Print

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

[MaximusPrime]

Hi,

I have a telephone cable that runs under my house directly into our main socket.

During bad weather my connection drops out.

Can you see from my tone graph if there is a problem with it:

cheers

[NewtronStar]

Just for clarification how did you find the pairs (Telephone cable) runs under the house ?

[broadstairs]

I was having an issue like this on my line primarily after rain. I found that using the phone would drop my broadband connection but when it was happening I also had a noisy phone line with background hiss and lots of crackles & pops (not it was not a popular cereal nearby  ) so reported it as a phone fault. The OR guy came and found a high resistance fault which he has repaired and although my broadband has not yet fully recovered from the DLM intervention caused by the fault my line has improved out of all recognition. So when this happens check you phone and if noisy report it.

Stuart

[MaximusPrime]

Just for clarification how did you find the pairs (Telephone cable) runs under the house ?

Openreach Engineers said

So when this happens check you phone and if noisy report it

I had noise on my line but I got that fixed but getting the tones voice is transmitted over fixed hasn't fixed my broadband

[licquorice]

 

[MaximusPrime]

@licquorice,

specifically looking for advice how I can use this graph to get the underground cable replaced with an overhead cable.

Those tones missing in the middle should be there imo?

[WWWombat]

Can you see from my tone graph if there is a problem with it:

On the face of it, no. Nothing stands out alone - save perhaps the gaps around tones 110 and 130. But we can't tell *why* they are not there.

To see static problems, we are likely to need to consider the Hlog, QLN, SNR/tone and bit/tone graphs all together - alongside "simple" information about the line from the "telnet data", such as the attenuation.

To see dynamic problems (such as a line dropping in bad weather), we are likely to need to consider graphs over time - particularly the attenuation, SNRM, sync speed, attainable speed graphs, plus the error-tracking graphs (FEC, CRC, ES, SES and retransmission graphs if they apply), plus the DLM-affected data (INP, delay etc from the "telnet data").

How long is the line? The distance from the cabinet, that is.

Those tones missing in the middle should be there imo?

It depends why they are missing.

The gap between 160 and 280 is likely to be because the cabinet is transmitting at reduced power to prevent interference with ADSL from the exchange. We expect a "notch" in both SNR/tone and bits/tone at *some* point in the spectrum (I've attached mine for comparison) - yours looks consistent with a cabinet that is about 3km from the exchange.

The small gaps at 110 and 130 are different. More data is perhaps needed there.

[MaximusPrime]

@WWWombat,

I'm 2.8km from my cabinet. I usually connect at 8Mb with a Billion 8800NL.

Yesterday I parked my car to wash it in a different place where the cable runs under & I swear it disconnected my internet & I'm down to 4Mb now.

I attached the extra graphs

[burakkucat]

Your circuit's Hlog plot shows that only the US0 and DS1 bands are in use. From the uppermost sub-carrier (tone) in operation, we can see that the circuit is just managing to be better than a good quality ADSL2+ service!

The QLN plot is very much as I was expecting to see. Not brilliant but neither outrageously poor.

The SNRM plot, with values of around 15 dB DS and 13 dB US, show that there is certainly more that the circuit could provide. 

[MaximusPrime]

@burakkucat,

what does HLog mean? Should it be better? If so how?

QLN means Quiet Line Noise but why are the y-axis values negative? The lower the number the more noise there is?

Usually my noise margin is between 6 & 7db

[burakkucat]

The units for both the Hlog (the transfer function of the circuit) and the QLN plots are based on decibel ratios (specifically dB and dB/Hz, respectively). Hence a negative value is an expression of the fact that numerator is smaller than the denominator. Both should really be shown as quadrant four plots . . . i.e. the zero point of the Y-axis at the top of the graph's frame.

The shape of your Hlog plot is good, you should expect to see a smooth, gradual decline with respect to frequency. (As the units are negative decibel values, you should expect to see larger negative values with increasing frequency.) A good Hlog plot would show the US0, DS1, US1, DS2, US2 & DS3 bands in use. (With green for US and red for DS you would see the characteristic green, red, green, red, green and red colouration across a good Hlog plot.)

[MaximusPrime]

@burakkucat,

when you say HLog is the transfer function what do you mean?

So the graphs are fine?

Any suggestions what data to record to get my line changed when the bad weather starts again?

I can't sit through another Winter with my broadband disconnecting for several days at a time

[burakkucat]

when you say HLog is the transfer function what do you mean?

It is a complex function, involving both real and imaginary numbers. (The equivalent Hlin function actually shows both real and imaginary values per sub-carrier.) In the broadest sense, it is often useful to consider the Hlog plot to show the equivalence of attenuation over frequency.

So the graphs are fine?

For what little we can see, there is no evidence of a "smoking gun" . . .

Any suggestions what data to record to get my line changed when the bad weather starts again?

Just as WWWombat has mentioned, above.

I can't sit through another Winter with my broadband disconnecting for several days at a time

Perhaps consider having a second line installed and once that has been actioned, then have the service on the first line terminated. 

[MaximusPrime]

@burakkucat,

I have considered getting a second line in but would Openreach try to use the single pair into my home to service both sockets?

We used to have a second line in our home during dial up internet days & they used the underground line into our house to service both sockets, there's only 1 pair in the underground cable

[Black Sheep]

We never used single-pair cable on underground fed homes, back in the day. I can't think of any single-pair underground cable in use, even today

I'm guessing you live on a 1960/70's estate, whereby the preferred method back then was to use DIG (Direct-in-ground) armoured cable. The premise was to lay the cable in the ground and the house was then built on top of it.
These cables were always 2-pair in nature ....... if you can find where the entry box is, you should see Pair 1 = Orange/White wires ..... Pair 2 = Green/Black wires.

Yes, we would definitely use the spare pair to provision the 2nd line over, if it was fault free.

Experience tells me that we do have issues with these old cables, especially with 'Earth contact' faults ...... ie: the insulation has broken down, along with the outer protective sheathing, and is in contact with the ground around it. However, for every cable we find faulty, there will be one that has stood the test of time. Only an engineer running tests could prove one way or another.