[Asterisk-Users] Chaining more than one zap echo canceller?

[Chad Scott]

Apparently you did not read my entire message.

I specifically stated it would be non-optimal and a bare-bones solution.

While 600 ohms may be the characteristic impedance of the wire run, a 
mismatch at either end will change the impedance of the entire path to 
some value that is the related to the severity of the mismatch and *the 
distance from it*.  The CPE "sees" an impedance of X because it is at Y 
distance from the mismatch.  You could have a horrible mismatch 
somewhere on the line and still see 600 ohms because you happened to be 
at the right distance.

And yes, the longer the run, the higher its DC resistance even though 
it's impedance is still 600 ohms.  However, in the AC world, there is a 
measure of impedance and loss per fixed distance.  So the loop might be 
20 miles long, 600 ohms impedance, and 1.5 dB loss per mile (as an 
example).  Since the resistance increases with distance, something must 
be happening within the wire to help overcome the resistance in order 
to stay at a fixed 600 ohms: capacitance and inductance.  However, the 
loss increases with distance, so you may have a 600 ohm match at the 
far end but no signal to pick up.

If you wanted to fix an impedance mismatch the "right" way, you'd use a 
matching network.  In it's simplest form, you could use a transformer 
to convert the 150 ohms impedance at the jack to 600 ohms for the 
equipment.  You could also use a dynamic matching network with variable 
capacitors and inductors to create just about any impedance you want.

On Oct 12, 2004, at 2:58 PM, Rich Adamson wrote:

> Adding resistance to one side of the line only begs for problems
> as it creates a tip-ring imbalance that will "cause" echo, etc,
> when other imperfections exist.
>
> If that approach works at all for anyone, its addressing a symptom
> and not the root cause.
>
> Try this one: Each customer loop is made up of copper and the longer
> the copper, the more resistance. Yet the impedance (in the US) is
> consistently 600 ohms. A short loop might be a 100 ohms while a long
> loop might be well over 1500 ohms; still both are 600 ohm impedance.
>
> ------------------------
>
>> Impedance is the measure of total opposition (resistance, capacitance,
>> and inductance) to alternating current flow.  Adding resistance will
>> raise the impedance of the line.
>>
>> On Oct 12, 2004, at 12:58 PM, Rich Adamson wrote:
>>
>>> Impedance does not equal resistance. Apples and oranges.
>>>
>>> ------------------------
>>>
>>>> If you're certain it is an impedance problem and the impedance of 
>>>> your
>>>> line is lower than that of the CO, you can increase the impedance of
>>>> your line by putting a potentiometer in-line and adjusting it until
>>>> the
>>>> sidetone disappears.  This is a bare-bones solution and decreases 
>>>> the
>>>> efficiency of the line because you're putting in pure resistance.
>>>>
>>>> If your impedance is higher than the CO, or if you want to be more
>>>> efficient, you'll need a more complicated impedance matching 
>>>> network.
>>>>
>>>> On Oct 12, 2004, at 10:26 AM, Kris Boutilier wrote:
>>>>
>>>>>  I have Asterisk connected to a channel bank via a t100p card. 
>>>>> There
>>>>> excessive sidetone generated on the analog side due to an impedance
>>>>> mismatch
>>>>> - I am very close to my serving CO which brings the line down to
>>>>> about
>>>>> 150ohms and the channel bank is expecting 600ohms. However, the 
>>>>> very
>>>>> loud
>>>>> sidetone is being fairly effectively supressed by the zap echo
>>>>> canceller and
>>>>> I have quite usable lines as a result.
>>>>>
>>>>>  From time to time calls are placed to other PSTN numbers (even
>>>>> terminating
>>>>> off of the same CO) that are introducing their own far end echo
>>>>> component on
>>>>> the line and it isn't being supressed at all - thus my outbound
>>>>> callers
>>>>> begin to hear thei          rownvoiceandgetveryfrustrated.Ihaveto
>>>>> assume the
>>>>> second reflection couldn't be supressed by design - the zap echo 
>>>>> can
>>>>> has
>>>>> already locked on to the sidetone echo after all.
>>>>>
>>>>>  So, I have a fix for the sidetone/impedance problem (PRI on order)
>>>>> however
>>>>> is it possible to insert another can into the system somehow? For
>>>>> example,
>>>>> if I were to run TDMoE to a second box and access to the
>>>>> t100p/channel
>>>>> bank
>>>>> through there?
>>>>>
>>>>> Any suggestions welcome (apart from curing the sidetone)  :-)
>>>>>
>>>>> Kris Boutilier
>>>>> Information Systems Coordinator
>>>>> Sunshine Coast Regional District
>>>>>
>>>>> _______________________________________________
>>>>> Asterisk-Users mailing list
>>>>> Asterisk-Users at lists.digium.com
>>>>> http://lists.digium.com/mailman/listinfo/asterisk-users
>>>>> To UNSUBSCRIBE or update options visit:
>>>>>    http://lists.digium.com/mailman/listinfo/asterisk-users
>>>>
>>>> _______________________________________________
>>>> Asterisk-Users mailing list
>>>> Asterisk-Users at lists.digium.com
>>>> http://lists.digium.com/mailman/listinfo/asterisk-users
>>>> To UNSUBSCRIBE or update options visit:
>>>>    http://lists.digium.com/mailman/listinfo/asterisk-users
>>>
>>> ---------------End of Original Message-----------------
>>>
>>>
>>> _______________________________________________
>>> Asterisk-Users mailing list
>>> Asterisk-Users at lists.digium.com
>>> http://lists.digium.com/mailman/listinfo/asterisk-users
>>> To UNSUBSCRIBE or update options visit:
>>>    http://lists.digium.com/mailman/listinfo/asterisk-users
>>
>> _______________________________________________
>> Asterisk-Users mailing list
>> Asterisk-Users at lists.digium.com
>> http://lists.digium.com/mailman/listinfo/asterisk-users
>> To UNSUBSCRIBE or update options visit:
>>    http://lists.digium.com/mailman/listinfo/asterisk-users
>
> ---------------End of Original Message-----------------
>
>
> _______________________________________________
> Asterisk-Users mailing list
> Asterisk-Users at lists.digium.com
> http://lists.digium.com/mailman/listinfo/asterisk-users
> To UNSUBSCRIBE or update options visit:
>    http://lists.digium.com/mailman/listinfo/asterisk-users