Line stats review please

Hi,

Long story short I have moved into a house in a very rural location. The BT checker advised I could get circa 3Mb however the actual speed I am getting is around 1Mb. Line stats follow but the engineer did have a hard time getting the phone line working (install was last Friday) so I am wondering if it's worth speaking to BT to have it all checked out again ?

Line state: Connected
Downstream: 1,056 Kbps
Upstream: 448 Kbps
Modulation: G.992.1 Annex A
Latency type: Fast
Noise margin (Down/Up): 10.3 dB / 18.0 dB
Line attenuation (Down/Up):52.0 dB / 30.5 dB
Output power (Down/Up): 15.8 dBm / 11.9 dBm
FEC Events (Down/Up): 0 / 28
CRC Events (Down/Up): 3083 / 29

Thanks in advance.

Matt

They are not good.

Are these figures when using the test socket, to rule out any extension wiring, faceplate faults

Are you still in the 10-day profiling period?

Your downstream noise margin is higher than it needs to be (6 dB is typical), but this extra margin may be necessary if your line gets noiser at night - particularly true of rural lines as they tend to run overground and pick up AM radio quite well once the sun goes down!

If you're outside the profiling period, try opening your master socket and connecting into the test socket to see if you get a better sync speed - home wiring can affect sync speed, and the test socket bypasses your home wiring.

[Just noticed your install was last Friday. I would be tempted to leave well alone for a while - and don't reset the router, as this can be interpreted as a dropped connection and adversely impact the profiling.]

Thanks both for the quick replies, I thought they looked bad but wasn't sure of the detail (especially the Attentuation numbers).

Yes I am still in the profiling period so I will leave it alone until that has finished and then try the test socket. The router is currently plugged into the master socket and nothing else is plugged in to any of the extentions throughout the house.

I'm concerned as the engineer took 1.5 hours just to get the line working and said the pair he had to use were "the last pair available", so I'm not sure what other options I will have available.

Thanks again

Matt

Edited by MattJessop (Mon 24-Sep-12 09:02:05)

There's an (approximate!) attenuation to line-length calculator here:

http://www.kitz.co.uk/adsl/max_speed_calc.php

Suggests your line is about 3.8 km. If that's in the right ballpark then your attenuation is more or less normal.

For what it's worth, my (rural) BT ADSL Max was syncing somewhere around 1.4 Mbps in the master socket (with no extensions plugged in) on a downstream attenuation of 37.0 dB.

In the test socket that went up to 3.8 Mbps sync speed and then 4.2 Mbps after a couple of weeks when the DSLAM twigged that the line had improved and automatically reset the SNR margin. Bad extension wiring in the house, clearly.

describe the internal wiring - do you have multiple sockets, which one is the router plugged into, do you have filters on every device plugged into the system ?

You should get 4M or more on that attenuation.

Hi, we've only just moved in but I'll try to describe.

We have the master BT socket in the hallway, the router is currently plugged into the normal socket (with a filter). Nothing else is currently plugged in (phones, sky boxes etc). There are numerous other sockets throughout the house, some of which work and some don't.

However of those that do work, they have varying degrees of "noise" on the line, so you may well be correct in that the internal wiring is causing the issue (it wouldn't surprise me, we've had all sorts of little "issues" crop up with other things).

I was thinking of getting someone in to sort the internal wiring out anyway, I'll get that sorted asap.

Does the master socket have a removable (half) faceplate? Is it the top one, NTE5, on this page http://www.kitz.co.uk/adsl/btsockets.htm ?

When you remove faceplate do all the other working sockets stop working?

Plug router with filter into revealed test socket. Do stats improve?

Plug corded phone with filter in test socket. Is it completely silent?

I will have a look tonight when I get home and give it a go..

In reply to a post by XRaySpeX:

Plug corded phone with filter in test socket. Is it completely silent?

Probably. Until someone rings in.

I assume you mean if used to do a Quiet line test, 17070 Option 2?

(Why the filter? Surely best done without that?)
_{Edit - strike out my garbage.}

Edited by RobertoS (Mon 24-Sep-12 16:04:38)

Guh! Of course I did; just forgot to state the obvious.

I thought filters were for filtering the BB out of the Voice, not t'other way round, so was needed with a isolated phone.

In reply to a post by XRaySpeX:

I thought filters were for filtering the BB out of the Voice, not t'other way round, so was needed with a isolated phone.

