[Asterisk-Users] Will Echo problems EVER be solved, I'm scared

[Steve Underwood]

Matt Fredrickson wrote:

>On Fri, Aug 26, 2005 at 02:00:54PM -0600, Rich Adamson wrote:
>  
>
>>Relative to the fxotune app, it would appear the app is specific
>>to the v2.4 kernels (/dev/zap*), which the v2.6 kernels don't use
>>    
>>
>
>It should with 2.4 and 2.6.  2.6 kernels with properly configured udev
>rules should create the /dev/zap/* entries dynamically.
>
>  
>
>>(but rather the udev equivalent). (When I had * running on a v2.4
>>kernel, the output from fxotune never deviated from all zero's. So
>>I'm assuming the default chipset values were already tweaked by the
>>chipset manufacturer to US telco lines. If that is true, then 
>>running fxotune in the US has very little value.)
>>    
>>
>
>Sometimes in the US you still deal with line impedance issues.  In fact,
>I was told by an engineer that worked for the company that designs the
>line interface part that the bulk of echo problems (with line interface 
>parts such as this) are related to AC impedance mismatches (which is one
>reason why I haven't done the digital hybrid tuning portion of fxotune
>still).  It should work the same regardless of which kernel (2.4 or 2.6)
>you are using.
>  
>
Everyone has to deal with line impedance issues. The hyrbids in 
exchanges usually use a compromise impedance, but there are tolerance in 
the circuits. The lines themselves vary far more, especially if loading 
coil, and other fudges have been used. To get pretty good hybrid 
performance you always need to individually tune.

What you have to remember about hybrids is they are not there to give 
great rejection. They are there to give enough rejection to prevent 
howling. Nothing more. The telephone approvals specs in most places only 
call for 12dB of suppression through a hybrid. The exchange specs tend 
to use a slightly higher figure, but only slightly.

I don't follow why knowing that impedance mismatch is the problem has 
stopped you making fxotune fix it. :-\ Where you the one who asked me 
how to make fxotune work well on IRC? Someone asked a while ago, and 
said they were working on a faster tuning algorithm for fxotune. I've 
forgotten who.

>If it doesn't, and you have udev setup correctly, something is fundamentally
>wrong in the setup.
>  
>
Regards,
Steve