BT BRAS Profile

[davej99]

Ok Dave, thanks for all the info - I will let you all know how I get on, which should be Monday evening sometime 

Thx for Karma/

There are pics of the old style non-split faceplaces here and the famous capacitors. One is primary and one secondary. Then the split plate already mentioned.

[T_M_D]

Thx for Karma/

There are pics of the old style non-split faceplaces here and the famous capacitors. One is primary and one secondary. Then the split plate already mentioned.

Brilliant - thanks. I am finding it all quite fascinating, maybe I should have been a BT engineer... 

[Rik]

New career moved, Tina?

[Tacitus]

New career moved, Tina?

You could do worse!  Take a look here.  A friend was quoted £100 plus £30 per extension for moving the main socket, a new drop wire (both very naughty since it's BTs property!) and, in his case, one extension.  The chap who did it was ex-BT and apparently did a first class job - even to the extent of checking the polarity on the main socket.  The external cabling was proper 6-core (4 core would have sufficed) external quality with polyethylene sheathing, unlike some cowboys who use black internal cabling and claim it to be external spec. 

Given the prices BT charge for what in some cases is 20 minutes work, I imagine a lot of this goes on.  The only reason my friend got someone in to do the work was the need for a new dropwire, otherwise it's well within the reach of anyone with a bit of common sense. 

[davej99]

THIS THREAD & POST may be helpful to some dealing with an old primary socket. BT do not look at all good. See the thread from the top.

THIS GUIDE shows how to do various things with house phone wiring to get better broadband performance, including fitting your own NTE5 filtered faceplate legally if there is no NTE5 at all. It's step by step DIY.

[T_M_D]

New career moved, Tina?

Hmmm, think I will pass on that one Rik 

[T_M_D]

You could do worse!  Take a look here.  A friend was quoted £100 plus £30 per extension for moving the main socket, a new drop wire (both very naughty since it's BTs property!) and, in his case, one extension.  The chap who did it was ex-BT and apparently did a first class job - even to the extent of checking the polarity on the main socket.  The external cabling was proper 6-core (4 core would have sufficed) external quality with polyethylene sheathing, unlike some cowboys who use black internal cabling and claim it to be external spec. 

Given the prices BT charge for what in some cases is 20 minutes work, I imagine a lot of this goes on.  The only reason my friend got someone in to do the work was the need for a new dropwire, otherwise it's well within the reach of anyone with a bit of common sense. 

Many of these big corporates are all the same, look at the energy companies and how they rip us off. They have us all over a barrel because we need them. 

[T_M_D]

THIS THREAD & POST may be helpful to some dealing with an old primary socket. BT do not look at all good. See the thread from the top.

THIS GUIDE shows how to do various things with house phone wiring to get better broadband performance, including fitting your own NTE5 filtered faceplate legally if there is no NTE5 at all. It's step by step DIY.

This whole post has certainly been incredibly useful to me and it will continue to be so as I will be keeping all the great and informative info in it for future reference.  Tina.

[T_M_D]

It has taken me some good time to get back to updagte this post, but here I am at last!

Here are the stats for before I tested from the master socket (behind split face plate):
Connection time 0 days, 00:58:46
Data Transmitted/Received (MB) 0.8 / 7.1
Broadband username bthomehub@btbroadband.com

ADSL line status
Connection Information
Line state Connected
Connection time 0 days, 01:02:10
Downstream 1,504 Kbps
Upstream 448 Kbps

 
ADSL Settings
VPI/VCI 0/38
Type PPPoA
Modulation G.992.1 Annex A
Latency type Interleaved
Noise margin (Down/Up) 11.5 dB / 18.0 dB
Line attenuation (Down/Up) 56.8 dB / 31.5 dB
Output power (Down/Up) 2.5 dBm / 1.4 dBm
Loss of Framing (Local/Remote) 0 / 0
Loss of Signal (Local/Remote) 0 / 0
Loss of Power (Local/Remote) 0 / 0
FEC Errors (Down/Up) 64 / 7
CRC Errors (Down/Up) 58 / 18
HEC Errors (Down/Up) 74 / 8
Error Seconds (Local/Remote) 44 / 6


