|
Posted by DerbyDad03 on January 13, 2007, 12:40 am
Todd,
Thanks for the detailed response. Happy viewing.
Todd H. wrote:
>
> > Could you explain the bi-directional amp comment?
>
> The amps aren't bi-directional really, it's just that they have a
> passive return path so as to minimize their interference with the
> upstream data transmission from the cable modem back to the head end.
> THe cable modem has its own gain/amplifier for that path, and part of
> it's "training up" in negotiation with the head end is setting that
> gain. Without a passive return in an amp, the upstream channel is
> pretty well doomed.
>
> However, it is best to not have an amp in the cable modem's way at
> all. The preferred method is to use a directional coupler (aka one
> way tap) that has minimal insertion loss on the through leg, and 6dB
> or so loss on the tap leg. The through leg goes to the cable modem
> relatively unmolested, and the -6dB depleted tap leg gets run
> that through an amplifier, to feed all bagillion outlets in the home
> if needed.
>
> In one case I had in a house the cable modem was on the 2nd floor far
> from place the cable came into the house. IN that case, they had to
> use 2 directional couplers, and even go to the higher quality, more
> directional couplers they refer to as DC9's to get low enough return
> path loss to get a usable upstream path. And because there were tv's
> on both floors of the home, 2 separate amplifiers were used, but none
> was in the path to/from the cable modem.
>
> The trouble with amplifiers, even passive return ones is that they add
> noise, and all you need is one class about information theory to know
> why noise is detrimental to data transmission.
>
> Best Regards,
> --
> Todd H.
> http://www.toddh.net/
| 
Yahoo! 
Google 
Windows Live 
del.icio.us 
digg 
Netscape 
XML Sitemap