[Asterisk-Users] TE411P Really Bad Echo

[asterisk at oneringnetworks.com]

I thought that if the VPM was detected then you didn't have any control as to which algorithm was used.  

I was under the impression that the algorithms were only used for the software echo cancellation.  

At this point I'll give anything a try.

Stagg Shelton
www.oneringnetworks.com

-----Original Message-----

From:  Rob Lith <rob at connection-telecom.com>
Subj:  Re: [Asterisk-Users] TE411P Really Bad Echo
Date:  Sat Feb 11, 2006 10:17 am
Size:  4K
To:  Asterisk Users Mailing List - Non-Commercial Discussion<asterisk-users at lists.digium.com>

Stagg

I don't think it's a matter of trying echocancel on and off, it is a matter
of tuning your system to your local PSTN - this is a combination of trying
the different echocan alogrithims (i.e. MG2), the echotrainign etc and
setting your txgain - too loud outgoing audio will result in echo.
http://www.asteriskdocs.org/modules/tinycontent/content/docbook/current/docs-html/x1695.html#AEN1718

Rob

On 2/11/06, Stagg Shelton <asterisk at oneringnetworks.com> wrote:
>
> Yes, right now we are only using span 1 on the quad span card with plans
> to pull in another T1 PRI when we get this echo problem solved.  The echo is
> only experienced when the call terminates to traditional analog circuits
> both local and long distance.  Calls to cell phones, and other known digital
> circuits do not exhibit the symptoms.  I've been sent a few articles off
> list which discussed the reasons why this may occur, imperfect impedance
> where the digital circuit is switched to the analog circuit.
>
> In testing I've set echocancel=no made calls, and echocancel=yes and made
> calls with no real audible difference.  I haven't done a zap show channel
> while on a call with echo, but I plan to this weekend.  I was informed that
> the number of taps would be shown.  I used a stop watch last night to try to
> get the delay, and it was about 1/2 to 1 second delay and was continual so
> long as I was talking.  It wasn't affected by differing acoustical
> variations.  I tested this using handset, headset, and speakerphone.
> Disabling VPM and recompiling zaptel or removing the VPM off the board
> completely is the only thing that has any effect on the echo.  Hopefully the
> zap show channel will provide to me another data point to help me determine
> if the HW module is active.
>
> More to come...
>
> Stagg Shelton
> www.oneringnetworks.com
>
>
> Cory Andrews wrote:
>
> Stagg - I know you get a full 128ms tail of echo can on the Sangoma.  I
> believe that on the TE411P, the 128ms tail is shared by all (4) spans, and
> as you add additional spans up to maximum of 4, the echo can tail amount
> decreases accordingly.  If you are running 4 spans, you have 32ms of echo
> can tail on each span, not the full 128ms.
>
> Now that I'm reading back through your post thread, it looks like you were
> only running 1 span on the TE411P, so you should have been getting the full
> 128ms of echo can tail.  You may not find an improvement with the Sangoma.
> I have never experience, nor heard of a 1-2 full second delay.  Trial and
> error is likely your course of action.
>
> Cory J Andrews
> ++++++++++++
> VOIPSupply.com
> 454 Sonwil Drive
> Buffalo, NY 14225
> ++++++++++++++
> voice - 716.630.1555 X22
> email - Cory at VOIPSupply.com
> AIM - B2CORY
>
> ----- Original Message -----
> *From:* Stagg Shelton <asterisk at oneringnetworks.com>
> *To:* Asterisk Users Mailing List - Non-Commercial Discussion<asterisk-users at lists.digium.com>
> *Sent:* Friday, February 10, 2006 11:34 PM
> *Subject:* Re: [Asterisk-Users] TE411P Really Bad Echo
>
>  It was Digium's opinion that perhaps the card had a VPM.  We got a
> replacement TE411P, I implemented it tonight and still the exact same echo
> problem.  At this point I feel like I can rule out failed hardware.
>
> I contacted Digium support and now they are telling me it's something with
> my carrier, and I should call them.  I called Bellsouth, and they ran a full
> stress test on the circuit taking me offline for about 30 minutes.
>
> The end result is that the circuit test passed with no errors.  Bellsouth
> says it's not in their network, Digium says its not their card, and I have a
> te411p with VPM disabled in the wct4xx kernel module because something
> doesn't work the way it should.  My customer is wanting to know about
> sangoma cards with the echo cancellation, and at this point I'm nervous to
> recommend any hardware.  I'm going to look into the sangoma that you
> suggested.  Are there any other kinds of products that I could look into
> either Passive or Active.
>
> Thanks
>
> Stagg Shelton
> www.oneringnetworks.com
>
>
> Matt wrote:
>
> try sangoma carrier grade 104d hardware EC card. we're using it ourself.
>
> Best Regards
>
> Matt
> ----- Original Message -----
> From: "Anthony Rodgers" <Anthony_Rodgers at dnv.org> <Anthony_Rodgers at dnv.org>
> To: "Asterisk Users Mailing List - Non-Commercial Discussion"
> <asterisk-users at lists.digium.com> <asterisk-users at lists.digium.com>
> Sent: Tuesday, February 07, 2006 12:57 PM
> Subject: Re: [Asterisk-Users] TE411P Really Bad Echo
>
>
>    For what it's worth, we have been going through very similar issues
> with a
--- message truncated ---