Print

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

[Bald_Eagle1]

Incidentally, and I've mentioned this on here before, my next door neighbours line has a lower attenuation than mine, around 10 db lower even though her line goes to the same pole and no doubt the same cabinet. She gets a faster speed than me because of this.

Have you a means of discovering your neighbour's FTTC speed estimate as a comparison against yours?

I just hope that the reason for the poor performance of my line up until now is before the cabinet and not after it as the fibre to the cabinet will solve that. (fingers crossed)

Fingers crossed indeed. A 10dB difference in attenuation levels does appear to make a substantial difference for FTTC connections, speed-wise.

[Maturecheese]

Sorry I should have said the cable between the Exchange and the cabinet fingers crossed

I will check my neighbours FTTC estimate tomorrow, could be interesting.

EDIT  I have tried to check but as my neighbour is not with bt I cannot input the tel No and for some reason the checker won't accept house No and postcode.

[Bald_Eagle1]

Sorry I should have said the cable between the Exchange and the cabinet fingers crossed

I will check my neighbours FTTC estimate tomorrow, could be interesting.

My next door neighbour's is the same as mine (30Mb)

[burakkucat]

EDIT  If the line is Aluminium then I'll just have to put up with slower speeds but it doesn't explain the discrepancies between my line and next doors as surely if mine is aluminium so is hers.

Two adjacent properties with drop-cables from the same pole, which are fed via the same (aluminium) multi-pair D-side cable from the PCP. How can one pair have 10 dB more attenuation than the other? Possibly because of more (defective) joints. Aluminium is an amphoteric metal; it reacts readily with both acids and alkalis. Aluminium is actually a very reactive metal. It appears to be fairly stable and unreactive because of the thin film of aluminium oxide that rapidly forms on its surface. Now get two pieces of aluminium in contact with yet another metal & atmospheric moisture and the aluminium will readily corrode. If it is expected that an electrical signal, at radio frequencies no less, will flow through those two pieces of joined aluminium then there will be a severe degradation of the signal. It will be attenuated. If those two pieces of aluminium are sections of a D-side cable pair and the other metal is an old-style "blue bean" type of crimp, them I'm sure I don't need to type much more. 

[Maturecheese]

EDIT  If the line is Aluminium then I'll just have to put up with slower speeds but it doesn't explain the discrepancies between my line and next doors as surely if mine is aluminium so is hers.

Two adjacent properties with drop-cables from the same pole, which are fed via the same (aluminium) multi-pair D-side cable from the PCP. How can one pair have 10 dB more attenuation than the other? Possibly because of more (defective) joints. Aluminium is an amphoteric metal; it reacts readily with both acids and alkalis. Aluminium is actually a very reactive metal. It appears to be fairly stable and unreactive because of the thin film of aluminium oxide that rapidly forms on its surface. Now get two pieces of aluminium in contact with yet another metal & atmospheric moisture and the aluminium will readily corrode. If it is expected that an electrical signal, at radio frequencies no less, will flow through those two pieces of joined aluminium then there will be a severe degradation of the signal. It will be attenuated. If those two pieces of aluminium are sections of a D-side cable pair and the other metal is an old-style "blue bean" type of crimp, them I'm sure I don't need to type much more. 

Wouldn't that senario show up as a High Resistance fault?

[asbokid]

Playing Devil's Advocate a bit here...

In 2009, Professor Barry Forde of Lancaster University produced a Technical Reference report on LLU (local loop unbundling) for the JANET Network Access Programme. [1]

There is a table on page 7 of Forde's report. It lists the range of wire types used historically in BT's network.  The table shows eight different types of copper cabling and four different types of aluminium cable.

Each type of cable is listed with a loop resistance (measured in ohms/km) and an insertion loss (measured in decibels/km @ 1600Hz).

That table from Forde's report is attached below.

