sent by SPA232D to Telstra switch.So we can decide who's guilty then.

According ...

the problem of every first outgoing call made resulting in engaged signal, and second attempt gets through

... follow the same advice. It seems to be signaling problem between SPA232D and Telstra switch, so monitor line condition. It will help to analyze the issue. Also "PSTN Dial Delay" and other PSTN related settings may affect the issue as well ...

Sorry, I have no simple answer for you. It is because many PSTN settings needs to be set according local conditions and there are no universal defaults for them. I don't know the Telstra specifications. Also, line settings needs to be set according answering machine expectation and there are no universal defaults as well. And I know no particular answering machine requirements ...

 

Rate helpfull advices. It will help others to found solutions.

 

Highlighted
Beginner

Once again, thanks for your

Once again, thanks for your extensive reply Dan.

For your information and anyone else who may find it helpful, we managed to resolve the problem by changing the following setting:

Voice > PSTN > PSTN Timer Values > PSTN Answer Delay > change from '16' to '60'.

Seems to be working fine for the customer now.

Thanks and kind regards

Highlighted
Advocate

Thank you for mentioning the

Thank you for mentioning the final solution. Glad to hear it worked now. Rate useful advice/advices (if you consider there has been the one) to help others.

Highlighted
Beginner

Re: SPA232D connected to analogue phone answering machine

Hi guys!

I am getting a similar issue. Not on NBN yet, but the PSTN service fluctuates wildly when there is an incoming call.

Setup: PSTN RJ11 standard landline to SPA232D G7 (AU), RJ45 from Router to SPA232D WAN (w/ WAN Managment enabled). Set-up to allow simultaneous VOIP and PSTN landline services via one cordless Uniden triple handset system. PSTN calls ring the FXS phone handsets and VOIP calls ring the FXS phone handsets.

 

Most the time the phone rings and is answered. However, I have noticed the voltage drops from around -44 to -45 (v) on-hook to -4 (v) volts off-hook (on the phone). Calling mobile to myself I cannot reproduce it. However 3rd party calling in seems to randomly happen.

When the problem occurs - the phone handsets ring normally, you pickup and answer the phone, it will have an engaged tone! The other party/caller also receives the same engaged tone. So when issue occurs the call is terminated (I suspect power / volts issue) and not connected to the handset on the FXS port.

 

Occurs only on the PSTN (no caller ID) inbound calls and not the VOIP as no evidence of call under status.

I pickup the mobile, call the number and the phones all ring, answer and no issue!

The voltage drops with all calls, which is understandable, but how low should it go? What do other people have their's at?

 

The interface so **bleep** slow, so you cannot see any real-time stats! Instead keep opening new tabs with http://192.168.1.100/admin/voice/ to avoid re-authenticating (with your own I.P. obviously). Then auto-logged out anyway after 5-10 minutes...

 

Thank you for your help. :D

Cheers!

 

### IDLE STATE / ON HOOK [2 COLOUMS] ###
PSTN Line Status
Hook State:	On	Line Voltage:	-45 (V)
Loop Current:	0.0 (mA)	Registration State:	Not Registered
Last Registration At:		Next Registration In:	
Last Called VoIP Number:	0123456789@127.0.0.1:5060	Last Called PSTN Number:	
Last VoIP Caller:		Last PSTN Caller:	,
Last PSTN Disconnect Reason:	VoIP Call Ended	PSTN Activity Timer:	300 (ms)
Mapped SIP Port:		Call Type:	
				VoIP State:	Idle
PSTN State:	Idle

### INBOUND CALL (ON HOOK) [2 COLOUMS] ###
PSTN Line Status
Hook State:	On		Line Voltage:	-43 (V)
Loop Current:	0.0 (mA)	Registration State:	Not Registered
Last Registration At:		Next Registration In:	
Last Called VoIP Number:	0123456789@127.0.0.1:5060	Last Called PSTN Number:	
Last VoIP Caller:		Last PSTN Caller:	,
Last PSTN Disconnect Reason:	VoIP Call Ended	PSTN Activity Timer:	300 (ms)
Mapped SIP Port:		Call Type:	
				VoIP State:	Proceeding
PSTN State:	Ringing	VoIP Tone:	
PSTN Tone:			VoIP Peer Name:	0
PSTN Peer Name:			VoIP Peer Number:	0123456789@127.0.0.1:5060
PSTN Peer Number:		VoIP Call Encoder:	G711u
VoIP Call Decoder:	G711u	VoIP Call FAX:	No
VoIP Call Remote Hold:	No	VoIP Call Duration:	
VoIP Call Packets Sent:	305	VoIP Call Packets Recv:	0
VoIP Call Bytes Sent:	48800	VoIP Call Bytes Recv:	0
VoIP Call Decode Latency:	0 ms	VoIP Call Jitter:	0 ms
VoIP Call Round Trip Delay:	0 ms	VoIP Call Packets Lost:	0
VoIP Call Packet Error:	0	VoIP Call Mapped RTP Port:	16454 >> 0