Yep! This time it's you that is right. I was confusing it with getting line stats.

So test socket with filter or no filter ?

Phone >> filter >> test socket.

Also, I don't know if it's been said, while you have the faceplate off make sure no other sockets in the house work.

Is there a phone-connected burglar alarm?

Will do and the alarm is not connected (mainly as it's a big hairy dog )

In reply to a post by XRaySpeX:

Guh! Of course I did; just forgot to state the obvious.

I thought filters were for filtering the BB out of the Voice, not t'other way round, so was needed with a isolated phone.

If the modem isn't connected, there won't be any broadband to filter so a filter is unnecessary when connecting the phone to the test socket.

OK now I'm confused...

Testing phone - use filter
Testing BB - no filter

Is that correct ?

What BatBoy says also makes sense. Now I think about it, I'm sure I've done phone testing with no filter.

Try it both ways. The worst that can happen is that without the filter you get a heck of a racket, and with it none or some. There should be none.

Broadband definitely without filter except the plug looks odd. A filter with nothing else connected to it except the router doesn't really create an issue.

Broadband definitely without filter except the plug looks odd. A filter with nothing else connected to it except the router doesn't really create an issue.

? Most DSL modem cables have a plug that only goes into the filter, not an unfiltered socket, surely?

Not that I've tried to make it fit, but the plug on a DSL modem cable is an RJ11, and the BT plug, er, isn't.

I admit I haven't tried it myself, but the specifications I've seen and the statements by many people on here who I believe implicitly say either plug will fit the standard phone socket and connect to the correct pins.

As far as the OP is concerned, it really doesn't matter. Either for phone testing or broadband testing. The important thing is that the test socket is used, with nothing else connected other than the device being tested, and that all other sockets on the premises are defunct when the faceplate is off.

In reply to a post by RobertoS:

I admit I haven't tried it myself, but the specifications I've seen and the statements by many people on here who I believe implicitly say either plug will fit the standard phone socket and connect to the correct pins.

Er, an RJ11 connector (which is smaller than the ethernet RJ45) will NOT fit a BT BS6312 connector. One is square and the other rectangular. The pins are in completely different places.

However you can get cables with BT BS6312 one end, and RJ11 the other - often supplied with laptops as "modem leads".

I admit I haven't tried it myself, but the specifications I've seen and the statements by many people on here who I believe implicitly say either plug will fit the standard phone socket and connect to the correct pins.

They won't. Be careful who you believe.

All UK DSL modems that I've seen come with RJ11 cables.

RJ11 is not physically compatible with a BT socket, so impossible to directly connect without a filter. It's much smaller - check out the adaptor here:

http://www.pmctelecom.co.uk/product_info.php?product...

I agree the key thing is the test socket. My concern was that the OP had (understandably) been somewhat confused by the information about filters.

In reply to a post by MattJessop:

OK now I'm confused...

Testing phone - use filter
Testing BB - no filter

Is that correct ?

Actually, it's the other way round, but only because of the plug differences for the modem. The filter doesn't filter the modem side - that's a straight-thru connection. And if the modem isn't connected, there's nothing to filter.

Ok got it, will report back later.

Thanks all

OK unplugged faceplate, plugged into test socket and got this


Line state: Connected

Connection time: 0 day, 00:01:24
Downstream: 4,448 Kbps
Upstream: 448 Kbps
ADSL settings
Modulation: G.992.1 Annex A
Latency type: Fast
Noise margin (Down/Up): 6.1 dB / 18.0 dB
Line attenuation (Down/Up): 51.0 dB / 30.0 dB
Output power (Down/Up):19.9 dBm / 12.1 dBm
FEC Events (Down/Up):0 / 0
CRC Events (Down/Up):488 / 0

So this obviously means an internal wiring issue, right ? I can get a man in for that.

Should I just leave the modem in the test socket for the time being, or should I try to disconnect the internal cable (bearing in mind I have no idea which one of the 3 it is) so I can put the faceplate back on.

I can post a pic if needed but one cable is blue, one is orange, one is white...

Thanks again

Matt

You're ok to leave it.

Duh! Of course! Not my day

Much better! You are now getting what you should

You have checked that none of the extension phone sockets work now?

Yup no other sockets working and I have a cabling engineer coming on Friday to sort it all out.

Thanks all !