Password Not configured

 
TCP/IP settings
Broadband network IP address XX.XX.XX.XX
Default gateway 217.41.221.146
Primary DNS 194.74.65.68
Secondary DNS 194.72.9.34

Here are the stats from the master socket:
ADSL line status
Connection Information
Line state Connected
Connection time 0 days, 00:13:12
Downstream 1,408 Kbps
Upstream 448 Kbps

 
ADSL Settings
VPI/VCI 0/38
Type PPPoA
Modulation G.992.1 Annex A
Latency type Interleaved
Noise margin (Down/Up) 11.9 dB / 19.0 dB
Line attenuation (Down/Up) 56.5 dB / 31.5 dB
Output power (Down/Up) 2.4 dBm / 1.4 dBm
Loss of Framing (Local/Remote) 0 / 0
Loss of Signal (Local/Remote) 0 / 0
Loss of Power (Local/Remote) 0 / 0
FEC Errors (Down/Up) 1 / 0
CRC Errors (Down/Up) 2 / 5
HEC Errors (Down/Up) 1 / 0
Error Seconds (Local/Remote) 2 / 1

I'm not quite sure how to analyse what the differences between them are? I don't know if the difference in the error stats between the two are significant? Any help with this would be appreciated.

I did a line noise test by calling the number and listening to the noise on the non-dect phone from the socket before removing the split face plate and then again after removing the split face plate and plugging the phone into the master socket. The noise was about twice as loud when tested in the master socket. Again, I don't know if this is significant or not?

I did remove all phones and extenstion leads from telephone sockets before doing all the tests, but unfortunately, I had forgotten to test that each extension was 'dead' after I removed the face plate from the master socket - does this make my testing completely redundant?

The external cabling into the house from the grey junction box is as follows: a very thin grey round cable (around a quarter of the width of mine, which is about the thickness of a laptop power cable) comes out of the junction box and along approx 12M of wall space before it enters the house through the wall into the downstairs cloak room. From that point inside the house, up to the master socket, the cable is approx 4M in length.

Cable comes out of the master socket in the hallway and after a few metres, links into another telephone wall socket in the hallway, cable comes out the other side of this non-master socket in the hallway and goes along the wall for around 5M before it goes through the wall into the dining room and feeds the double socket that the Hub is connected to.

Cable also goes out of the back of the non-master socket in the hallway into the lounge and travels around 7-8M before going into the wall socket in the lounge. Cable comes out the back of this socket and through the lounge wall to the outside of the house then travels around 8-9M along the wall before feeding through the wall into the bedroom wall socket.

The hub is plugged into one of the ports in the double socket in the dining room via a filter, an extension cable that is very long, with much of it in a heap, is also plugged into the filter and a phone is then plugged into the end of that long extension cable. One of the first things I did when trying to trouble-shoot was unplug the extension cable so that only the hub was plugged into the wall socket via the filter. It made no difference however, to the router stats

I plugged a telephone into the other socket in the double socket and it worked from that socket proving it was live, however, when dialling out from that phone in that socket, it made the hub (plugged into the other socket of the twin socket) lose connection and some of its lights flash from blue to orange - I don't know what that indicates about the double socket, if anything?

I did all this testing around a week and a half ago and at about the same time, I became aware that BT Open Reach were in the locale digging up a couple of sections of road and pavement, one section they dug up was just a few metres from the house and I noticed that they came and lifted up a BT cover right outside the front of the house and were putting something down it - I didn't have the chance to go and query what they were doing unfortunately.

I did a BT Speed test after I tested in the master socket and was surprised to find that the Max Achievealble speed had gone up from 200-1000kbps to 400-2000kbps and the IP profile had gone from 1000kbps to 1250Kbs, though the DSL connection rate had gone down from 1472kbps to 1408kbps. The download speed achieved during the speed test from the master socket was 1130kbps as compared to 923kbps when the test was run without using the master socket. Is it the case that a BT speed test from the master socket will always yield superior results to one performed from a non-master socket?

I am not sure what all of the above really tells me, but I ran a speedtest today (not from the master socket), the first since doing all the above testing and I only had time to use speedtest.net. The download speed has risen from what was a consistent 900+kbps to 1.3mbps. I will of course continue to monitor it using the BT speed test whenever possible. I still think the speed is pretty poor and that she should be getting between 3-4mbps, but I don't really know where to go with it from here....anyone any ideas? Tina.


