Concerned about a potential poor line affecting speeds

Hi everyone

There is fibre availability in my area and I'm looking into it now (have been for some time).

The line it would need to be connected to is, as far as I recall, probably at least 20 years old if not considerably longer.

It's only within the last couple of years that the phoneline has had a standard face plate put on it, before that it was connected directly to the phone as the phone was one of those old ones with the spin dial on it.

It had to be changed over due to crackling on the line, whether it was the phone or the line I don't know but it stopped after the face plate was put in and a new phone was added.

Anyway...long and the short is...can a poor quality or old phone line affect FTTC speed? I assume it will but I thought i'd ask.

Is there any way of finding out the quality of the line? There is no broadband connection on the line at all at the moment.

Thanks very much!

[edit]
Is this the same premises as this thread, or is it somewhere else? If somewhere else, then continue with the checker I suggest. Unless you've already done that and forgotten to tell us what it says .

Other posters - check out that thread. It's meaty.
[/edit]

This checker will give basic information about what performance you could expect. The age of the copper is usually irrelevant.

Please copy and paste what the checker says, so we can interpret the results for you. It can be a bit daunting, and some of the figures shouldn't be taken at face value. (Don't paste your phone number, of course ).

Edited by RobertoS (Fri 30-Nov-12 17:48:18)

Yes line quality is actually a bigger factor for VDSL2

Quality for carrying a signal up to 17 MHz can really only be learnt by running the line with the VDSL2 service on it.

Or in plain English it is suck it and see

Andrew - note he has a rather detailed previous thread, see my edit above, but apparently a different line.

Oh blooming nora, I'd completely forgotten about that thread!!! *facepalm*

OK...in the house we have two phone lines...

One upstairs which currently has:
O2 broadband
O2 voice
VOIP

Downstairs
A bog standard phoneline with BT
A credit card machine which dials out using that line (home based business if you were wondering)

What I'm wanting to do is to kill the O2 line upstairs, which is about 14 years old, it was installed and wired up in 1998.

Enable fibre on the downstairs line and have:
Phone line with anytime calls on the line
Fibre
Credit card machine, upgraded to a broadband terminal so it doesn't tie up the phone line
VOIP number

All coming off one line...hopefully decrease the bills a bit.

I hope that'll all be OK running off one line!

*edit - checker details added*


ADSL2+ broadband line speed of 13Mbps; typically the line speed would range between 11.5Mbps and 14.5Mbps. Our test also indicates that your line could support an estimated ADSL 2+ Annex-M broadband upstream line speed of 1Mbps and downstream line speed of 13Mbps; typically the downstream speed would range between 11.5Mbps and 14.5Mbps.

Our test also indicates that your line currently supports a fibre technology with an estimated WBC FTTC Broadband where consumers have received downstream line speed of 25.7Mbps and upstream line speed of 5.7Mbps.


Interestingly I've just run the same test with my upstairs phone, and it's absolutely accurate with the downstream on ADSL2+, I do indeed get download speeds of between 10-12mbps, when downloading it tops out at an absolute maximum of 1.4MB/s - it also shows that the downstairs phone line should support slightly faster ADSL2+ speeds!

your line currently supports an estimated ADSL2+ broadband line speed of 11Mbps; typically the line speed would range between 10Mbps and 12Mbps. Our test also indicates that your line could support an estimated ADSL 2+ Annex-M broadband upstream line speed of 1Mbps and downstream line speed of 11Mbps; typically the downstream speed would range between 10Mbps and 12Mbps.

Our test also indicates that your line currently supports a fibre technology with an estimated WBC FTTC Broadband where consumers have received downstream line speed of 25.7Mbps and upstream line speed of 5.7Mbps.


It predicts the fibre will be the same speed for both lines...disappointingly lower than I had hoped. The upstream is good compared to the 1mbps we have now though...don't upload much though, except for backing up to cloud storage.

Looking on Google at the distance to the cabinet using the distance checker and comparing on the TB fibre board I was hoping for around 34mbps as we are around 700 yards from the exchange.

