I have an incoming POTS line connection to an FXO port on a 1760 gateway. This POTS line is only providing -24 voltage rather than the normal -48 voltage. This seems to be preventing the FXO interface on the router from sending DTMF that is acceptable to the POTS line. I have read that I need to change the FXO line in use voltage on the gateway to a lower value in order to allow DTMF to pass correctly to the POTS line. Does anyone know how to change the FXO line in use voltage, or know if it is possible to convert the -24 to -48 between the FXO port and POTS line?
A few things:
1) I've never seen line voltage affect DTMF specifically. This sounds like a red herring.
2) The line voltage is platform/card specific. You won't be able to change your line voltage 24 volts with a CLI.
It is possible that the line voltage may not be affecting the DTMF. Incoming calls work on this POTS line, maybe that indicates the voltage is irrelevant in this case? When outbound calls are attempt to this FXO port the port goes off hook and dial tone is transferred properly. This issue is that I am unable to break the dial tone once the FXO port has gone off hook. Is there a way to inject DTMF from the gateway directly?
No, there is no way. When dialing out, the card wants loop current (indicated as "dialtone"), in lack of it the call fails. Perhaps Nick knows of some hidden command to skip that step.
What is entering the dtmf?
You can try adding this to your dial peer:
This adds 3 seconds of delay and then DTMF 12345 into the stream.
Nick, the problem with your suggestion above, is that when the port does not detect loop current, the call is aborted before any prefix or dialed digit is sent.
I would like to add that incoming calls to the DID on this POTS line route through the FXO port just fine. Also if I connect a regular analog phone directly to the POTS line DTMF is accepted and both incoming and outgoing calls route without an issue. What I don't understand is how DTMF from the FXO is different than the analog phone? If they are different how can I more closely replicate DTMF from the FXO to match that of an analog phone?
If the router wasn't detecting dial tone the user wouldn't be hearing dial tone as well. As well "dial tone is transferred properly" seems to sounds like the current may not be the issue.
In the case the router doesn't detect dial tone it would hunt to a different port.
It sounds like there may be a DTMF negotiation failure, rather than a current problem. This is why I suggested having the dial peer manually enter DTMF.
Nick, to my knowledge the router never detects dial-tone, although the vpm debug print a string to that effect.
What the router does for outgoing calls, is detecting current loop after going off-hook. So for example when the FXO port is not connected, you are able to tell, again by looking at the output of debug vpm signal.
Instead, for incoming calls, router detects AC ring current and goes off-hook without further check.
So if my theory here is that the attached equipment is not providing correctline voltage or loop current, if I'm correct the proof would be identical output of "debug vpm singal", either with the line connected, or disconnected.
I'm not sure if it's the loop current, dial tone, or both that is actually detected to allow an outgoing call. I haven't seen a case where only of the two is valid/invalid. vpm signal surely does seem to indicate that it has detected dial tone.
But, if he's hearing the dial tone on the phone, neither are the issue.
For incoming calls the ABCD bits are toggled in a way to signal an incoming ringing call.