Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
   Glossary   Glossary
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Author Topic: A question on noise margin fluctuations  (Read 5956 times)

renluop

  • Kitizen
  • ****
  • Posts: 3326
A question on noise margin fluctuations
« on: December 21, 2012, 09:14:46 PM »

I have noticed that upstream margins fluctuate more erratically, sample to sample, than the downstream.

I am curious why that should be.

Education time again! ;D
Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33881
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: A question on noise margin fluctuations
« Reply #1 on: December 21, 2012, 10:30:32 PM »

I cant give you a definative answer why this would be. 
The obvious of course is that there is more fluctuation at those channels.

Something else that may very well affect this....  is that SNRM is usually shown as an average over the whole range. 
The upstream channel has far less sub-tones than the downstream channel does. Therefore a fluctuation in just one of the upstream frequencies is going to have more impact on the overall SNRM than what it would across the downstream range. 
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

roseway

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 43467
  • Penguins CAN fly
    • DSLstats
Re: A question on noise margin fluctuations
« Reply #2 on: December 21, 2012, 10:43:47 PM »

It's something which I see on my connection too. It must be some kind of interference which is predominantly below ~140 kHz, so it's not likely to be radio interference. I would guess some nearby electrical equipment with a dodgy power supply, or something of that nature.
Logged
  Eric

sheddyian

  • Kitizen
  • ****
  • Posts: 1159
    • My Shed Blog
Re: A question on noise margin fluctuations
« Reply #3 on: December 21, 2012, 11:22:23 PM »


I too see this.  My downstream SNRM is a remarkably flat 6dB with almost no variation, but the upstream will typically vary between 8.9 and 10.1dB.

Ian
Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33881
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: A question on noise margin fluctuations
« Reply #4 on: December 22, 2012, 12:23:06 AM »

Just theorising..   (warning  :drink: )

It must be some kind of interference which is predominantly below ~140 kHz, so it's not likely to be radio interference.

Problems with the upstream are often caused by faulty wiring/incorrect filtering.  
Without filters (or even cheapy filters) its not unknown for fax machines & sky boxes etc to encroach into the upstream channel range.

In the home we use low pass filters.  
What about crosstalk on the line/FeXT.  How many lines are unfiltered due to not having dsl, but still use sky or fax machines?
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

JGO

  • Reg Member
  • ***
  • Posts: 729
Re: A question on noise margin fluctuations
« Reply #5 on: December 22, 2012, 08:06:57 AM »

" so it's not likely to be radio interference. I would guess some nearby electrical equipment with a dodgy power supply, or something of that nature." [/quote]

Why is it assumed that electrical noise is caused only by FAULTY equiplent ? In my experience supression only reduces noise, always assuming someone hasn't removed it. I recal another posting where someone was in denial,  his broadband speed fell by 95% whenever he switched on his new and expensive plasma TV . It COULN'T be the TV !
Logged

roseway

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 43467
  • Penguins CAN fly
    • DSLstats
Re: A question on noise margin fluctuations
« Reply #6 on: December 22, 2012, 10:07:14 AM »

Quote
Why is it assumed that electrical noise is caused only by FAULTY equiplent ?

Nobody is making any assumptions here, just suggestions. Because of the comparatively low frequency, it's not so likely to be transmitted electrical interference (although it could be). It's more likely to be dodgy electrical equipment or, as Kitz says, a wiring/filtering issue. It could also be a line fault, but the kind of fault which has this effect tends also to produce audible noise on the telephone.
Logged
  Eric

renluop

  • Kitizen
  • ****
  • Posts: 3326
Re: A question on noise margin fluctuations
« Reply #7 on: December 22, 2012, 10:59:21 AM »

Thanks for replies
Quote
...It's more likely to be dodgy electrical equipment or, as Kitz says, a wiring/filtering issue. ...
Any most likely  culprits come to mind? Our tree is illuminated but lights are static ones, is that even so a possible?

Router is connected by a hard wired in cavity extension out of the back of the master. I thought filtering was to do with downstream. As far as that aspect is concerned all seems OK, unless the router's not recovering its target SNRM during daylight is an indicator. Comments?
Logged

roseway

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 43467
  • Penguins CAN fly
    • DSLstats
Re: A question on noise margin fluctuations
« Reply #8 on: December 22, 2012, 11:28:03 AM »

Quote
I thought filtering was to do with downstream.

No, filtering is intended to separate the telephone frequencies (up to 4 kHz) from the ADSL frequencies (above 25 kHz). The ADSL frequencies starting at ~25 kHz are the upstream ones, and these are the frequencies which are more likely to be affected by imperfect filtering.

As to the culprits, there's large range of possibilities. Normally the first thing to do is to connect the router to the test socket inside the master socket, to eliminate the internal wiring. But if I understand you correctly, your extensions are directly wired to the back of the socket and not into an NTE5 faceplate. If you haven't got an NTE5 then the best technical option would be to get an NTE5 fitted by BT.

