[BC] audio cable termination

[R A Meuser]

This is all covered in the original AES paper from 1980.

http://www.richardhess.com/be/aes-80.htm



Peter Smerdon wrote:
> Why do we terminate cables? Because broadcast borrowed the practice from 
> the telephone companies, and old habits (and legacy installations) die 
> hard.
> 
> My understanding is that properly terminating a cable requires the 
> source impedance, cable characteristic impedance, and load impedance to 
> be the same
> - and -
> You would only terminate a cable when either:
> a) the cable length was such that it was acting as a transmission line - 
> rule of thumb (heard from Bill Whitlock of Jensen Transformers - who's a 
> bit of an evangelist on this subject) is that for a cable length to 
> start looking like a transmission line it has to be at least 10% of a 
> wavelength - and that is 4000 feet at 20kHz (to put Cowboy's "short" and 
> "long" into perspective)
> -or-
> b) you are interested in maximum power transfer, not maximum voltage 
> transfer.
> 
> Here I'm assuming the application is "short" cable lengths at audio 
> frequencies.
> 
> In distributing audio, we are interested in voltage, not power, and if 
> we were to terminate, we'd be just throwing away 6dB in level 
> (headroom/noise) for no benefit.
> For audio signals the modern practice of low source impedance (say 50 
> ohms) and high input impedance (10K+) eliminates the need for terminations.
> 
> Ya just gotta watch out for interfaces to long lines, your telco or 
> campus network - and if interfacing with a transformer designed for 600 
> ohms, where failure to terminate adversely affects the transformer 
> performance.
> 
> btw: the 600 ohms line impedance comes from the fact that that is the 
> characteristic impedance of the typical "open wire" telephone line 
> strung on poles with wooden crossbars and porcelain insulators. Audio 
> cable Zo is usually somewhere between 80 and 100 ohms.
> It's just as well maintaining correctly sourced and terminated cables 
> was unnecessary, 'cos they were designed for the wrong impedance.
> 
> Cowboy wrote:
> 
>> On Thursday 03 January 2008 11:51 pm, Mario Hieb wrote:
>>
>>>  Terminating audio lines and other cables is a subject that interests 
>>>  me and tends to come up now and again.
>>
>>
>>
>>>  So I throw the question out to all of you; why do we terminate 
>>>  cables? When should we terminate cables?
>>
>>
>>  Depends on why there is a cable there at all !
>>
>>  In most cases, the objective is to take a signal, or energy, and
>>  simply transport it from one location to the other, unchanged.
>>  In other cases, it is desirable to change the signal in some fashion.
>>  In any case, the cable itself is a circuit component, with its own
>>  characteristics, which may be desirable characteristics, or
>>  undesirable characteristics and will have some affect on the
>>  signal or energy being transported, which may or may not
>>  be significant.
>>  If the cable is "short" the effects may be insignificant, in which case
>>  the decision to terminate will depend solely on what affect on the
>>  signal is created by the termination itself.
>>  If the cable is "long" and the effects significant, then the decision
>>  to terminate will be dependent on whether it is desired to augment,
>>  mitigate, or in some other fashion affect some control on the effect
>>  of the cable on the signal, in concert with the affects of the 
>> termination.
>>  Therefore, the decision to terminate, and with what, is a design
>>  decision based on the function of the system as a whole.
>>
>