Solved: Incomplete PSTN dialed numbers and echo

markgruber · ‎05-07-2009

I have had the following problem at a couple (not all) of my installations (SPA9000 + SPA400 + SPA962 + SPA942 -
AT&T Chicago area PSTN only):

Fairly often when the user dials the complete called number (e.g. 312 55-1212) not all of the digits make it to the
PSTN and we get a message back from AT&T about dialing the complete number or dialing a one first, etc.).

At first I thought it was a user problem or a dial plan problem but traces confirmed the complete number was making
it to the spa400.

I read somewhere online that it may have to do with an impedance mismatch causing many of the trailing digits to be
dropped out. The article suggested to boost the transmit power (from the default of 0). I did this going up to +4
and it seems to help a little but now there is lots of echo from our end (our callers here themselved talking)
although there is no echo from the remote end.

Any idea how to solve the partial dial problem and the echo problem. I have experimented with the SPA400 impedence
settings but there are so many combinations of rx and tx gain plus impedence settings.

Thanks for your help.

Alberto Montilla · ‎05-07-2009

Dear Sir;

The issue you mention seems to be related to the transmit gain as you say. The best good be if you can collect SPA400 logs to see what the exact issue might be. +4 might be to high, so my suggestion would be to start with +2 and see what happens.

Increasing the gain will definitively increase echo so don't change the receive gain.

Regards
Alberto

PS: How to get SPA400 logs is described here...

/docs/DOC-9864

View solution in original post

donwbpa3donwbpa3 · ‎05-07-2009

I have been unable to resolve the echo issues I have, but have seen a document on Cisco's site related to their other VOIP products. It describes different impedance mismatch scenarios and options. My question here to Cisco is: when the impedance setting is changes on the spa400 (or the gain settings), does the spa400 need a reboot or do those changes take effect immediately?

My suggestion on the mis-dials is to slow down the dialing (and possibly increase the wait time until the dial event starts after the line is taken off-hook). Not sure if this did what I think it did, but increasing DTMF power, increasing dial out wait, and changing the tone out on and tone out off timers to basically wait a few extra ms after the spa400 grabs a line before dialing, using louder DTMF tones, and using longer DTMF tones. It increases the amount of time for a call to dial/process....but it seems to have worked for me...Let me know if this helps or if I'm off base.

View solution in original post

Alberto Montilla · ‎05-07-2009

Dear Sir;

The issue you mention seems to be related to the transmit gain as you say. The best good be if you can collect SPA400 logs to see what the exact issue might be. +4 might be to high, so my suggestion would be to start with +2 and see what happens.

Increasing the gain will definitively increase echo so don't change the receive gain.

Regards
Alberto

PS: How to get SPA400 logs is described here...

/docs/DOC-9864

donwbpa3donwbpa3 · ‎05-07-2009

I have been unable to resolve the echo issues I have, but have seen a document on Cisco's site related to their other VOIP products. It describes different impedance mismatch scenarios and options. My question here to Cisco is: when the impedance setting is changes on the spa400 (or the gain settings), does the spa400 need a reboot or do those changes take effect immediately?

My suggestion on the mis-dials is to slow down the dialing (and possibly increase the wait time until the dial event starts after the line is taken off-hook). Not sure if this did what I think it did, but increasing DTMF power, increasing dial out wait, and changing the tone out on and tone out off timers to basically wait a few extra ms after the spa400 grabs a line before dialing, using louder DTMF tones, and using longer DTMF tones. It increases the amount of time for a call to dial/process....but it seems to have worked for me...Let me know if this helps or if I'm off base.

Alberto Montilla · ‎05-08-2009

Dear Sir;

Thanks for your comments, they are all valid, see below:

donwbpa3donwbpa3 wrote:

I have been unable to resolve the echo issues I have, but have seen a document on Cisco's site related to their other VOIP products. It describes different impedance mismatch scenarios and options. My question here to Cisco is: when the impedance setting is changes on the spa400 (or the gain settings), does the spa400 need a reboot or do those changes take effect immediately?

[Alberto] I suggest to reboot the unit.

My suggestion on the mis-dials is to slow down the dialing (and possibly increase the wait time until the dial event starts after the line is taken off-hook). Not sure if this did what I think it did, but increasing DTMF power, increasing dial out wait, and changing the tone out on and tone out off timers to basically wait a few extra ms after the spa400 grabs a line before dialing, using louder DTMF tones, and using longer DTMF tones. It increases the amount of time for a call to dial/process....but it seems to have worked for me...Let me know if this helps or if I'm off base.

[Alberto] You are right on spot! Gain is the typical cause for missed digits, but could be also a timing issue. So your suggestion is very valid. Unfortunately, there is no a single answer for this issue, as there may be different root causes (though most of the time it works without tuning).

markgruber · ‎05-08-2009

Thanks for all of your suggestions. I will collect the logs from the SPA400 and try some of the things mentioned.

Note that from my experience it does not seem a reboot is needed when changing the gain but I am not sure with the impedance values. I will reboot just to make sure.

I too have seen the SPA400 work perfectly with the default configuration (no faulty dialing) in some locations but not in others. All have been using local AT&T service. One interesting note is that in one location I was using a different manufacturer product (not the SPA400) and saw the same problem!

Mark