You might be able to do it yourself and save some cash by Removing the Ring Wire

In reply to a post by XRaySpeX:

Removing the Ring Wire

"The i-plate is not compatible with the new (2007) NTE5 faceplates marked Openreach which already have a bell wire choke installed."

That's an interesting note from the Kitz page - I didn't know that

In reply to a post by 4M2:

That's an interesting note from the Kitz page - I didn't know that

Not compatible is a bit strong, they just won't do anything useful !

In reply to a post by jchamier:

Not compatible is a bit strong, they just won't do anything useful !

I'm not sure it isn't true. I seem to remember this being discussed soon after the new NTE5A came out, and there being something about the shape of the back of the NTE5A to accommodate the choke making the two not fit together.

It might be my imagination of course.

Edited by RobertoS (Mon 24-Sep-12 22:44:50)

In reply to a post by jchamier:

In reply to a post by 4M2:

That's an interesting note from the Kitz page - I didn't know that

Not compatible is a bit strong, they just won't do anything useful !

Sure, but presumably the choke on the faceplate fitted to an "Openreach" marked NTE5 will stop the negative effects of the bell wire on the adsl signal in a similar way that an i-plate does with earlier versions of NTE5 faceplates.

Personally I still prefer to remove the bell wire from an extension and use a filtered faceplate, but I did find Kitz's mention of a choke fitted to new faceplates to be interesting http://www.kitz.co.uk/adsl/images/phone/NTE5_facepla... - choke marked with a red circle.

In reply to a post by RobertoS:

...I seem to remember this being discussed soon after the new NTE5A came out, and there being something about the shape of the back of the NTE5A to accommodate the choke making the two not fit together.

The new NTE5A does fit onto an older NTE5 - I've just checked. Since the terminals on a new NTE5A are reduced to three and the position of the choke is where the other three terminals are on an older NTE5A

In reply to a post by 4M2:

Personally I still prefer to remove the bell wire from an extension and use a filtered faceplate, but I did find Kitz's mention of a choke fitted to new faceplates to be interesting http://www.kitz.co.uk/adsl/images/phone/NTE5_facepla... - choke marked with a red circle.

Yes, I've seen at least two installed. Helpful OR engineers had fit them in both cases when fixing other PSTN faults.

That's not the question Of course it fits an NTE5.

The question is does it still fit if an iPlate is there.

In reply to a post by RobertoS:

That's not the question Of course it fits an NTE5.

The question is does it still fit if an iPlate is there.

Ah! Good point, I don't have an i-plate to check unfortunately

A quick update..

The cabling guy was here about 3 hours sorting out all the internal wiring (only cost £80, I was happy with that).

Broadband originally connected at 4000Kbps but had a few stability issues, the SNR margin was low at 3Db.

A couple of days later it seems to have settled at the following -


Line state: Connected
Connection time:1 day, 21:30:20
Downstream:3,456 Kbps
Upstream: 448 Kbps
VPI/VCI:0/38
Type:PPPoA
Modulation: G.992.1 Annex A
Latency type: Fast
Noise margin (Down/Up): 6.1 dB / 16.0 dB
Line attenuation (Down/Up): 52.0 dB / 30.5 dB
Output power (Down/Up): 19.2 dBm / 12.1 dBm
FEC Events (Down/Up): 0 / 22
CRC Events (Down/Up): 7332 / 235
Loss of Framing (Local/Remote): 0 / 0
Loss of Signal (Local/Remote): 0 / 0
Loss of Power (Local/Remote): 0 / 0
Loss of Link (Remote): 0
HEC Errors (Down/Up): 7874 / 130
Error Seconds (Local/Remote): 0 / 0

I'm hoping the line will now stay stable and my IP Profile will go up to 3Mb (it's currently stuck at 2Mb).

You were getting a downstream sync of 4448Kbps from the test socket on the 24th September, so it looks like you have somehow lost nearly1000Kbps?

Yes caused by internal wiring issues, I can live with 3mb, as I said its a very rural location.

Did your "Engineer" fit a filtered faceplate, and then run all the extensions from the 2 and 5 terminals on the back of that?

To me, that is by far the best way to do it.

Sounds to me like the "Engineer" has just connected all the extensions (maybe removed a star-wired layout) but left you with plug-in filters at all points? Less than optimal, shall we say.

No he didn't, because that is not what I needed him to do.

I'm not sure why you felt the need to put engineer in speech marks, but I can hazard a guess.