Print

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

[Loading]

why does the bit on the end change? what does it mean?

[kitz]

to be honest Im not sure.   Ive seen it on mine on the odd occasion and not thought anything further about it.
I also see lines line that at the end of my QLN, so thought it may have something to do with band masks.

[Loading]

also there are lots of noise spikes on my qln

[kitz]

also there are lots of noise spikes on my qln

That doesnt look too healthly    Small spikes and slight jitters would be considered normal, but theres some quite steep spikes.
QLN data can indicate cross-talk from a neighbouring line :/   

[Loading]

high no of errors, attenuation increase and induced noise from a tie pair mod?, unless on a faulty port  bit odd

[kitz]

high no of errors, attenuation increase and induced noise from a tie pair mod?, unless on a faulty port  bit odd

Can be caused lots of other causes aside from the dslam port.  Could you be on a different pair which is why things have changed?

[Loading]

No as apparently there were no available e-side pairs and sometimes I can hear clicking and beeps in the background

[kitz]

They usually replace d-side before the e-side..  was just checking it was ONLY the tie pair that was swapped.  Do you know this for sure? 

Im not sure why they did a tie pair mod and Im now trying to track down your other posts to find out whats gone on without me swapping back and forward between windows.

[kitz]

Ok im still not clear despite scanning your other posts.

The noise only occurred after the tie-pair mod yes?
The first visit was for slow speeds...  yet now they are even slower and your attenuation is worse?

When looking at your stats in your last posts one of the possibles that I suggested was a HR fault.

Ive now found another of your posts which I previously hadnt seen... in which Blacksheep last month thought it was definitely a small HR fault.
Hes responded to your 1st thread.  Take note what he says.  Blacksheep knows his stuff when it comes to things like this
http://forum.kitz.co.uk/index.php/topic,13002.msg245477.html#msg245477

[Loading]

I take on board what you say, but there is sometimes a clicking noise, and beeps when nothing is connected, that is the only reason why I can't believe it to be a hr fault, and it the noise margin never changes depending on the weather? Yea the noise is only after a tie pair mod, the first visit was the noise margin just rising in jumps of 3db, even when there wasn't a lot of dropouts.

[kitz]

I think you need to pester your ISP again to get another visit.   That audio file is useful evidence that something is wrong.
HR faults unfortunately arent the easiest to track down because the symptoms can be so varied.  :/

Im not saying its not a port issue, but the tie pair is relatively short when compared to the d-side which can be exposed to the elements and more likely to suffer from faults.  It may be that the Tie pair mod has just triggered off something.   I know when I had a HR fault, just the engineer testing from the cab made my line much worse for a couple of weeks afterwards until they actually rectified it properly.  Theres no reason why the testing should have made it worse, but it did by increasing my atten and SNR slightly.  Unfortunately though BT didnt see this increase in atten as a fault & it was only a week or so later when voice started to fail and become unusable that they took note.

[burakkucat]

. . .  there is sometimes a clicking noise, and beeps when nothing is connected, . . .

The above facts now suggest, to me, that there is the possibility of a 'battery contact fault' on your pair.

Initially, in your other thread, I speculated that there might be an 'earth contact fault' which would account to the 50 Hz 'mains hum' and the slower than normal broadband connection.

As I mentioned previously, without any physical access to the copper pair then remote diagnosis is not really accurate.

[Loading]

after testing the line with his tester he said that the line was perfect, even the attenuation decreased slightly, so somehow the testing helped slightly?, also could the battery contact and earth be in the exchange? and the line is mostly the same in all weathers little or no variation, also how could it be both earth and battery?. there are also quite a significant number of errors on the downstream. when bearing in mind before the tie pair mod there were only 10-20 hec errors and no ocd/lcd errors

    Downstream   Upstream
Line Coding(Trellis):   On   On
SNR Margin (0.1 dB):   153   130
Attenuation (0.1 dB):   635   315
Output Power (0.1 dBm):   146   123
Attainable Rate (Kbps):   576   756
 
    Path 0       Path 1   
    Downstream   Upstream   Downstream   Upstream
Rate (Kbps):   480   448   0   0
 
K (number of bytes in DMT frame):   16   15   0   0
R (number of check bytes in RS code word):   8   4   0   0
S (RS code word size in DMT frame):   4.00   4.00   0.0   0.0
D (interleaver depth):   8   8   0   0
Delay (msec):   8.00   8.00   0.0   0.0
INP (DMT symbol):   0.44   0.21   0.0   0.0
 
Super Frames:   4562528   4562528   0   0
Super Frame Errors:   334   21   0   0
RS Words:   77562968   77563231   0   0
RS Correctable Errors:   15567   129   0   0
RS Uncorrectable Errors:   1226   0   0   0
 
HEC Errors:   372   11   0   0
OCD Errors:   11   0   0   0
LCD Errors:   11   0   0   0
Total Cells:   87806690   0   0   0
Data Cells:   3258497   0   0   0
Bit Errors:   0   0   0   0
 
Total ES:   241   0         
Total SES:   0   0        
Total UAS:   9   9         
   
also added the new hlog and qln graphs?

[burakkucat]

. . . also could the battery contact and earth be in the exchange? . . .

Yes, it could be. When a 'frames' engineer was working nearby, a small off-cut of jumper wiring could have dropped down and partially shorted one leg of one circuit to one leg of another.

. . . also how could it be both earth and battery? . . .

It is highly unlikely to be both. Take it as an example of how fraught with error remote fault-finding by description (without access to the pair) can be. What I heard in your recording could have been due to an 'earth contact fault'. What you have just described, above, could be due to a 'battery contact fault'.

Only by showing evidence to your CP/ISP will you be able to convince them to book an Openreach engineering appointment. That evidence should then be shown to the engineer.

also added the new hlog and qln graphs?

For comparison purposes, you can see the Hlog and QLN graphs for my line here.

[Loading]

What does your connection stats look like, as you have spikes in your qln graph, I only want a decent connection around 2mb as it was before!