Print

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

[kitz]

No thats just the effect of power cut back to help prevent crosstalk for the longer lines.   Those on lines near the exchange will have the most PCB in their U0 and U1 bands.   For example mine are -6, and -26.7dBm

Mine is -18 and -32
Oh so it means because mine line is short it wont need as much power is that right thats why its -  ?

Yep   You can still get the maximum speeds without need for more power.  It they let your line run at full power, then it would drown out the signal for longer lines.

[aparat]

Thanks Kitz, Bald Eagle, and others you are very helpful thanks a lot. If Ill need more answers  I know where to ask.

Thank you

[aparat]

Oh BTW can someone explain Hlog for me? What Im looking at on my Hlog, bacause I seen others Hlogs and they usualy going down much further tthen mine thanks

[kitz]

You cant really explain hlog... or any of the graphs come to that.   

Interpreting line stats from graphs is something that comes over time and experience.   Ive been 'reading' line stats for about 10yrs now and sometimes you just have to go on gut instinct and you get 'a feel' for the line.   
No-one really likes saying it should be this or it should be that.   Because then you will get someone else come along and think they see a similar pattern on their own graph.. and then that person goes into panic mode thinking there is something wrong with their line, when in all reality it is within normal & acceptable line parameters for their line. 

hlog and QLN are relative newcomers into the the world of graphing, and even those of us who have been reading line stats for years are still learning.  The main thing about hlog is that it should be reasonably smooth. 

The fact that yours doesnt go down as far as some other graphs is because your line is short and you have excellent attenuation.   As (copper) line get longer then the more the hlog figures will go down.  Its not unusual to see lines begin to drop quite rapidly when you get into the higher frequencies. For example my own hlog drops very rapidly after tone 3800 - you'll see that tail end curve on quite a lot of lines.

[Bald_Eagle1]

As mentioned by Kitz, a smooth Hlog curve demonstrates a physically O.K. line.
Short lines have closer to horizontal Hlog graphs whereas longer lines have fairly steep downward graphs.

I have attached snapshot montages from a longish connection where a bridged tap fault was firstly identified & then resolved.

A bridged tap is caused by a 'branch' off the main line, often caused by star wiring inside a user's premises, but occasionally seen from alarm companies branching the line to their kit before the faceplate filtered connections.


The examples clearly show the negative effect of the bridged tap (the 'V' shape in the Hlog graph) & if the line speed hadn't already been capped by DLM at 19998Kbps DS sync speed, removal of the bridged would have allowed even higher sync speed.

[aparat]

Thanks a lot guys I understand it now  Thanks for explaining it to me. If Ill need more answers Ill be back

[aparat]

Something new have a look at US SNR, over half day at 18 and then at 24 what could it be. I noticed it about 3 days ago

Stats at MDWS: aparatos1

[NewtronStar]

Something new have a look at US SNR, over half day at 18 and then at 24 what could it be. I noticed it about 3 days ago

Stats at MDWS: aparatos1

It could be, have noticed my US SNRM being lower and the US sync rate has gone down by 800 kbps over the last 7 months so crosstalk can hit your upstream.

[kitz]

If you look at your Sync attainable, you will see it follows exactly the same pattern.

The spike at 2am is typical of what you'd see when a crosstalk disturber resyncs their line.   If you look at my SNRm graph for the past 10 days, you can see that my line frequently does the same thing.    Whenever their router is switched off, then your stats are better because they arent introducing crosstalk on to your line.

-
edit NS just beat me to it

[aparat]

Thanks for reply  That must be new connection because US SNR was never affected until 3 days ago . Oh well nothing I can do anyway. Ill keep eye on it .......