Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[ken0062]

Hi

I am trying to squeeze a bit more performance out of my poor line (63db attenuation).
Looking at the snr and tones graph it appears that I have a very poor SNR on bits 33 to 37.
Is this normal or is there likely to be anything I can do to trace the problem ( I already running router on master socket with NTE5 face plate)
My line stats are :-
Noise Margin:     7.6   dB
Connection Rate:  2048  Kbps
Line Attenuation: 63.5  dB
Power:            16.9  dBm

SNR margin is usually very stable usually only fluctuates by +- 1dB 

Thanks

[attachment deleted by admin]

[roseway]

I don't know the reason, but I don't think it's particularly unusual. My connection also has no bitloading on tones 33-37, and if you look at the Kitz bitloading page (near the bottom) you'll see that two of the three bitloading graphs there have the same behaviour. Maybe someone else can explain it.

[burakkucat]

Like Eric, I do not know the answer. However, let's see what can be deduced.

The tones in question, 33 to 37, have a mid point of 35. Going to Kitz's appropriate page and making use of the ADSL Frequency Bins & Tones converter, we can see that tone 35 equates to approximately 151 kHz. So, the question is: "Is there any powerful transmission at that approximate frequency within the UK?". That frequency is stirring something in my memory. What is the wavelength? Hmm . . .

300000000 / 151000 ~= 1986 metres.

 

[roseway]

There are a few European radio stations at 153 kHz, but they're far enough away that I doubt they could be the cause.

[waltergmw]

@ Eric,

I too have no idea why it should be so but I suspect there's some logic within the bitloading algorithm.
Perhaps one for a guru to investigate further?

Below are two (Horrible) 2Wire diagnostic pictures of a 21CN ADSL1 service with a small gap and a 20CN ADSL1 service with a large gap at the same address fed from the same 20 pr cable.

Kind regards,
Walter

[attachment deleted by admin]

[Oranged]

Must be an idiosyncrasy of long lines or perhaps BT IPStream ?

My attenuation is 38dB and tones 33 to 85 are all either 14 or 15 bits on O2 LLU.

[jeffbb]

Hi

quote@ Oranged Must be an idiosyncrasy of long lines or perhaps BT IPStream ?

My attenuation is 30db (adsl1) So not a long line but I have the same sort of effect on tones  33 to 36  see attachment . A point to note is that there is an  apparent increase in attenuation on those tones forcing the SNR margin down  .  I wonder if its an extension of the spectral mask ?.

quote  from Kitz Also note the smooth curves at on the upstream causing a inverted 'U' and also at the beginning of the downstream. This is likely due to spectral masks applied at the DSLAM which ensures that power levels are cut back to reduce the likely hood of cross-talk on those frequencies.

Read more: http://www.kitz.co.uk/adsl/adsl_technology.htm#bit_loading#ixzz1MWJmYVEX

quote @ Oranged My attenuation is 38dB and tones 33 to 85 are all either 14 or 15 bits on O2 LLU.
Must be that the SNR is very good ,around 50db for those tones
Regards Jeff



[attachment deleted by admin]

[waltergmw]

Gentlefolk,

Here's the data on a PlusNet 21CN ADSL1 regrade indicated by the change to the Infineon vendor. To contradict my earlier pictures, the tone 31 32 gap has remained constant. The new BT speedtester results are similar to those results observed on the 20CN BT speedtester. The TBB speed test also shows a similar speed of about 820 Kbps before and after the regrade.
This suggests the BT & TBB calculations are different and perhaps the way they handle the other internet traffic caused by my LogMeIn connection.
After I took the observations I did a remote soft reboot of the modem but it had no noticeable effect on most readings.
The line stability has improved but it's still early days for those observations.

Kind regards,
Walter

[attachment deleted by admin]