Kitz Forum

Broadband Related => Broadband Technology => Topic started by: BigAl on November 25, 2011, 05:51:23 PM

Title: Overnight SNR variations
Post by: BigAl on November 25, 2011, 05:51:23 PM
Hi all,

My SNR margin is still the default 6db. My connection is solid and my download speeds are excellent considering where I live and my line length. I am not about to interfere and upset the apple cart. However, I like to keep an eye on things and therefore log and record my stats  using RouterStats. My modem is a Netgear DG834G v4 running the firmware current).

I am puzzled though. It is regularly stated that the SNR will drop during evenings. This is certainly true. However, It is also regularly stated that this is due in a good part to extraneous domestic equipment starting up when everyone gets home from work. That does not add up to me as that would produce not only an evening drop but also a morning drop and a rise in the intervening period. That is not the case. My SNR drop: line gets more noisy,  certainly starts during the late afternoon/evening but then persists throughout the night more or less the same until it returns to its usual daytime levels at about 7:30 -8.:00 ish. (and that is an odd time when I think about it). Another possible source of noise that is put forward are street lights. We live in a rural location some 2 km from my exchange, by road and cable route 1.7 km as the crow flies. There are 6 street lights at the exchange end of the run. None anywhere else. So what else is there? I am wondering if it has something to do with the lack of daylight although the majority of my 'run' is underground.

I am I alone in wondering what is actually going on?


Title: Re: Overnight SNR variations
Post by: roseway on November 25, 2011, 06:44:56 PM
One other significant night-time effect is AM radio interference. During darkness, AM radio signals propagate further and more powerfully. If you look at the bitloading for an ADSL connection, it's quite common to see one or more holes whose frequency corresponds with powerful nearby transmitters, and during darkness many more transmitters start to have an influence.
Title: Re: Overnight SNR variations
Post by: BigAl on November 25, 2011, 07:21:43 PM
Our nearest transmitter is up on the Preseli Hills, some 35 miles away. I suppose that it still could have an impact but confess to being dubious. Both Digital TV and analogue radio signals are very poor here. Both 'drop out' frequently at least partially because we are down below a local hill line and well out of 'line of sight'. Additionally the bulk of our cable run is underground.

My local set up is through the BT NTE5 socket with the No. 3 ring wire disconnected. I have cut off all of our 'phones a couple of times and tried overnight straight into the test socket but the results were more or less identical: can't do that too often as I get complaints from, well, I'll let you guess who. The point is though, that I can more or less eliminate all domestic wiring as a source.

As I said, we are quite rural and I understand from a friendly BT engineer there are not actually many connections on our particular bit of the exchange internals. SNR seems to be such a grey area with so many variable I can't see how anyone could properly understand what is actually going on. Professional or, like me, amateur.

Thank you for taking the time to reply though.


Title: Re: Overnight SNR variations
Post by: waltergmw on November 26, 2011, 11:38:43 PM
Hi BigAl,

Ezzer here has said he's picked up REIN on both overhead and underground cables. One contributory factor is that the majority of all BT twisted pair cables are unscreened.

Radio propagation involves day / night changes in the Ionosphere (If it's still called that) as Ham radio people know. Local military sources might vary so it's not just public broadcasting frequencies which are involved.

Kind regards,
Title: Re: Overnight SNR variations
Post by: GunJack on November 27, 2011, 07:55:01 AM
sounds like totally normal behaviour for snrm :)  I run at 2.5-3dB daylight sync, drops to 1-2dB evening/night, rising again as daylight approaches. Walter's spot-on with his Ionosphere observations (and yes it is still called that ;) )
Title: Re: Overnight SNR variations
Post by: BigAl on November 27, 2011, 09:27:47 AM
Hi Gunjack

Thanks for that. I rather guessed that it was normal. It was just that with my limited knowledge of these things I didn't understand WHY it was normal.

And thank you waltergmw. Another thing added to my limited pool of knowledge. I had to look up REIN on the Kitz site ( but I am slowly 'getting it'

Cheers both.

Title: Re: Overnight SNR variations
Post by: Weaver on January 13, 2012, 05:51:56 AM
More crosstalk from evening activity on neighbours' lines? (dodgy assumption about some domestic users behaviour, they keep switching off dsl h/w in the day when they're not using it.) that possible?
Title: Re: Overnight SNR variations
Post by: jeffbb on January 15, 2012, 07:36:14 PM
To summarize SNR margin drop has many causes
Radio interference as already explained ,does sometimes show itself in the bit loading as a definite blip in the number of Data bits loaded where there is a significant affect from a particular trans mitter see list of UK   MW radio stations   ( and their frequencies . But as well as the obvious effect from a particular transmitter tthere may be many other smaller effects across the ADSL spectrum .
As also mentioned other sources of noise will become apparent as more users are on line through some cross talk .
both these affects tend to change the SNR margin gradually . If there is a sharp increase /decrease in SNR margin at various intervals or definite times  then this is most likely some equipment actually being switched on/off at those times .
Regards Jeff