Kitz Forum
Chat => Tech Chat => Topic started by: kitz on January 31, 2008, 09:50:33 PM
-
Posted in this thread because I do not want several threads where customers have issues with their connections taken off topic, by this debate.
This forum is to be constructive in the help of all users regardless of their ISP.
---------------------------------------
My response to Roseway, if what you are attempting to perport was true then some of the basic laws of Physics would have to be re-written. The example I must site would be a line of 5 kilometers using your analagy would give a line attenuation of 20db per kilometer over say 5 kilometers the result 100db of line loss. You therefore will be able to supply a broadband signal of usable level to all peoples in all corners of the UK with little difficulty.
My figure gives approx 50db for a five (5) kilometer length of line attenuation which is the reason why broadband will function up to approx 65db and six(6) kilometers. QED!
I must take issue with your last post Kitz and state that the figure I a,m quoting is more than generous at 10db per kilometer and I can state that was obtained some years ago after extensive and refereed research across the specrum in use.
I still fail to see, nor have I ever seen, any evidence that 10dB per km is a generous figure. I still stand by my statement that in the UK the average loop loss is around 15dB per km. Actually perhaps I should have said 14dB.
Its one thing saying bt should give the users 10dB per km, but I fail to see how this is possible based on adsl technology and the fact that its carried over copper wire. Attenuation isnt something that BT - or any telecom co - can "dish out" to a magic number of 10.
I have however seen evidence that 14dB per km is about right. The evidence wasnt supplied by BT but its a world wide figure therefore there cant be any conspiracy theories that the figure has been cooked by BTw. The survey was conducted about 2 years ago by Consultel in specific relation to adsl.
When I coded the max speed checker I based them on those figures, and once it was complete, I then spent the best part of a whole afternoon and evening checking every single line stat I could find against it to make sure it was as accurate as it could be in the UK.
I literally checked it against a few hundred line stats from various different sources.
I am not going to spend several days coding something then make a idiot of myself by putting it up on the main site if it was going to be wrong, without doing some research of my own first. :(
If you feel my figures are nonsense then you should also approach Mr_Saffron (who is one of the best adsl experts in the UK) and tell him that he's working on wrong figures too.
Both Andrew and myself spend a huge amount of time trying to help asdl users in the UK.
Im sorry that if you mistakenly feel that some of those that spend a huge amount of time trying to help adsl users, if they dont always agree with what youre saying must be in the pockets of BT.
I was quite shocked to hear a few weeks back that you perhaps class me in that category.. I can 100% assure you Im not. Anyone who knows me and my history of run-ins with BT would know just how ludicrous that statement is.
The biggest reason this site is here in the first place... and why I funded it myself and spent countless hours for so many years.. is because of the problems Ive had with BTw and therefore I wanted to try and help others too.
I would like to think that I try if possible to give a balanced and factual view of things ... rather than point fingers in the wrong direction and blame BT at every opportunity.
I am annoyed however that yet again a thread which is supposed to help a user went off topic. You already know my thoughts on the only way that uk adsl is going to get decent speeds.. and that certainly isnt going to be possible without a huge investment and FTTC.
-
quote (http://forum.kitz.co.uk/index.php/topic,1357.msg49438.html#msg49438)
I must also add that I stated yesterday a figure of 10db per kilometer I should have said 10db per mile as the maximum line attenuation that should be accepted. That is a figure that was accepted in addition up to the few honest staff in BT up until one year or so ago. All the acceptance tests in obtaining those figures were the result of refereed papers from such research. So I refute any critisism.
I just noticed that you are now saying per mile rather than km
1 kilometer = 0.6213 mile @ 10dB per mile
So therefore are you are now saying this report stated that attenuation should be no more than 6.213 dB per km?
I can honestly say Ive never seen a line recorded at such a low looploss !!
Heres my proof that loop loss is considered to be 7dB per 0.5km in relation to adsl
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fwww.internode.on.net%2Fmedia%2Fimages%2Finternode-adsl2-dist07.jpg&hash=da22c91d7e19e304485fb7f9955fcb22f9aa8a03)
-
You're wasting your time kitz. Ignore him as he doesn't have the first idea what he is on about and nor is he willing to correct his abysmal lack of knowledge by listening. Works for me ;)
-
......... and the last time i had a bt engineer out to my place he said 15db per km is what they work on ...........
-
looking back in my log book:
line lenght 1080m, downstream loss 21 db
6440m 43 db
3040m 49.5 db
4306m 53 db
2350m 35.5 db
4180m 57 db
3030m 41 db
113m 16 db
4480m 55 db
263m 10.5 db
893m 19 db
as a genereralisation 14-15db/km seems about right, assuming all the wire is 0.5mm copper. some cables & routes have some variations 0.4-1.2mm copper and similar in aluminium, all will affect the overall result & is some times part of what an engineer may look into during a fault. I've got broadband working ok at 9.3km though this was mainly 0.63mm copper. I've even seen a modem sync at 11.5km (though only for 10-30 seconds at a time)
-
With regards to the conductor size making a difference, is that sometimes why pairs are doubled up on very long lines?
