My line failed the test due to the noise on it being in excess of the 60db
limit. Does anyone know what the limit is for the 10km limit they are
testing.

the distance doesnt matter, with a signal to noise ratio of that, adsl just
wont work, sorry
"Richard Rollins" <deadeyedic@tiscali.co.uk> wrote in message
news:40e1a741_2@mk-nntp-2.news.uk.tiscali.com...

"Richard Rollins" <deadeyedic@tiscali.co.uk> wrote in message
news:40e1a741_2@mk-nntp-2.news.uk.tiscali.com
There is no limit for the trial - they are trying everything to see what
works and what doesn't..

I would say the new limit will be at least 70dB.

Regards
Sunil

news:cbsg6c$1le$1@hercules.btinternet.com
In excess of 60 db would be an exceptionally good signal to noise ratio
(lots of signal and very little noise).

However it is a high line loss (which is related to distance) and that is
why you may not be able to get ADSL.

--
Old Codger
e-mail use reply to field

What matters in politics is not what happens, but what you can make people
believe has happened. [Janet Daley 27/8/2003]

marky <mark.farmer@net.com> wrote:
<looks at current attenuation>

63.9dB.
Working well.
SNR of 11dB.

"Brian Morrison" <scrapspam@fenrir.org.uk> wrote in message
newsan.2004.06.29.21.22.37.250548@fenrir.org.uk
My downstream attenuation is 52/53 dB. S/N is rarely above 10 dB, and I
have never seen it above 15 dB. Currently S/N is 6 dB. I do get periods
when sync will drop for very short intervals, sometimes annoyingly often,
but when the S/N is down at 5 or 6 dB the connection is usually as solid as
a rock.

--
Old Codger
e-mail use reply to field

What matters in politics is not what happens, but what you can make people
believe has happened. [Janet Daley 27/8/2003]

being a broadband engineer for BT, I was just pointing out that with a line
loss such as 60db, it will not be stable, the other guy that said he had 63
db, how do you know that,?.....we have various ways to test, but not any
available to the public.


"Sunil Sood" <news@soods.freeserve.co.uk> wrote in message
news:2ke1kbF1a014U1@uni-berlin.de...

marky <mark.farmer@net.com> wrote:
I captured a BT engineer and have him tied down in my dungeon, as well
as assorted plumbers and other tradesmen.

Be warned, they are picky eaters, he turned his nose up at the Sky
engineer, eventually I had to provide Ketchup.

"Old Codger" <Oldcodger@anyoldwhere.net> wrote in message
news:CelEc.848$Fc7.198649@stones.force9.net...

I use Solwise 715 routers and when I connect via telnet I can ask for ADSL
performance figures and this is what I see

2 Meg Plusnet connection been working fine for over 2 and a half years now


Login successful

--> adsl show perf
Downstream :
Noise Margin : 6.5 dB
Attenuation : 64.5 dB
Output Power : 19
Upstream :
Noise Margin : 8.5 dB
Attenuation : 31.5 dB
Output Power : 9

-->

2 Meg Plusnet connection been working fine for over 1 year now

Login successful

--> adsl show perf
Downstream :
Noise Margin : 7.0 dB
Attenuation : 63.5 dB
Output Power : 19
Upstream :
Noise Margin : 9.5 dB
Attenuation : 30.5 dB
Output Power : 9

-->

So if 60db is the maximum how the hell have I got 2 fully working 2 meg
connections well it ain't what you know its who you know


Are these figures bad then?

cheers for all replies

In article <pan.2004.06.29.21.22.37.250548@fenrir.org.uk>,
Brian Morrison <scrapspam@fenrir.org.uk> writes:
Given the power level of the ADSL signal at launch (which I don't know),
how much attenuation is required to take that signal down to the thermal
noise level?

--
Tim Clark

On Thu, 08 Jul 2004 20:32:18 GMT, "Tim Clark"
<Timothy.Clark@btinternet.com> wrote:

about 60 dB takes it into the background noise level in some places,
maybe 70 dB in others and perhaps 55 dB in a few cases.

its not the thermal noise its all the other stuff in the ADSL
frequency band (AM radio for example)

Phil
--
spamcop.net address commissioned 18/06/04
Come on down !

