Return to KLUBNL.PL main page

rsgb_lf_group
[Top] [All Lists]

LF: Re: Re: ADSL EMC with LF receiving?

To: [email protected]
Subject: LF: Re: Re: ADSL EMC with LF receiving?
From: "Stewart Nelson" <[email protected]>
Date: Thu, 5 Dec 2002 18:22:40 +0100
References: <[email protected]>
Reply-to: [email protected]
Sender: <[email protected]>
Hi Markus and all,

Interference to LF reception by DSL:  The +12 dBm figure is total power.
I believe that the spec requires the PSD to be a maximum of -30 dBm/Hz;
looks like my modem is running about -34 dBm/Hz.  It's possible that
some providers run illegal power on noisy lines (sound familiar?).  A
typical CPE ADSL line driver can put out 40 V p-p differential into 100
ohms.  If there are 20 upstream bins, linear operation would entail no
more than 2 V p-p per bin, which corresponds to about -26 dBm/Hz.  The
above figures assume spectral peaks 3 dB above the average.

Even at the higher figure, I would think that with Cat-5 cable from the
modem to where the aerial or underground phone cable enters the house,
you should be fine, as long as your antenna was at least a few meters
from the Cat-5, and you put the DSL low pass filter near the cable entry.
If you still had trouble, I would first suspect the SMPS for the modem!

Interference to DSL downstream from LF Tx:  I agree with Markus that
DSL is very robust; disruption of TCP traffic is extremely unlikely.
If your application is harmed by individual lost packets, e.g. Internet
telephony, then you might lose a little audio each time your Tx keys on :)

73,

Stewart KK7KA

----- Original Message ----- From: <[email protected]>
To: <[email protected]>
Sent: Wednesday, December 04, 2002 11:51 PM
Subject: LF: Re: ADSL EMC with LF receiving?


Hi all,

most of the messages in this thread have focussed on potential interference from the LF-TX to the ADSL connection. Being digital, I would assume that below a certain threshold there would be no noticable interference at all, and even above that level, error correction protocols and adaptive channel allocation would make it relatively robust.

However, the emissions from an inadequately shielded ADSL line could be a significant noise source for LF reception. - The background noise level here on a quiet day is around -25dBuV/m in 1Hz, which on the terminals of my Marconi (having -31dB efficiency) is about -107dBm/Hz or -81dBm/4kHz. Stewart's log excerpt indicates a modem power of 12dBm, is that power per channel? If so, the isolation between ADSL and the LF antenna would have to be better than 93dB, which seems fairly demanding. With small receive loops in the vicinity of the wiring, the situation could be even worse.

As I am still using plain old analog telephony and a V.90 modem here, this is only a rough theoretical estimate, and I wonder if somebody has experienced any problems in actual practice.

Regards
Markus, DF6NM




<Prev in Thread] Current Thread [Next in Thread>