Edited by PhotoFiend (Fri 30-Nov-12 18:28:26)

Thanks for the reply....hmmm...I thought that might be the case...can't tell how good/bad it is until I've taken out at least a years contract.

I started an alternative suggestion, then decided it might be risky, came back to check something and lost it all anyway !

The estimates on a line do adjust over time to be close to the actual. Assuming you didn't put the upstairs number in by mistake, it looks as though they have picked the nearest comparable line to use as an estimate for your location ... which is ... .

Part of what I lost - it's almost certain there is only one cable coming to your house. That will have either two pairs or three pairs in it.

At the point it reaches the house one pair is supplying upstairs and one downstairs. The ages of the two pairs should be the same unless one has been replaced between that divergence point and a socket. The fact the two estimates are the same probably implies the two e-sides, (the lines from the cabinet to the exchange), follow the same route. They may or may not be down the same cable at that point, so ages could be different. It should be irrelevant.

As Andrew says, it is suck it and see, as what can make a difference is the state of every joint/junction over the whole route from socket to exchange.

My alternative, in short, was to discontinue the BT line and temporarily use the upstairs one for the credit card machine, then migrate the upstairs line and broadband to your chosen (BT?) line and FTTC supplier. The engineer at install line should be able easily to alter the link at the divergence point to send it all downstairs.

The risk is that the simultaneous migrations might fail to happen on the same day. The upside is that you already know the quality of the line, whereas the downstairs one may turn out to be better or worse.

In reply to a post by PhotoFiend:

Looking on Google at the distance to the cabinet using the distance checker and comparing on the TB fibre board I was hoping for around 34mbps as we are around 700 yards from the exchange.

It's distance from the cabinet to your property that matters, not the exchange with FTTC, it is possible your cabinet is further than the exchange.

In reply to a post by PhotoFiend:

it also shows that the downstairs phone line should support slightly faster ADSL2+ speeds!

No ,it doesn't! It is a pure figure plucked out the air. That line has no history of BB for the BT db to know anything about it other than its number.

The line to my place was fitted in 1967, still works, gets 70+ meg sync rate.

In reply to a post by R0NSKI:

It's distance from the cabinet to your property that matters, not the exchange with FTTC, it is possible your cabinet is further than the exchange.

Whoops, wording mistake, I'm not used to talking about cabinets! That's what I meant though, I used the Google maps route planner to trace down the roads (seemingly where the wires go) and it was 700 yards to the cabinet.

In reply to a post by Zarjaz:

The line to my place was fitted in 1967, still works, gets 70+ meg sync rate.

Oh...I hoped it might be a bit more scientific than that.

In reply to a post by Zarjaz:

The line to my place was fitted in 1967, still works, gets 70+ meg sync rate.

Good to know!

In reply to a post by R0NSKI:

In reply to a post by PhotoFiend:

Looking on Google at the distance to the cabinet using the distance checker and comparing on the TB fibre board I was hoping for around 34mbps as we are around 700 yards from the exchange.

In reply to a post by RobertoS:

I started an alternative suggestion, then decided it might be risky, came back to check something and lost it all anyway !

The estimates on a line do adjust over time to be close to the actual. Assuming you didn't put the upstairs number in by mistake, it looks as though they have picked the nearest comparable line to use as an estimate for your location ... which is ... .

Part of what I lost - it's almost certain there is only one cable coming to your house. That will have either two pairs or three pairs in it.

At the point it reaches the house one pair is supplying upstairs and one downstairs. The ages of the two pairs should be the same unless one has been replaced between that divergence point and a socket. The fact the two estimates are the same probably implies the two e-sides, (the lines from the cabinet to the exchange), follow the same route. They may or may not be down the same cable at that point, so ages could be different. It should be irrelevant.

As Andrew says, it is suck it and see, as what can make a difference is the state of every joint/junction over the whole route from socket to exchange.

My alternative, in short, was to discontinue the BT line and temporarily use the upstairs one for the credit card machine, then migrate the upstairs line and broadband to your chosen (BT?) line and FTTC supplier. The engineer at install line should be able easily to alter the link at the divergence point to send it all downstairs.