In article <m7dre0dv135lait9qfmrkb0ofquuihsbfm@4ax.com>,
Phil Thompson <phil.thompson@spamcop.net> writes:
Thanks Phil. It shows that BT's talk of extending ADSL provision to
perhaps 10 km, has to be taken with a large pinch of salt if average
line loss on a typical phone line is around 10 dB/km.

I've answered my own question, prompted by a statement by
Mugwump <me@privacy.net> in article
<MPG.1b5639381b02298d989c28@news.claranews.com>:

Using the Web I refreshed my memories of Boltzman's constant and
calculated the following (^ denotes exponentiation):

Boltzman's constant k = 1.38 x 10^-23
Thermal noise at temperature T is kT
Thus for a typical temperature around 290 kelvin (17 celcius) thermal
noise is 4 x 10^-21 W or 4 x 10^-18 mW hence -186 dBm.

Attenuation of 148 dB would therefore reduce the DSLAM launch signal
(at -38 dBm) to the level of thermal noise (-186 dBm).

It's of theoretical interest only of course, as Phil points out,
noise from EMI etc. is at far higher levels.

If there are any errors in my calculations, please point them out, it's
been a good 30 years since I last had occasion to use Boltzman's
constant.

--
Tim Clark

Tim Clark wrote:

That's in a 1Hz bandwidth. What's the bandwidth of DSL? 1MHz? That puts
the noise power up 60dB to -126dBm.

On Fri, 09 Jul 2004 10:12:50 GMT, "Tim Clark"
<Timothy.Clark@btinternet.com> wrote:

I have 45 dB loss with 27 dB SNR *margin* = 33 dB of SNR (I assume) so
the signal at the DSLAM output is 78 dB more than the noise floor.
Guessing the latter is at -80 dBm would put the DSLAM at -2 dBm of
output. I think this is well below full output, it wil be reduced as
the SNR is good.

I suspect the output maximum is likely to be +38 dBm not -38 as Wifi
kit is +20 dBm output.

http://www.linear-tech.com/ezone/dsl.html suggests 20 dBm downstream
and 13 dBm upstream.

I suspect the *upstream* is the real limit. RADSL allows adaption of
rate in the upstream to compensate for line loss. It does this by
using less clever coding to put fewer bits into each frequency bucket
in the *same frequency range* as ADSL without rate adaption. In other
words it works with a lower SNR in the upstream band, leaving the
downstream alone.

So the limiting factor may be how much of your modems output arrives
at the DSLAM relative to the amount of noise the DSLAM receives.

My upstream margin is also 27 dB and the loss 24 dB, so 57 dB is the
difference between my modem output and the DSLAM noise floor - much
less than the 78 dB on the downstream on account of the lower
frequency. This suggests my line could double in loss and maintain
QoS, assuming the modem can't turn itself up to overcome the extra
loss. I'm 2.5 km from the exchange.

If the noise floor is at -70 at both ends and we need 10 dB of SNR for
good operation then the received signal has to be -60 dB, with a DSLAM
output of 20 dBm the line loss downstream can be 80 dB and with a
modem output of 13 dBm the upstream loss can be 73 dBm.

Lower noise floors, lower acceptable SNRs and bigger outputs all allow
operation with more attenuation. 10km might just creep in with +20 DBm
at the DSLAM, 100 dB attenuation and -86 dBm or less noise.


Phil
--
spamcop.net address commissioned 18/06/04
Come on down !

"Tim Clark" <Timothy.Clark@btinternet.com> wrote in message news:<m0nlcc.p41.ln@basil>...
You did leave out the bandwidth, so assuming a bandwidth of about 1MHz
the calculation becomes:

1.38 x 10^-23 x 290 x 10^6 = 4 x 10^-15 W

or 4 x 10^-12 mW

or -114 dBm

John

In article <m7dre0dv135lait9qfmrkb0ofquuihsbfm@4ax.com>, Phil Thompson
<phil.thompson@spamcop.net> writes

it isn't thermal noise that matters - it's Near-End & Far-End Crosstalk
isn't it??
--
robert w hall