The figures for insertion loss are interesting. They appear to show that heavier gauge aluminium cabling (0.8mm) can outperform lighter gauge copper cabling (0.5mm), in terms of insertion loss at 1600Hz.

However, 1600Hz is at the low-end of the voiceband. It may be unrepresentative of the insertion losses found at the much higher frequencies used in xDSL broadband access systems.

[1] http://www.ja.net/documents/development/llu/llu-technical-document.pdf



[attachment deleted by admin]

[Maturecheese]

Is it possible to officially find out if a line is Aluminium and the thickness of it?  If one can and it turns out it is aluminium is there a hope in hell of getting it replaced by either copper of fibre in the next decade?  I suspect the costs involved will rule it out indefinitely.

[asbokid]

Is it possible to officially find out if a line is Aluminium and the thickness of it?  If one can and it turns out it is aluminium is there a hope in hell of getting it replaced by either copper of fibre in the next decade?  I suspect the costs involved will rule it out indefinitely.

It would be reassuring to learn that BT Openreach has a database in which it records the cable type(s) and lengths used in every subscriber line.   Though, in practice, the records are probably very vague.  Maybe the use of aluminium conductors in the larger distribution bundles is recorded?

Openreach must have a long-term programme to replace all that aluminium with copper.  Has that programme ever been disclosed?  Copper will be obsolete by the time the aluminium is eventually replaced, especially if Walter has his wicked way with FTTP rollout!

You can bet that the replacement programme would accelerate rapidly if the price of aluminium were to soar and the copper price were to plunge. However, that doesn't seem likely in the near future. Here are the Friday close (cash) prices on the London Metal Exchange for aluminium and copper.  The prices are quoted in US dollars per metric ton.

Aluminum  2184.00
Copper      8230.50

As for discovering whether your own twisted pair has aluminium conductors, without access to Openreach records, it's a good question.  Maybe the line characteristics data obtained from the modem could give a strong indication. It's a difficult problem.  The line could be 'inhomogeneous' along its length, comprising multiple sections of different cable types.  And maybe the succession of cable joints is causing more damage to the signal than the types of cable in use?

cheers, a

[burakkucat]

EDIT  If the line is Aluminium then I'll just have to put up with slower speeds but it doesn't explain the discrepancies between my line and next doors as surely if mine is aluminium so is hers.

Two adjacent properties with drop-cables from the same pole, which are fed via the same (aluminium) multi-pair D-side cable from the PCP. How can one pair have 10 dB more attenuation than the other? Possibly because of more (defective) joints. Aluminium is an amphoteric metal; it reacts readily with both acids and alkalis. Aluminium is actually a very reactive metal. It appears to be fairly stable and unreactive because of the thin film of aluminium oxide that rapidly forms on its surface. Now get two pieces of aluminium in contact with yet another metal & atmospheric moisture and the aluminium will readily corrode. If it is expected that an electrical signal, at radio frequencies no less, will flow through those two pieces of joined aluminium then there will be a severe degradation of the signal. It will be attenuated. If those two pieces of aluminium are sections of a D-side cable pair and the other metal is an old-style "blue bean" type of crimp, them I'm sure I don't need to type much more. 

Wouldn't that senario show up as a High Resistance fault?

Yes, you are correct. A HR fault should be found -- if one (or more) were potentially being investigated.

[Bald_Eagle1]

Yes, you are correct. A HR fault should be found -- if one (or more) were potentially being investigated.

"if" being the operative word.

In my experience, it has been rather difficult, to say the least, to get anyone from the provider's end to show any real interest whatsoever in potential HR faults, especially when a given connection falls within "acceptable Fault Threshold Rates & is performing above estimated speeds".

[burakkucat]

The HR faults that one can fairly easily get investigated and fixed are those that have an adverse affect on the telephony service. With hisses, crackles, plops, pops, etc, it is a bit difficult for a representative of an ISP / CP to deny the presence of such a fault when the line in question is being used to report the problem. Unless, of course, s/he suffers from a "Nelsonian Ear". 