Edit: IP address removed

[Rik]

Connection time 0 days, 00:58:46

Here are the stats from the master socket:

Connection time 0 days, 00:13:12

The difference in connection time makes it hard to do any real comparison, Tina, but I'd say the change in sync speed is just down to the ambient noise at the time you connected, while the figures from the master socket suggest a lower error rate than from the normal configuration. To have any real validity, though, you need to be connected each way for about 24 hours.

I am puzzled by the IP address you show, though (now removed), it isn't an IDNet one, nor the one you're using here.

I did a line noise test by calling the number and listening to the noise on the non-dect phone from the socket before removing the split face plate and then again after removing the split face plate and plugging the phone into the master socket. The noise was about twice as loud when tested in the master socket. Again, I don't know if this is significant or not?

The master socket is the face plate, the test socket is the one you reveal behind it. It should, if anything, be quieter. I'm puzzled on that.

I did remove all phones and extenstion leads from telephone sockets before doing all the tests, but unfortunately, I had forgotten to test that each extension was 'dead' after I removed the face plate from the master socket - does this make my testing completely redundant?

No, but it leaves the chance of a wiring issue, so ideally you need to test that.

The hub is plugged into one of the ports in the double socket in the dining room via a filter, an extension cable that is very long, with much of it in a heap, is also plugged into the filter and a phone is then plugged into the end of that long extension cable. One of the first things I did when trying to trouble-shoot was unplug the extension cable so that only the hub was plugged into the wall socket via the filter. It made no difference however, to the router stats.

I plugged a telephone into the other socket in the double socket and it worked from that socket proving it was live, however, when dialling out from that phone in that socket, it made the hub (plugged into the other socket of the twin socket) lose connection and some of its lights flash from blue to orange - I don't know what that indicates about the double socket, if anything?

But that same phone, plugged into the same filter as the router, is OK?

I did a BT Speed test after I tested in the master socket and was surprised to find that the Max Achievealble speed had gone up from 200-1000kbps to 400-2000kbps and the IP profile had gone from 1000kbps to 1250Kbs, though the DSL connection rate had gone down from 1472kbps to 1408kbps. The download speed achieved during the speed test from the master socket was 1130kbps as compared to 923kbps when the test was run without using the master socket. Is it the case that a BT speed test from the master socket will always yield superior results to one performed from a non-master socket?

Your profile has risen to match your sync speed, suggesting flapping. If the wiring is OK, and all other variables don't change, then the difference in throughput between the test socket and any other should be minimal. You need to repeat that test several times throughout the day to establish whether it was a one-off result.

I am not sure what all of the above really tells me, but I ran a speedtest today (not from the master socket), the first since doing all the above testing and I only had time to use speedtest.net. The download speed has risen from what was a consistent 900+kbps to 1.3mbps. I will of course continue to monitor it using the BT speed test whenever possible. I still think the speed is pretty poor and that she should be getting between 3-4mbps, but I don't really know where to go with it from here....anyone any ideas? Tina.

You need to establish that all other sockets are dead when you are using the test socket, then you need to run the router in the test socket for 24 hours, doing a number of speedtests across that period (ideally BT, but if not, then speedtest.net using the Maidenhead server). Repeat with the wiring in its normal state and let us have the results for both.

[Steve]

I am puzzled by the IP address you show, though (now removed), it isn't an IDNet one, nor the one you're using here.

Its her friend's BT broadband connection,Rik

[Rik]

This thread's been running too long. 

[T_M_D]

This thread's been running too long. 

Steve/Rik,

thanks for all your work in sticking with this and your comprehensive answers to my lengthy text Rik.

But that same phone, plugged into the same filter as the router, is OK?

Yes, the phone does work in the same filter as the router - I have had to put it back in there so it will be interesting to see if that makes the speed go back down below 1mb.

