Print

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

[Bald_Eagle1]

This seems to be the effect of crosstalk on my connection:-



Sync speed was around 30Mb 27th December, had reduced to around 26Mb by 5th January & reduced again to not much over 23Mb 19th April.

Error seconds have now increased from around 100 or so per day to around 250 per day (sometimes exceeding 300) & CRC/HEC & FEC errors have all increased, showing bursts:-








I have eliminated any internal interference by running the modem from a car battery & grabbing the stats with a battery powered laptop whilst all the internal power was completely switched off at the incoming mains consumer unit.




I don't THINK this is a physical fault, as such, as my connection is otherwise very stable with very few disconnections & SNRM tends to hover pretty steadily around the target 6 dB.

I believe BT are currently trialling Vectoring which may improve crosstalk matters.

However, from what I have read, as my D-side bundle also includes other users' ADSL connections over 1000m or more, Vectoring will have quite a limited improvement as it only works with like for like VDSL2 services from the same DSLAM.
ADSL 'interference' will still apparently have quite an effect & will still require various levels of Interleaving & INP to ensure stability for VDSL2 connections.

Anyone else seen similar connection changes & does anyone know if the HG612 modem is 'ready' for Vectoring?

I have no idea when/if BT will be making Vectoring available/compulsory for all VDSL2 users - any idea, anyone?

[c6em]

My thoughts:

1. ADSl2+ interference problem might be reduced as it only uses the lower frequencies of VDSL

2.  X-talk will affect the longer distance users most - so introducing vectoring might an attempt to improve the speed capability of those further away from the cabinet - rather than improve anything for those closest.

3.  That said, vectoring itself is a method of x-talk cancellation so it cannot improve speeds beyond that obtainable at the distance used - only prevent them being reduced (further) by noise.

4.  BT will have QA standards for suppliers and somewhere in there will be a requirement (or not yet) that all stuff such as the HG612 and the DSLAMs are vectoring compliant - or not.  Even if it IS now in the QA stuff whether it WAS in the QA when the first cabinets and modems were being put in is another matter.  In other words, assuming it's now in the procedures it may be that early cabinets do not have the capability.  (same thoughts apply to profile 30 - though the modem currently used has a 100mbps limit so would need changing anyway if you were close to the cabinet)

5.  If BT wanted to be really difficult it could offer vectoring as a cost add on for your circuit so that (I speculate) they could offer some sort of high speed stability guarantee for vectored lines which would be met by reducing speeds on no vectored lines - so you get a priority based system much like 999 calls are prioritised by the exchange over everything else............no, it too complicated to do and impossible to verify the effect for the customer.

6.  Vectoring should make it possible for BT to remotely diagnose imminently failing lines.  A single line which the data returned showed was giving off large amounts of noise leading to a high level of vectoring intervention by the DSLAM on the others can be ID'd for investigation.

[waltergmw]

@ c6em,

Thanks for those ideas. I  agree that para 5 doesn't seem very likely.
Re your para 6, given what is observed now, I strongly suspect that at least one agency within the BT Group has the capability of knowing precisely how every VDSL cct is performing. It is equally clear that there is no intention whatsoever to attempt any remedial works for those unfortunates suffering with quite appalling speeds. It is also known that some subcontractor VDSL installations without a JDSU or EXFO cannot give any attention to the line quality and sometimes the existing house wiring isn't even observed, let alone tested for it's suitability. (We also know that there are still a substantial number of all broadband houses with sub-standard internal wiring, often where the EU is in blissful ignorance as to what (s)he should do about it.)

Kind regards,
Walter

[ColinS]

It is also known that some subcontractor VDSL installations without a JDSU or EXFO cannot give any attention to the line quality and sometimes the existing house wiring isn't even observed, let alone tested for it's suitability. (We also know that there are still a substantial number of all broadband houses with sub-standard internal wiring, often where the EU is in blissful ignorance as to what (s)he should do about it.)

