+11dBmV DS - Power level(s)

[George_w]

Just A quick question, as I know I can expect a quicker reponse from public members than on the VM forums, Just looking for feedback really.

So here's the score, As of this morning (8am - My US was 56dBmV - and DS (4 channels) was all between 4-6dBmV depending on the time of day.

VM Engineer come out at 9am, checked the line, put new connecters on and a new splitter outside, but after all that failed and it become 9dBmV and up to 56dBmV. He rang for Central Networks? to come out.

They arrived at 11ish, (Didn't knock on, just seen them outside) - Net went off a couple of times and then notice they cleared off after dinner, leaving my power levels at over 10dBmV (current sitting at 11dBmV) and US at 49.8dBmV.

My question is, how safe are these levels for the superhub?

I'm awaiting a reponse from the admin(s) from the VM Forum for the next step. Just annoyed that I've had today off and really it seems like it was for nothing. Why couldn't this of been sorted the day they put the SH in (2 weeks ago) - The levels have been :s since the thing was put in.

Check the graphs below, as you will see around the times I have said above, you will notice the fluxuations while they was doing work on the cab.

Download Failed (1)

[Sephiroth]

Because your downstream power has gone up by c. 5 dBmv and your upstream has fallen by c. 5 dBmv, it seems that all they did was to move your tap point - perhaps from 29 dB attenuation down to 22 dB. The should have put a 10 dB FPA on your modem to compensate for the high DS power.

But they won't have cured the root cause of hig upstream power. I imagine you're close to the cabinet, in which case your upstream power should have been much lower - closer to or below 40 dBmv.

So, how far from the street cab are you and what do you think the network bods did? Your power levels are too high and can affect performance (e.g. crosstalk). The DS levels will rise fuurther as it gets colder.

[George_w]

The little green one they looked at (I Could see them out the window) so it's really close, it's 2 houses next to me.

If you look at the graphs between 11-1 - thats when they was there messing about. It was 9dBmV earier, it's risen to 11dBmV right now.

[Sephiroth]

So, what I say holds. Your downstream power may be causing crosstalk and there's possibly an upstream impairment (maybe at the optical node) to make matters worse.

[George_w]

I Guess that means bad news for me then... Just waiting back on a response. Apparantly the case has been closed but I don't see how when my power levels are looking like this!

[Chrysalis]

wow how you managing to graph the data? a friend of mine on tbb is trying to setup the same with some scripts.

[jb66]

Network techs won't have just moved up a tap, they dont do simple jobs like that. this sounds like Sid architecture and they've tweaked the power of the amp or swapped the tap for a less attenuated one

[Sephiroth]

Network techs won't have just moved up a tap, they dont do simple jobs like that. this sounds like Sid architecture and they've tweaked the power of the amp or swapped the tap for a less attenuated one

That's what I said - JB. I said "perhaps from 29 dB attenuation down to 22 dB". Whether that's up a tap or down a tap depends on the box layout!

[Ignitionnet]

Those levels are fine going back to the original question, so long as it's working it's all cool.

---------- Post added at 14:39 ---------- Previous post was at 14:36 ----------

So, what I say holds. Your downstream power may be causing crosstalk and there's possibly an upstream impairment (maybe at the optical node) to make matters worse.

Downstream power causing crosstalk?

Please explain, you've my attention!

As a reminder the DOCSIS 3 modems digitally sample a big band and extract channels from there so if you're thinking of the isolation between channels being overpowered it's not going to happen, moreover the isolation between frequencies on standard modems must be sufficient to handle the DOCSIS specifications.

---------- Post added at 14:41 ---------- Previous post was at 14:39 ----------

That's what I said - JB. I said "perhaps from 29 dB attenuation down to 22 dB". Whether that's up a tap or down a tap depends on the box layout!

He's talking about a line distribution architecture where each home's drop is tapped off a line running along the pavement, rather than there being a collection of taps to choose from in a street cabinet. Every so often on the coaxial line running along the pavement there will be a tap serving a few premises with a line extender at some point when power budget requires it.

This is done by network techs, moving a customer's drop from one tap to another in a cabinet is easily doable by service.

[Milambar]

wow how you managing to graph the data? a friend of mine on tbb is trying to setup the same with some scripts.

Thats a good question. With the old Ambit modems, you could query the modem along the lines of

wget http://username:password@192.168.0.1/RgConnect.asp

Then redirect the output to a script, which scraped the data out of the webpage.

I just tried that, and, at least as far as the superhubs go, it no longer works. Its piqued my curiosity, I think I'll poke around a bit.

[Milambar]

Okay, wasted about an hour poking around with this. Its now a two step process.

You need to feed the login script with your username and password, then save the cookie it gives you.

Then you need to call the RgConnect.asp page, feeding it your previously saved cookie and save the resulting page for parsing.

Clunky as hell but it works.

[George_w]

My Superhub doesn't supply you with a cookie. Just keeps u logged on based on your IP address and activity.