I don't think my friend and her husband would want all extenstions unplugged and the laptop and hub connected to the test socket in the hallway for 24 hours somehow or me hanging around doing numerous speed tests throughout that period (would be difficult for me to find the time to do it anyway). Why is it such a convoluted process to ascertain whether the problem is with the internal wiring or the external wiring? Surely most people don't do this level of troubleshooting to establish where the fault lies? I suppose many people would just call out a BT engineer to check it all out and then pay the charge if BT prove the problem isn't with the external wiring? Is it worth me trying to get BT to run a line fault test do you think? No wonder some people just put up with internet speed problems. Tina.

[Rik]

You can try, Tina, but BT are renowned for the best customer service. Ultimately, the 24 hour test is the only protection your friend has again a £160 minimum BT charge if an engineer is called out and finds no fault on BT's side of that test socket. One thing you might consider is fitting a filtered faceplate instead of that dual one. It will isolate all the extensions from the ADSL service, but would require the router to be close to the master socket.

[Steve]

Have we checked the ring wire can't recall?

[Rik]

I would have thought so, Steve, but like you, I can't remember...

[Rik]

We don't appear to. Tina, you need to check every socket and remove any wire connected to terminal 3. You only need wires on terminals 2 & 5, and they should be a pair, eg blue/white & white/blue.

[T_M_D]

You can try, Tina, but BT are renowned for the best customer service. Ultimately, the 24 hour test is the only protection your friend has again a £160 minimum BT charge if an engineer is called out and finds no fault on BT's side of that test socket. One thing you might consider is fitting a filtered faceplate instead of that dual one. It will isolate all the extensions from the ADSL service, but would require the router to be close to the master socket.

Filtered face plate will be a no-no then as it would mean them having a chunky desktop stuck in the (narrow) hall.

Forgive my obtuseness but why would running router in test socket over 24 hours prove anything? What would I be looking out for in the speed tests and what would/how would those tests prove or disprove BT responsibility? Not that I can do it anyway on this occasion, but I am curious in case I ever need to do it for myself. Tina.

[Rik]

Hi Tina

Snapshots of line conditions are just that. To get a real idea of what's happening, you need to run a program like Routerstats for 24 hours, thus charting the rise and fall of the noise margin and any resyncs that occur. In your friend's situation, it would pay to do two tests, one in the test socket and one in the normal socket. That would demonstrate what impact the internal wiring was having on the connection.

Did you see the bit about the ringwire, which we seem to have missed telling you about explicitly?

[T_M_D]

We don't appear to. Tina, you need to check every socket and remove any wire connected to terminal 3. You only need wires on terminals 2 & 5, and they should be a pair, eg blue/white & white/blue.

Does that include the master socket Rik/Steve and if so, are the wires you speak of in the upper part of the master socket? Could I do untold damage fiddling with these wires in the sockets? Would it be obvious which terminals are 2, 3 and 5? Is that something that would need to be done on a permanent basis, or is is something that is done and then put back after doing all the other tests?

[Rik]

They are numbered, most often the wire is orange/white, it's a permanent thing (the ring wire is a hangover from 'bell' phones), you should be able to lift out the wire with a small pair of pliers without any difficulty. The ring wire acts as an antenna for noise, so can really create havoc on a circuit.

[T_M_D]

Hi Tina

Snapshots of line conditions are just that. To get a real idea of what's happening, you need to run a program like Routerstats for 24 hours, thus charting the rise and fall of the noise margin and any resyncs that occur. In your friend's situation, it would pay to do two tests, one in the test socket and one in the normal socket. That would demonstrate what impact the internal wiring was having on the connection.

So I wouldn't have to sit around for several hours doing test after test then using the BT Hub stats? I would have a program running on the laptop monitoring it? Is the program Routerstats you mention expensive?

[Rik]

It's free.

http://www.vwlowen.co.uk/internet/files.htm
http://www.vwlowen.co.uk/moreinternet/files.htm

[T_M_D]

It's free.

http://www.vwlowen.co.uk/internet/files.htm
http://www.vwlowen.co.uk/moreinternet/files.htm

Thanks V. Much.  Interesting about the ring wire....Perhaps I should check that on my own master socket - my one and only socket in the whole house! It is rather old so it could well have the naughty wire lurking behind the socket. I think I will check it tomorrow in the day time so I can see it all more clearly.

[Rik]

It won't have a ring wire if it's the only socket, Tina.