I've seen some of this myself Walter on my own recent contractor installation.  I believe they are simply relying on the SSFP to isolate whatever horrors may or may not lurk in the EU's internal wiring because it's beyond the NTE.  Of course, this could, most likely will, cause issues for the proposed PCP-only connections, depending upon whether or not the ISP provides an SSFP or dongles (centralised or distributed filtering models).  Because I was once a BT engineer myself, I know the state of my own wiring and have been able to move to a distributed filtering model.  The reduction in attainable rates is entirely consistent with an extra ~10m being added to the existing ~150m D-side.

[ColinS]

This seems to be the effect of crosstalk on my connection:-

BE - I never cease to be impressed with your ability to demonstate what's going on on your line! 

Anyone else seen similar connection changes & does anyone know if the HG612 modem is 'ready' for Vectoring?

Not yet & I have seen BTOR documents (somewhere, if only I could remember where ) that said that it would require (only) firmware updates to both the DSLAM and the VDSL modem, but as to whether Huawei have such an update ready for the existing HG612s, or whether EU's would need a modem swap-out is anybody's guess at this point.

I have no idea when/if BT will be making Vectoring available/compulsory for all VDSL2 users - any idea, anyone?

I think it is already on their plans to use it to try to offset FEXT (prevent reduced sync rates) as VDSL penetration levels increase.  Just the impression I got from what I read.  (note to self: go look for the source again!)

[Chrysalis]

To me whats most interesting is if ECI dslams will get vectoring.

The past 2 weeks has seen lots of news press releases about vectoring which has led me to doing some research, what I found was the existing HG dslams can probably be made to work at minimal cost, but the ECI seems much mmore questionable because possibly on ECI cabinets the entire line cards will need replacing something I expect BT wont be too keen on doing as they only just installed them.

It wouldnt surprise me worringly if BT rollout vectoring its not a 100% coverage across the HG footprint and might be exclusive to HG areas.

[ColinS]

To me whats most interesting is if ECI dslams will get vectoring.

The past 2 weeks has seen lots of news press releases about vectoring which has led me to doing some research, what I found was the existing HG dslams can probably be made to work at minimal cost, but the ECI seems much mmore questionable because possibly on ECI cabinets the entire line cards will need replacing something I expect BT wont be too keen on doing as they only just installed them.

It wouldnt surprise me worringly if BT rollout vectoring its not a 100% coverage across the HG footprint and might be exclusive to HG areas.

That would indeed be unfortunate    Can you give us all a reference to the need to replace the ECI line cards, as that would indeed be an interesting read?  As much fun as it may be to give Beattie grief over such matters, I would like to hope that they weren't quite as short sighted as have been sold a pup by ECI (in not being firmware upgradeable for vectoring).   

[Chrysalis]

Basically the M41 cards according to ECI's site do not have vectoring or at least vectoring isnt mentioned.

A guy on another forum who said he was part of 21CN design said BT use V41's and V41 do support vectoring.

However multiple people on tbb claim (without any reference to how they know) BT use M41's.  I find it shocking that BT would be rolling out M41's right now whilst also testing vectoring, that seems very shortsighted and just silly.

[ColinS]

Basically the M41 cards according to ECI's site do not have vectoring or at least vectoring isnt mentioned.

A guy on another forum who said he was part of 21CN design said BT use V41's and V41 do support vectoring.

However multiple people on tbb claim (without any reference to how they know) BT use M41's.  I find it shocking that BT would be rolling out M41's right now whilst also testing vectoring, that seems very shortsighted and just silly.

Yes, that would indeed be silly of them.  But, despite their bad press, I would like to hope that, like the BT guy said, it is perhaps just a case of 'mistaken identity' i.e. M41s for V41s? 

[snadge]

Is it not possible to install vectoring for both adsl & vdsl...?

Sent from my Sony Xperia Miro on Tapatalk

[ColinS]

Is it not possible to install vectoring for both adsl & vdsl...?