### CALL PICKUP (OFF HOOK) [2 COLOUMS] ###
PSTN Line Status
Hook State:	Off		Line Voltage:	-6 (V)
Loop Current:	33.0 (mA)	Registration State:	Not Registered
Last Registration At:		Next Registration In:	
Last Called VoIP Number:	0123456789@127.0.0.1:5060	Last Called PSTN Number:	
Last VoIP Caller:		Last PSTN Caller:	,
Last PSTN Disconnect Reason:	VoIP Call Ended	PSTN Activity Timer:	300 (ms)
Mapped SIP Port:		Call Type:	PSTN To Line 1
				VoIP State:	Connected
PSTN State:	PSTN Caller Accepted	VoIP Tone:	None
PSTN Tone:	None			VoIP Peer Name:	0
PSTN Peer Name:				VoIP Peer Number:	0123456789@127.0.0.1:5060
PSTN Peer Number:			VoIP Call Encoder:	G711u
VoIP Call Decoder:	G711u	VoIP Call FAX:	No
VoIP Call Remote Hold:	No	VoIP Call Duration:	24:13:16
VoIP Call Packets Sent:	9073	VoIP Call Packets Recv:	8431
VoIP Call Bytes Sent:	1451680	VoIP Call Bytes Recv:	1348960
VoIP Call Decode Latency:	30 ms	VoIP Call Jitter:	8 ms
VoIP Call Round Trip Delay:	0 ms	VoIP Call Packets Lost:	0
VoIP Call Packet Error:	1	VoIP Call Mapped RTP Port:	16454 >> 0

 

Highlighted
Advocate

Re: SPA232D connected to analogue phone answering machine

 

I have noticed the voltage drops from around -44 to -45 (v) on-hook to -4 (v) volts off-hook

It's how the POTS line works. On-hook state is high impedance state (=almost disconnected phone) thus you see (almost) full voltage. Off-hook state is low impedance state, thus power drop. Connect an classic analog phone directly to PSTN line, you will observe the same behavior. Voltage drop is not symptom of an issue.

 

the phone handsets ring normally, you pickup and answer the phone, it will have an engaged tone! The other party/caller also receives the same engaged tone. So when issue occurs the call is terminated

OK. Lets allow me to rephrase.

  1. Your side is on-hook, e.g. high impedance=high voltage.
  2. Call is comming - impedance and voltage is still high, ringing signal (AC voltage superimposed on DC voltage) is supplied by PBX.
  3. SPA consider to pick-up - it's done by lowering the impedance; PBX removes ringing signal; ...

Now call shall be considered connected and you should observe low voltage on the line.

 

It seems the call is teared down instead. We need to identify who has ordered it.

 

Signal "lets tear down the call" is high impedance on answering side (e.g. high voltage on wire). So, unless you see voltage spike, the call has not been cancelled from your side.

 

Check and share results.

 

The voltage drops with all calls, which is understandable, but how low should it go?

There's nothing like "standard for POTS line". Nominal DC voltage, what impedance is high enough to be considered high or low to be considered low, exact AC voltage and wave form of ringing signal - all those parameters are country specific. And even phone company specific sometime.

 

But generally, high impedance can't be so high and low impedance can't be so low. Even if you will short the wires of PSTN line (thus you can't measure more than pure zero volts) it will be considered off-hook (e.g. connected in the case of for incoming call).

 

Occurs only on the PSTN (no caller ID) inbound calls

We need to decide who's tearing call (as discussed above). But Caller ID may be behind some issues. Thus I'm asking - is CallerID provider by your's Telco ? If yes, what kind of CallerID it use (there are so many kinds of CallerID protocol over POTS).

 

Just side note (urelated to the issue you described). The stats you disclosed shows:

VoIP Call Encoder:	G711u
VoIP Call Decoder:	G711u

Native codec for AU is G711a. While it doesn't harm to use other codec, it may be considered suboptimal.