The risk is that the simultaneous migrations might fail to happen on the same day. The upside is that you already know the quality of the line, whereas the downstairs one may turn out to be better or worse.

Thanks for the post, that's really detailed!I could have a look this morning and see what wires go where...not really sure what I'm looking for though.

I see what you both mean with every junction making a difference.

Hmmm, it's an interesting thought I hadn't considered! However that plan would mean I would have to port the downstairs number (been in use for decades, can't lose it) to a VOIP which would put the costs up and starts to defeat the purpose of consolidating everything into one line, one bill.

In reply to a post by PhotoFiend:

Hmmm, it's an interesting thought I hadn't considered! However that plan would mean I would have to port the downstairs number (been in use for decades, can't lose it) to a VOIP which would put the costs up and starts to defeat the purpose of consolidating everything into one line, one bill.

Rules my idea out anyway then . Also eliminates the risk of downtime if anything went wrong with my suggestion, as you can have the new services all up and running before terminating the essential existing broadband.

700 metres from the cabinet is a fair distance. VDSL2 degrades quite badly at longer lengths. I have a good d-side, (line from cabinet to house), of 600 metres, and interleaved connection speeds as per my sig. Esimate:-

Our test also indicates that your line currently supports a fibre technology with an estimated WBC FTTC Broadband where consumers have received downstream line speed of 51.7Mbps and upstream line speed of 13.2Mbps.

Edited by RobertoS (Sat 01-Dec-12 10:11:20)

On the other hand I have a bad D side, only 450 meters, speeds as per my signature - note the dates, the slower one is the most recent, and I'm connected at 44Mbps

Estimate:

Our test also indicates that your line currently supports a fibre technology with an estimated WBC FTTC Broadband where consumers have received downstream line speed of 57 Mbps and upstream line speed of 20 Mbps.

Thank you both for those replies.

Hmmm 600m and still getting 50mbps sounds great to me, I do understand about it dropping off at distance though.

450m and 44mbps doesn't sound wonderful but as you say I suppose it depends on cabling.

Is there any way of finding out what my D side is? I guess I can't check the downstairs with no data currently enabled on the line but the upstairs with be interesting to know.

RobertoS, you're spot on with the downtime issue, I will be enabling the downstairs and making it all runs perfectly before shutting down the upstairs line, can't lose my precious!

Is there any way of finding out what my D side is? I guess I can't check the downstairs with no data currently enabled on the line but the upstairs with be interesting to know.

The best we can do on that is look at the overall performance of that line compared to its length and those estimates.

Speed test results are not a lot of use, so don't bother with those. What we need are the full stats, which if you are using the O2 box you can find by following these instructions.

Ah perfect, thanks very much! I wondered why I couldn't see everything when I was logged in as admin!

Stats are...


Link Information

Uptime: 7 days, 8:29:39

DSL Type: ITU-T G.992.5

Bandwidth (Up/Down) [kbps/kbps]: 1.308 / 13.470

Data Transferred (Sent/Received) [GB/GB]: 3,63 / 1,07

Output Power (Up/Down) [dBm]: 12,0 / 19,1

Line Attenuation (Up/Down) [dB]: 20,7 / 36,0

SN Margin (Up/Down) [dB]: 6,1 / 2,6

System Vendor ID (Local/Remote): TMMB / ----

Chipset Vendor ID (Local/Remote): BDCM / BDCM

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): -

Error Seconds (Local/Remote): 1.226 / 0

FEC Errors (Up/Down): 0 / 29.231.831

CRC Errors (Up/Down): 1.037 / 5.942

HEC Errors (Up/Down): 431 / 149.120

"Bandwidth (Up/Down) [kbps/kbps]: 1.308 / 13.470" is fairly good. It would be excellent but for "SN Margin (Up/Down) [dB]: 6,1 / 2,6".

Has the sync-time noise margin been set to 3dB by support, or is it 6dB and has sunk from it to 2.6dB?

Please can you do a reconnection tomorrow, some time between 10am and 3pm? (Best to power off the router, wait a minute, then power back on). Take and post the stats immediately afterwards.

I'm not too happy about the error seconds and FEC count.