Sent from my Sony Xperia Miro on Tapatalk

As I understand it  , ADSL lines in the same cabinet will (or would) be more affected by crosstalk from the VDSL lines, because their signal strengths at that point are much lower than the VDSL ones as a result of the longer loop lengths.  At the moment this is mitigated to some extent by the PSD masks, and DPBO (Downstream Power Back Off) on VDSL US. G.vector -ing is only defined for FEXT between VDSL lines, but by reducing the power levels affecting the VDSL lines in the cabinet, the ADSL lines will also benefit (potentially more) from the reduced crosstalk in the same cabinet.
The quieter it is in the cabinet as a whole, the quieter it is in any individual xDSL line. 
[1] http://www.fie-epn.net/profesores/phidalgo/trabajos/trab_7.pdf is an early paper explaining some of the issues.
HTH 

[ColinS]

IMO this 2010 Alcatel-Lucent paper gives a fairly balanced view of the issues and possible solutions in the various FTTx approaches deployed to date.

[1] http://www.ospmag.com/issue/article/vdsl2-turning-copper-gold

[burakkucat]

I can provide a ZIP file containing a few selected images which clearly show that ECI HiFOCuS M41s have been deployed by Beattie.

Anyone interested? The file is 9.4 Mbytes in size, so please send me a PM containing your e-mail address to which it may be sent.

[ColinS]

I can provide a ZIP file containing a few selected images which clearly show that ECI HiFOCuS M41s have been deployed by Beattie.

Anyone interested? The file is 9.4 Mbytes in size, so please send me a PM containing your e-mail address to which it may be sent.

No, why would we doubt you? Not that I doubted Chrysalis' concerns either, just that I hoped they wouldn't be that stupid. Well, that answers that then, they have been rather short-sighted about this.

Are these deployed in any particular types of locations, e.g. smaller cabinets possibly, or in rural locations, or were they earlier or later deployments?  It would be interesting to know.  Since they have done that, then presumably either a kit swap-out would be required, or at worse, a fork-lift upgrade!

Another reason I'm glad I'm (lucky to be) on a Huawei cabinet then?!! 

[EDIT] This article, http://www.gazettabyte.com/home/2012/11/9/vdsl2-vectoring-explained.html apart from giving a quite lucid explanation of the vectoring process and its difficulties, does hold out some hope for CPEs (modems) that are not vectoring 'friendly' i.e. can't be upgraded for it (see the last paragraph).  But it doesn't say anything about what can be done about cabinets wearing 'the wrong trousers'. 

[waltergmw]

@ Coin,

As far as I'm aware, there is only one physical size of ECI cabinet which can also be identified externally from a Huawei one, as it has many more ventilation slots especially designed for friendly "visiting" dogs !

The internals are also identical for ECI 128s and ECI 256s, except for the number of line cards (2 or 4), terminating IDC connector blocks and, more importantly, the number of dual Telco 64 cables/connectors. Why they haven't put all four dual Telco 64 cables in beggars belief as any that require upgrades now must require a degree of dismantling live-working cabinets. I also can't fathom why BT don't include at least dual ducts nor why they throw away 28 or 56 connections as they never seem (at least in this part of Surrey) to install 100 pr AND a 50 pr tie cable sets.

BT were especially stupid in Ewhurst as they only installed ECI 128s despite VERY strong comments as our Vtesse Networks project that BT destroyed has specified cabinets with a total capacity of 500 services each. Two of the cabinets also only have single ducts to the PCPs whereas BT always install double ducts when 2 pairs of 100 pr tie cables are installed. The third cabinet BT "saved money" by using an existing duct with other cables installed. They got the tie cables and the fibre one in but then had to try three separate visits before they managed to insert a rod to pull in the 5 pr telemetry cable. Only the Almighty knows (at least at present) how much damage the other cables have suffered. When any expansion of ANY service - including the mythical FoD (Ha Ha Ha) fibre cables - is required, road excavation will be mandatory.

Kitz has posted pictures of ECI 256 FTTC and I have an ECI 128 pictures if anybody wants it.

Kind regards,
Walter