[Bald_Eagle1]

Update 05/02/2012:


Just as an experiment while my sync speeds are still low, at my request, Plusnet have recently experimented with resetting their throughput profile for my connection.

BT's SIN 498 document states that ISPs should try to set throughput profiles to closely match achievable speeds to avoid traffic losses etc.

I was seeing really high error counts (millions at a time), even following the latest engineering work to replace a section of dropwire, that did actually help to further stabilise my connection.

I was on a 37Mb Plusnet profile, which means full speed throughput for the currently capped 40Mb offering.

As my speeds have not exceeded 30Mb since July, I asked Plusnet to lower my throughput profile to 32Mb.
Error counts have reduced dramatically & likewise the frequent "on the fly" re-syncs, usually at lower speeds, have also reduced.

My sync speeds have been gradually increasing & I actually achieved just over 30Mb sync speed a few days ago (for the first time since July!).

Hoping, that lower error counts & less frequent re-syncs would be seen by DLM as a more stable connection, I was hoping for a prolonged error-free connection up time.

Everything was looking positive until I arrived home from work on Thursday, when I spotted an engineer at the top of the pole nearest to my house.

Of course, I rushed inside just to check everthing was still O.K. with my connection.
It wasn't.
I could see that for around 1/2 hour my SNRM had been all over the place & there was evidence of a few re-syncs in quick succession.

I spoke with the engineer & he denied touching my connection as he was working to provide a new one for someone else.
I ended up with a sync speed of not much above 20Mb.

as it was dark by the time the engineer left, my subsequent re-sync was O.K. speed-wise, but below 30Mb again.

Anyway, I contacted Plusnet just to let them know what had been causing this recent instability.

When I retuned from work on Friday, everything seemed fine again, until I noticed my connection re-syncing again.
Oh no. I thought. That enginer has messed up my connection again.
Just as a check, I looked out of the window & there was a different engineer at the top of the same pole.
He also denied touching my connection, but I lost a few more Mb in sync speeds again.

I rebooted the modem yesterday morning & achieved just below 30Mb sync speed.

I have reported both these incidents to Plusnet & requested an engineer's visit to thoroughly test my whole D-side using TDR and/or whatever else would be necessary & if no "issues" are detected, to INSIST on BT resetting my connection to 40Mb as it has never actually been confirmed that it has been reset to 40Mb following ANY of the previous engineers' visits.

So the message is that ISPs attempting to send too high a throughput on a poorly performing connection can actually cause even more instability & actually lower ongoing sync speeds.

Watch this space for any updates.


Paul.

[waltergmw]

[burakkucat]

Everything was looking positive until I arrived home from work on Thursday, when I spotted an engineer at the top of the pole nearest to my house.

Of course, I rushed inside just to check everthing was still O.K. with my connection.
It wasn't.
I could see that for around 1/2 hour my SNRM had been all over the place & there was evidence of a few re-syncs in quick succession.

I spoke with the engineer & he denied touching my connection as he was working to provide a new one for someone else.
I ended up with a sync speed of not much above 20Mb.

. . .

When I retuned from work on Friday, everything seemed fine again, until I noticed my connection re-syncing again.
Oh no. I thought. That enginer has messed up my connection again.
Just as a check, I looked out of the window & there was a different engineer at the top of the same pole.
He also denied touching my connection, but I lost a few more Mb in sync speeds again.

To me that is the evidence of a defective DP, which needs to be rectified by Openreach.    A total "refit" of the pole should be implemented.

[waltergmw]

@ BKK,

Although that might well seem to be the most obvious component (culprit ?), there are probably others.
I very much doubt if BT Openreach would even consider any cause postulated by their clients, especially as there won't be any official cross reference between Paul's service and the new service(s).

Perhaps the most that could be asked for is sight of the TDR data from the JDSU. That also assumes that PlusNet have the capability to obtain action from BT Openreach.

Kind regards,
Walter