Kitz Forum
Broadband Related => Broadband Technology => Topic started by: burakkucat on October 02, 2014, 06:00:03 PM
-
Earlier today a message arrived in the e-mail in-box at The Cattery. Attached to it were a pair of Hlog graphs, which I've attached below. The bulk of the message reads as follows --
I invite you to compare the Hlog plots (yes they are both from the same physical line) from 22:00 last night and 08:30 this morning. Note the pretty standing wave effect.
The latter was taken from the test socket as I awaited the arrival of an SFI to invitigate an intermittent (but otherwise bloody obvious) (physical connection) fault which started generating audible noise on voice on 12th August, and since then has been responsible for something like 70 LOS and in excess of 600 LOM (down to an unbelievable -8dB more than once) since then as SNRm deteriorated.
The good news is that the OR SFI arrived, traced it to a failing wire at the BT where the drop cable comes in - and which fell off as soon as he touched it. He replaced it. Attenuation dropped and SNRm jumped up 3dB immediately. And since it was my birthday (or at least that's what it felt like) he reset the profile to open, and now I'm back to 80/20 fastpath no INP.
As I wrote for the subject line of this thread, we should use caution when interpreting a Hlog graph. Certainly I would not have expected to see such a plot as a result of a simple failing connecting at a block terminal!
-
Thanks for the Hlog comparison graphs, the only reason why the wire would just fall of during the OR inspection would be down to rust, bare metal wires hates moisture, so my first thought is how the moisture is getting in from the drop wires and into the block terminal.
Or it was badly terminated in the first place.
-
Wow... that is something :D Thank you for sharing.
.. and talking of wires that disintegrated as soon as the engineer poked it with his screwdriver, heres mine in case anyone is interested.
Graph 1 - with corroded terminals in the BT66
Graph 2 - terminals rewired and terminal block replaced. Dip at circa tone 60 gone and U2 now smooth.
Still had HR issues with the line though and upstream SNRm would drop drastically if the phone rang - note the squiggles in U1 band
Graph 3 - Fault fixed - new pair.
-
interesting, check my hlog, it seems to have dips although not too severe but are there.
-
interesting, check my hlog, it seems to have dips although not too severe but are there.
:hmm: Hmm . . . apart from that "untidiness" in your U1 band the overall plot seems to be quite acceptable.
-
Burakkucat has said (in the other thread) that HR faults are not noticable in Hlog graphs.
-
interesting, check my hlog, it seems to have dips although not too severe but are there.
:hmm: Hmm . . . apart from that "untidiness" in your U1 band the overall plot seems to be quite acceptable.
I am confused it has similarities to the one you posted.
-
Now I am confused. ???
How is Chrysalis_Hlog.png (http://forum.kitz.co.uk/index.php?action=dlattach;topic=14507.0;attach=14464) (almost purrfect) similar to Hlog-20141002-0830.png (http://forum.kitz.co.uk/index.php?action=dlattach;topic=14507.0;attach=14440) (clearly defective)? :-\
-
Found this interesting Web Site it has all those old and new internal block terminals.
http://www.britishtelephones.com/bterm.htm (http://www.britishtelephones.com/bterm.htm)
-
Found this interesting Web Site it has all those old and new internal block terminals.
http://www.britishtelephones.com/bterm.htm (http://www.britishtelephones.com/bterm.htm)
Yes, indeed. And as a companion for that site, I would recommend Telephones UK (http://www.telephonesuk.co.uk/). It will be something very odd or rare that is not mentioned on either of those two sites. :)
-
I did prefer the BT80 RF3 screw terminals as it was a lot easier to terminate than the BT80A as it needed more wire stripping and it was kind of fiddly to get the curl right before screwing down.
-
Ah yes. The "chocolate block" style termination as distinct to a screw and washer combination. ;)
-
Now I am confused. ???
How is Chrysalis_Hlog.png (http://forum.kitz.co.uk/index.php?action=dlattach;topic=14507.0;attach=14464) (almost purrfect) similar to Hlog-20141002-0830.png (http://forum.kitz.co.uk/index.php?action=dlattach;topic=14507.0;attach=14440) (clearly defective)? :-\
I have some dips, not as severe but there. on the d3 and u3.
-
Hmm . . . A subtle, little wave. In my opinion nothing to be concerned about.
Take a look at Kitz' third graph (http://forum.kitz.co.uk/index.php?action=dlattach;topic=14507.0;attach=14454), shown above. That is more like your Hlog graph (http://forum.kitz.co.uk/index.php?action=dlattach;topic=14507.0;attach=14464).
-
sorry I meant my graph is similar to kitz's first, with the dip.
-
Wow... that is something :D Thank you for sharing.
.. and talking of wires that disintegrated as soon as the engineer poked it with his screwdriver, heres mine in case anyone is interested.
Graph 1 - with corroded terminals in the BT66
Graph 2 - terminals rewired and terminal block replaced. Dip at circa tone 60 gone and U2 now smooth.
Still had HR issues with the line though and upstream SNRm would drop drastically if the phone rang - note the squiggles in U1 band
Graph 3 - Fault fixed - new pair.
Could you attach Hlog text datas?
Best regards
konrado5
-
Which ones.. and what would they show thats not in the graph? What are you looking for? VDSL has a heck of a lot more tones and use different bands than adsl. The plink files are 400KB each and more than the forum attachment limitation.
-
All ones. I want to do graphs first 511 tones (only ADSL2+) and compare with my graph.