Edited by RobertoS (Sat 01-Dec-12 16:32:15)

As you might have guessed I'm not very clued up on these figures, I don't know what they mean (know of a handy explanation guide?) so I'm going to have to answer simply...

Sync-time noise margin hasn't been touched as far as I know, I received a Wireless box V a few months ago and instantly got another 2mbps which I put down to better processing in the box (I really have no idea), but as far as I know, no settings have been changed by O2.

Sure thing, I'll do a reconnection in the morning!

What do the error seconds and FEC count mean? Is that the router or the line at fault?


Thanks again for all your help in this and the other thread, it's really appreciated! This forum should have a thanks button.

So, what Bob is saying is, if it is set to a 3db downstream SNR target, then that looks OK, might also the high-ish error count he mentions.

If it's set to 6db and has gone down to 3db, then that's not so good, and would mean the error count is evidence of iffyness on your line.

My original reply was just trying to state that age of the pair is not a factor, the condition of that pair IS, but the two don't go hand in hand.

Have you posted your postcode, some clever bunny may well be able to pinpoint your cab for you.

My line stats (under ADSL2+) are *almost* identical, except with a slightly lower attenuation value (so I am slightly closer to the exchange).

I too have a target SNRM of 3dB downstream (6dB upstream), and it hardly varies from that. This (as Zarjaz says) is a good thing.

The only downside is that it racks up the FEC counter - currently at 28 million after 2.5 days of uptime. Yours is only around a third of that rate, so isn't *too* bad.

The 2nd line here comes through the same drop cable - drop cable dates back to 1996, while the 2nd line itself was installed in 2002. It worked just fine on ADSL1 (8Mbps, 10 errors per hour), and used to have FTTC on it until a year ago - just before the 80/20 packages were available. It is about 550-600 metres from the cabinet.

When first connected, FTTC could hold a 40/10 connection during the first 2 days, but the number of errors on the line would trigger DLM to add FEC and interleaving, and drop the sync speed to 36Mbps (and download speeds to about 33 Mbps). Once BT re-organised the frequency bands in preparation for the arrival of 80/20, the modem suddenly got back the ability to sync at full 40/10 speeds (download at 37) - partly because the new bandplan gives a bit more capacity to downstream. However, it kept the FEC and interleaving overhead.

At the time the bandplan changed, I unlocked the modem, and was told that it had a maximum attainable speed of 60Mbps - so I suspect It'd achieve around 50Mbps sync speeds while the rest was used for the FEC overhead. The checker currently believes I'd have 51/11.

Does any of this mean anything for *your* line?

We have similar performance on ADSL2+ over the full line length (to exchange), with mine slightly worse. If the errors in both our cases come from the full line, then you would expect that you have a slightly better line (in terms of quality), but slightly worse (in terms of length to cabinet).

My gut feel says you'll probably be OK. If you're not, there are people here who can help - but you'll have to learn a bit about the technology & terminology

My ADSL2+ stats:

Link Information

Uptime:	2 days, 9:16:09

DSL Type:	G.992.5 annex A

Bandwidth (Up/Down) [kbps/kbps]:	1,268 / 14,103

Data Transferred (Sent/Received) [kB/kB]:	0.00 / 1.00

Output Power (Up/Down) [dBm]:	12.5 / 0.0

Line Attenuation (Up/Down) [dB]:	19.0 / 32.5

SN Margin (Up/Down) [dB]:	6.5 / 2.5

Vendor ID (Local/Remote):	TMMB / IFTN

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

Error Seconds (Local/Remote):	28 / 0

FEC Errors (Up/Down):	0 / 28,317,935

CRC Errors (Up/Down):	0 / 2,995

HEC Errors (Up/Down):	0 / 2,772

Morning everyone

Please find below the tests for this morning...

