Line Attenuation question.

Started by Bat, Jul 10, 2007, 17:00:30

Previous topic - Next topic

0 Members and 2 Guests are viewing this topic.

Bat

Hi,
If I live 1.6 miles from the exchange what should I expect my line attenuation to be, or doesn't it work like that?
Cheers,
Gavin :)

Rik

It doesn't work like that, your line could arrive by a very circuitous route, or have some iffy joints on it, or be carried on aluminium rather than copper etc... Now, if you find out the attenuation, we can give a good idea of the speed you should achieve. :)
Rik
--------------------

This post reflects my own views, opinions and experience, not those of IDNet.

Bat

Hi,
Between 44 and 39 dB. Seems to make up it's own mind!
Currently 39dB.
I'm 1.6 driving miles from the exchange, builtup area if it makes any difference?
Cheers,
Gavin :)

ReDGryphoN

Hey Rik,

I bet you or I would l love that attenuation !!

Drool drool

ReD

MAABOF
BILLION 8800NL USER FTTP

If a man does not keep pace with his companions, perhaps it is because he hears a different drummer. Let him step to the music which he hears, however measured or far away.
Henry David Thoreau

Lance

Quote from: Bat on Jul 10, 2007, 20:13:36
Hi,
Between 44 and 39 dB. Seems to make up it's own mind!
Currently 39dB.
I'm 1.6 driving miles from the exchange, builtup area if it makes any difference?
Cheers,
Gavin :)

I'll leave Rik to make the guess on the attenuation. (let me know how you find/work it out Rik!)

Although you are 1.6 driving miles from the exchange, the route the BT cable could be much longer. I remember reading somewhere that you work out the length if the cable by dividing your attenuation by 13.81 (as the copper cables used by BT lose 13.81db of signal for every mile - or something like that!). I'm not sure how accurate it is though.

An example of the route a cable could take is: imagine a cul-de-sac. You live on one side at the entrance to it so you might think you are closer to the exchange then someone who lives right at the end. You could easily be wrong though, as your house could be the last one on the street the cable goes to, ie it starts over the road, goes all the way down on one side, and back up on your side. From this, you can see although you might be the same distance as the person over the road from the exchange, the cable distance is much further. I hope this makes sense to you!

Quote from: ReDGryphoN on Jul 10, 2007, 21:53:30
Hey Rik,

I bet you or I would l love that attenuation !!

Drool drool

ReD

MAABOF

Not forgetting me with my attenuation of 55db!
Lance
_____

This post reflects my own views, opinions and experience, not those of IDNet.

DorsetBoy

These are the stats Rik wants........


Bandwidth (Up/Down) [kbps/kbps]:   448 / 8,128

Data Transferred (Sent/Received) [KB/KB]:   0.00 / 2.00

Output Power (Up/Down) [dBm]:   11.5 / 19.5

Line Attenuation (Up/Down) [dB]:   13.5 / 25.0

SN Margin (Up/Down) [dB]:   21.0 / 9.0

Vendor ID (Local/Remote):   TMMB / TSTC

Bat

Hi,
I know you lot like numbers :D

16/11/06 shortly after joining IDNet

25/12/06

05/07/07

I imagine these were done if something wasn't right at the time!  ::)
Cheers,
Gavin :)

Lance

From looking at these screenshots, it is clear you have a fault somewhere on your line.

As the screenshots get later, you can see the noise margin increases and sync decreases. The noise margin is set at a default 6db by BT, but if their equipment senses a repeated fault, it will increase automatically in 3db increments.

I would suggest having a read through the FAQ, especially the part on internal wiring, and trying things such as swapping filters.

If you have a NTE5 type socket, plug the router into the test socket and then post the stats from the table in the screenshot (copy/paste will be fine). If the stats improve, then the fault is with your internal wiring.

Hope this helps.

Lance
Lance
_____

This post reflects my own views, opinions and experience, not those of IDNet.

Bat

Hi,
I've already tried the router in the test socket, as you mentioned, stats unchanged :(
Wel I think I got a 0.5 dB improvement on Att. but then it's usually differtent each time anyway.
I've tried another router and also upgraded the firmware on my router, which is why the last one looks different! But again, no improvement!
Cheers,
Gavin :)

Rik

All the signs indicate a wiring issue, Gavin. Although attenuation does change slightly with sync speed, you have had a significant change. This may be as a result of work done by BT at the exchange, or perhaps as a result of changes you have made to your internal wiring. What does not make sense is that, as attenuation has dropped, so has your sync speed. In part, this would appear to be because your noise margin has increased, probably to a 15db target, but your connection appears to be going in two opposite directions.

For this reason, I would suggest you contact support, quoting them your three sets of d/s attenuation, sync speed and noise margin figures, and ask them to test the line.
Rik
--------------------

This post reflects my own views, opinions and experience, not those of IDNet.

Bat

Hi,
Here's what I've got from support when I got home....

On testing the line there does appear to be a speed issue, however it can be routed to the connection stability.

The connection has been up and down 26 times in total today - because the line is rate adaptive the speed has come down to try and compensate. The line test shows that these issues are not as a result of any known BT issue, and all tests are showing a healthy line. For this reason, further tests will be needed to narrow down the issue and resolve it. Drops in connection are commonly caused by peaks of noise on the line. To minimise noise you need to simplify the connection as much as possible. The best way to do this is have your modem/router connected direct to the master socket with a brand new filter. If the connection still continues to drop at this point then you can be sure it is unrelated to your equipment and wiring. These tests are needed to ensure that if we need to send a bt engineer, the chance of you being charged for the engineer identifying the problem on your side is minimal.


Cheers,
Gavin :)

Rik

Quote from: Bat on Jul 11, 2007, 18:20:24
These tests are needed to ensure that if we need to send a bt engineer, the chance of you being charged for the engineer identifying the problem on your side is minimal.[/i]

Sounds reasonable, Gavin. If you're getting that many disconnections, something is wrong - it's now a case of establishing where. As support say, the last thing you want is for a BT engineer to turn up, find the issue is in your wiring and raise a £150+ bill. If you can, as far as possible, eliminate your own wiring, anything after the master socket, then you should be spared any unpleasant surprises. The only way to do that will be to use the test socket behind the faceplate of the master for a minimum of 24 hours. Does your router log disconnections? If not, advise support when you plan to start the test and then check with them the following day.
Rik
--------------------

This post reflects my own views, opinions and experience, not those of IDNet.

Bat

#12
 Hi,
That was my plan  ;) :D
It doesn't log disconnections, so I was unaware of what was occuring  :o
I'll move the router and let support know.
Cheers,
Gavin :)