That being said, it may not be worth the bother. If your connection is stable and adequately fast, the variation in upstream SNR margin is really of no consequence, so the best thing to do may be nothing.
Logged
  Eric

renluop

  • Kitizen
  • ****
  • Posts: 3326
Re: A question on noise margin fluctuations
« Reply #9 on: December 22, 2012, 11:52:53 AM »

I was having one of those elderly "what/who was it" moments. The master is the modern split one, so is NTE5. :-[ :blush:
Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33881
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: A question on noise margin fluctuations
« Reply #10 on: December 23, 2012, 07:50:14 AM »

Because of the comparatively low frequency, it's not so likely to be transmitted electrical interference (although it could be). It's more likely to be dodgy electrical equipment or, as Kitz says, a wiring/filtering issue. It could also be a line fault, but the kind of fault which has this effect tends also to produce audible noise on the telephone.

To expand further on my rambling the other night, which I perhaps didnt explain too well.   

I was wondering about the effect on the lower tones of cross talk.  These frequencies may often be encroached by fax machines etc so I was wondering of what the possibility could be from neighbouring lines which may be unfiltered.

---

I still think the much lower number of upstream bins in use when compared to 10-20x times more downstream bins and because the SNRm is reported as an average will show fluctuations more in the upstream.    eg a fluctuation of just 2 or 3 upstream bins will more likely affect the upstream SNRM and be reported... but fluctuations of 2 or 3 downstream bins wouldnt make much impact on the overall downstream SNRM.
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

renluop

  • Kitizen
  • ****
  • Posts: 3326
Re: A question on noise margin fluctuations
« Reply #11 on: January 01, 2013, 02:10:49 PM »

Happened to notice that when phone rang this morning US SNRM dropped from 5.2 to 2dB. Nearest phone to router is DECT slave ca. on desk 4ft from router, but not in sight. Main phone is in hall downstairs.

Something to watch if it happens again?
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: A question on noise margin fluctuations
« Reply #12 on: January 01, 2013, 02:56:14 PM »

Something to watch if it happens again?

Yes, indeed. It is beginning to look as if you have a joint, somewhere in your pair, that is beginning to show either HR or semiconducting tendencies.

If you were to perform a quiet-line test (call 170170, option 2) would you hear anything other than silence?
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

renluop

  • Kitizen
  • ****
  • Posts: 3326
Re: A question on noise margin fluctuations
« Reply #13 on: January 01, 2013, 06:44:36 PM »

Phone rang this afternoon and there was again a dip but not as much as before. Did QLT with analogue phone and that too produced a dip, but according to wife ( better hearing) no noise.

Code: [Select]
17:03:53 Rx-Noise=1.1 Tx-Noise=4.8 Rx-Sync=7918 Tx-Sync=1029
17:04:53 Rx-Noise=0.9 Tx-Noise=5.6 Rx-Sync=7918 Tx-Sync=1029
17:06:43 Rx-Noise=0.8 Tx-Noise=3.9 Rx-Sync=7918 Tx-Sync=1029
17:07:43 Rx-Noise=0.8 Tx-Noise=3.6 Rx-Sync=7918 Tx-Sync=1029
17:08:43 Rx-Noise=0.8 Tx-Noise=3.9 Rx-Sync=7918 Tx-Sync=1029
17:09:43 Rx-Noise=0.7 Tx-Noise=4.5 Rx-Sync=7918 Tx-Sync=1029
17:10:43 Rx-Noise=0.9 Tx-Noise=6.0 Rx-Sync=7918 Tx-Sync=1029

18:05:44 Rx-Noise=1.1 Tx-Noise=5.4 Rx-Sync=7918 Tx-Sync=1029
18:06:44 Rx-Noise=1.0 Tx-Noise=5.4 Rx-Sync=7918 Tx-Sync=1029
18:07:44 Rx-Noise=1.0 Tx-Noise=4.1 Rx-Sync=7918 Tx-Sync=1029
18:08:44 Rx-Noise=1.0 Tx-Noise=3.2 Rx-Sync=7918 Tx-Sync=1029
18:09:44 Rx-Noise=1.1 Tx-Noise=5.5 Rx-Sync=7918 Tx-Sync=1029
18:10:44 Rx-Noise=1.1 Tx-Noise=5.4 Rx-Sync=7918 Tx-Sync=1029
18:11:44 Rx-Noise=1.0 Tx-Noise=5.5 Rx-Sync=7918 Tx-Sync=1029

TBH I wouldn't be surprised if there was some water ingress with the weather. Muck and water pours off the Forest on to our cul de sac, and is slow to drain.

Watch and wait I guess. :)
Logged
 

anything