DSL Connection

Link Information

Uptime: 0 days, 0:01:54

DSL Type: ITU-T G.992.5

Bandwidth (Up/Down) [kbps/kbps]: 1.282 / 13.485

Data Transferred (Sent/Received) [MB/MB]: 1,92 / 2,44

Output Power (Up/Down) [dBm]: 12,1 / 19,3

Line Attenuation (Up/Down) [dB]: 20,5 / 35,5

SN Margin (Up/Down) [dB]: 7,0 / 6,2

System Vendor ID (Local/Remote): TMMB / ----

Chipset Vendor ID (Local/Remote): BDCM / BDCM

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): -

Error Seconds (Local/Remote): 0 / 0

FEC Errors (Up/Down): 0 / 898

CRC Errors (Up/Down): 0 / 0

HEC Errors (Up/Down): 0 / 1

In reply to a post by Zarjaz:

So, what Bob is saying is, if it is set to a 3db downstream SNR target, then that looks OK, might also the high-ish error count he mentions.

If it's set to 6db and has gone down to 3db, then that's not so good, and would mean the error count is evidence of iffyness on your line.

My original reply was just trying to state that age of the pair is not a factor, the condition of that pair IS, but the two don't go hand in hand.

Have you posted your postcode, some clever bunny may well be able to pinpoint your cab for you.

Thanks for the post, to be honest I'm not sure what I'm looking at with the stats.

That's what I meant with age really, logically age = worse condition but of course it's not always the case.


I haven't put my post code on, if I did you would be able to see 'me' completely due to what I do and where I am.

We are pretty sure we have located the correct cabinet, it's the next one along, on it's way to the exchange, if someone can tell me how I can check properly myself I will double check.

In reply to a post by WWWombat:

My line stats (under ADSL2+) are *almost* identical, except with a slightly lower attenuation value (so I am slightly closer to the exchange).

I too have a target SNRM of 3dB downstream (6dB upstream), and it hardly varies from that. This (as Zarjaz says) is a good thing.

The only downside is that it racks up the FEC counter - currently at 28 million after 2.5 days of uptime. Yours is only around a third of that rate, so isn't *too* bad.

The 2nd line here comes through the same drop cable - drop cable dates back to 1996, while the 2nd line itself was installed in 2002. It worked just fine on ADSL1 (8Mbps, 10 errors per hour), and used to have FTTC on it until a year ago - just before the 80/20 packages were available. It is about 550-600 metres from the cabinet.

When first connected, FTTC could hold a 40/10 connection during the first 2 days, but the number of errors on the line would trigger DLM to add FEC and interleaving, and drop the sync speed to 36Mbps (and download speeds to about 33 Mbps). Once BT re-organised the frequency bands in preparation for the arrival of 80/20, the modem suddenly got back the ability to sync at full 40/10 speeds (download at 37) - partly because the new bandplan gives a bit more capacity to downstream. However, it kept the FEC and interleaving overhead.

At the time the bandplan changed, I unlocked the modem, and was told that it had a maximum attainable speed of 60Mbps - so I suspect It'd achieve around 50Mbps sync speeds while the rest was used for the FEC overhead. The checker currently believes I'd have 51/11.

Does any of this mean anything for *your* line?

We have similar performance on ADSL2+ over the full line length (to exchange), with mine slightly worse. If the errors in both our cases come from the full line, then you would expect that you have a slightly better line (in terms of quality), but slightly worse (in terms of length to cabinet).

My gut feel says you'll probably be OK. If you're not, there are people here who can help - but you'll have to learn a bit about the technology & terminology

Thanks for the detailed reply!

From your post if our lines are comparible it sounds good to get fibre on the line I can test, the only issue is it's going to be going on the downstairs line which doesn't even have broadband on it so I have no way of checking the quality of the line which is a pain.

You're right, I really do need to learn about these stats, I'm shamefaced I'm so interested in the subject while knowing so little about it!

I've discovered Bobs website (thanks footer in his posts!) and I'll have a good read through that, hopefully it'll explain everything.

Edited by PhotoFiend (Sun 02-Dec-12 11:43:12)

http://www.coolwebhome.co.uk/fibre/checker.php may give you your cabinet number (no logs of entries are made on that site). Then its a case of finding the number written on the cab you think that is yours.

To be frank, I think without an engineer enabled FTTC at the cabinet there is no real way of knowing for sure what speeds you will get, as some many variables are at play.