[chambohambo]

Just A quick question, as I know I can expect a quicker reponse from public members than on the VM forums, Just looking for feedback really.

So here's the score, As of this morning (8am - My US was 56dBmV - and DS (4 channels) was all between 4-6dBmV depending on the time of day.

VM Engineer come out at 9am, checked the line, put new connecters on and a new splitter outside, but after all that failed and it become 9dBmV and up to 56dBmV. He rang for Central Networks? to come out.

They arrived at 11ish, (Didn't knock on, just seen them outside) - Net went off a couple of times and then notice they cleared off after dinner, leaving my power levels at over 10dBmV (current sitting at 11dBmV) and US at 49.8dBmV.

My question is, how safe are these levels for the superhub?

I'm awaiting a reponse from the admin(s) from the VM Forum for the next step. Just annoyed that I've had today off and really it seems like it was for nothing. Why couldn't this of been sorted the day they put the SH in (2 weeks ago) - The levels have been :s since the thing was put in.

Check the graphs below, as you will see around the times I have said above, you will notice the fluxuations while they was doing work on the cab.

http://d3v.me.uk/pgraph.jpg

How have you done that? I mean record the power levels? is it possible for me to do?

[George_w]

Those levels are fine going back to the original question, so long as it's working it's all cool.

Hey Ignition, long time no speak.. (even though you won't remember me) anyways, My connection seems to keep randomly "dropping" but nothing is showing in the event log, and once again it's a cold night and it's managed to creep up to +12dBmV on the DS channel... hmmm

[Milambar]

For those who may be interested, this is a really simple (and somewhat hacky) perl program that will log in to the superhub, interrogate the stats page, and output the power and noise levels.

The program could easily be modified to store the data in a database for graphing.


#!/usr/bin/perl
#
# (*) Released into the PUBLIC DOMAIN on 07/10/2011.
# (*) Use at your own risk. No garuntees made.
# (*) Note: A bug; The SH also uses a 302 for an invalid login. Need to think of a workaround.

use strict; # Enforce variable scoping.
use LWP::UserAgent; # Module for retrieving webpages from a PSN.
use HTTP::Response; # Module for processing a webpage retrieved by LWP. Not needed, but it simplifies things.
use HTML::TreeBuilder; # Currently unused, but left in so I can investigate its capabilities later.
use Data::Dumper; # Used for debug output, can be removed once the program works.

USER_CONFIGURATIONS:
# Define the superhub login credentials here.
# The defaults will work UNLESS you have changed the superhubs login name/password.
my %credentials = (
'loginUsername' => 'admin',
'loginPassword' => 'changeme'
);

my $superFlub = '192.168.0.1'; # IP address of the router on your network.

################################################## ################################################## ################
################### You shouldn't need to alter anything below this line. ##########################################
################################################## ################################################## ################

INTERNAL_CONFIGURATIONS:
my $loginPage = "/goform/valogin"; # Address of the login handler.
my $statsPage = "/RgConnect.asp"; # Address of the page containing the stats you want.
my $uri = "http://"; # For completeness. Its unlikely to require changing.
my @power; # Global scope for use later.
my @noise; # Global scope for use later.
my $ua = LWP::UserAgent->new; # Create a new instance of LWP::UserAgent

SUPERHUB_LOGIN:
# We now use the LWP object to send a post request to the server. It returns a HTML::Response object.
my $login = $ua->post("${uri}${superFlub}$loginPage", ["loginUsername"=>$credentials{loginUsername}, "loginPassword"=>$credentials{loginPassword}]);

RETRIEVE_STATISTICS:
if ($login->code != 302) {
# Superhub Login failed. Report this to STDOUT and terminate.
print "\nThere was an error logging into the router.\n";
exit;
} else {
# Got a valid login, now retreieve and process the statistics page.
my $sp = $ua->get("${uri}${superFlub}$statsPage");
if ($sp->code != 200) {
# Superhub failed to supply us with the statistics page. Report to STDOUT and terminate.
print "\nRetrieval failure.\n";
exit;
}

# Convert the response object into a HTML content object. This is an implicit call
# to HTML::Response. This is required, as the $sp object is blessed (private object).
my $content = $sp->content;

# This is a really HACKY way of retrieving the numeric values for Power and SNR.
# Must investigate TreeBuilder for a cleaner method. But, despite that, it works.
while ($content =~/(\d+\.\d+)\sdBmV/g) {
push (@power, $1);
}
while ($content =~/(\d+\.\d+)\sdB\</g) {
push (@noise, $1);
}
}

OUTPUT_DATA:
print "Downstream\n";
print "Channel 1\t$power[0]\t$noise[0]\n";
print "Channel 2\t$power[1]\t$noise[1]\n";
print "Channel 3\t$power[2]\t$noise[2]\n";
print "Channel 4\t$power[3]\t$noise[3]\n";

print "Upstream\n";
print "Channel 1\t$power[8]\n";