Attached is a pic of the statistics from an ADSL circuit in an exchange building. It doesn't get much better than that.
[attachment deleted by admin]
-
......... and the last time i had a bt engineer out to my place he said 15db per km is what they work on ...........
Early on they were using 10 dB/km so for example the 55 dB limit equated to 5.5 km which then extended to 60 dB / 6km. See for example http://www.bb4mk.org/ADSL%20extended%20reach.htm
I'm not sure everyone has a good handle on their actual line length, and a further question arises over the definition of attenuation - is this at 300 kHz or at a weighted average of the frequencies in use or.....
If longer lines have fatter copper to compensate (on voice) then the dB/km will be lower in any case, I've seen numbers like 62 dB over 10 km on a thick overhead line.
Phil
-
Thank you Phil,
I believe there are certain individuals on this site acting like "barrack room lawyers" in the sense of being "pretend competent engineers" sadly misleading those seeking advice. Where I pay credit to the wealth of information that is available on this site. No one can condone quoting figures that they are neither competent or with the facilities to measure. Phil the facts you quote are absolutley correct. In fact the work was originally per mile. We must I believe consider accepting the figures these days per kilometer.
Now with reference to the brightly coloured graph in this thread it I believe has little practical relevance to the practical situation in the Uk but is an interesting guide to the overall picture and is of course drawn up by whatever means by a group in Australia. So please in future quote from work with the credibility of the IERE or IEE in the UK.
I shall stick to and believe customers should expect a figure of 10db loss per kilometer to be an acceptable line or the provider should be held responsible for bringing such line up to an acceptable service standard. To start accepting 15 or 20db loss per kilometer is to enter the slippery slope of those customers paying for a service which they would not be receiving. Therefore a line indicating 15 or even 20db loss per kilometer is clearly one which requires rapid maintenance or a cable "uplift".
-
Ignoring your offensive remarks about 'barrack room lawyers' and 'pretend competent engineers', the facts still are that a loop loss of around 15 dB/km is typical. It isn't the best which can be achieved, fairly obviously, but lower loss would require more expensive cable. This discussion isn't about what you would like (and presumably are prepared to pay for) but about what IS.
If you choose to believe that anything over 10 dB/km is unacceptable, that's your prerogative, but all considerations of this nature are about what's achievable at a price the customer is prepared to pay. Anyone can grab a figure out of the air and say that's what they want, but here we live in the real world.
-
I am not plucking a figure from the air and I am living in the real world. 10db per kilometer is what the network should provide anything less is NOT what the customer is paying for. We also must insure that the network is maintained with the standard of cable used that is appropriate for its length, if that is not the case then those responsible must be called "to book". If you wish to go to the garage and purchase a gallon of petrol and go away with 4 pints or even 7 1/2 pint I bet you would be one of the first to call in trading standards. Im not in this world to "sell the customer short" as it would appear some would appear to condone.
-
The customer is not paying to for 10dB per km. Most customers are only after the cheapest monthly tariff they can get from ANY broadband supplier. Obviously BT foots the bill for anything to do with the access network and in the present climate of Ofcom driving market competition there is absolutely no way such infrastructure upgrades are going to happen. Since the majority of customers are oblivious to the term attenuation, I don't think they are going to get to upset by the difference between 10dB/km and 15dB/km.
-
Hi Soms. Thanks for that we shall have to see what comes from the Ombusman and OFCOM in the near future. BT do have a responsibility to maintain the network to an acceptable standard and as Yarwell has brought out a document that is in the public domain we shall have to see what penalties are brought to bear, where there are service failures
Its sad but for anyone to take advantage of the public as you say not being aware is for those apointed responsible bodies to take care of the public. If cables that are not up to standard for the job have been laid then it must be corrected.
it is not for anyone here or anywhere else to condone a substandard network or suggest to customers that they put up with 15db or 20db, they should point out that the customer has statutory rights and for those to be upheld and if necessary competent and qualified persons who are independent to be called in to report and correct.
-
Your views on BT are well known, but they are simply your opinion. Kitz has done extensive research which shows clearly that ~15 dB/km is quite typical, not only in the UK, but around the world. We're not here to defend BT or anyone else, but unless you can actually produce some facts to support your opinions this discussion is a waste of time. Describing the BT network as substandard is just so much hot air unless you can justify it with properly researched information. You've been asked to do this several times, but have failed to do so.
-
Roseway, you simply must start to understand basics and stop using work and figures not substantiated or carried out under any form of control. "Yarwell" by putting up the link that he has, has spoken for the public. I repeat,
"Early on they (BT)were using 10 dB/km so for example the 55 dB limit equated to 5.5 km which then extended to 60 dB / 6km. See for example http://www.bb4mk.org/ADSL%20extended%20reach.htm.
Taking the attitude that you are taking can leave me and others with the only view that you are clearly in support of a degrading network, one must therefore question your motive and I believe it will leave the public at large with a clear view of how your view will benefit you. I am not the owner of the rights to the research and testing so am not in a position to place such on this site. So I and those other qualified professionals will continue to state that any line loss greater than 10db per kilometer is more than that which should be acceptable to any customer. "Ezzer" in this string has shown quite clearly how poor the quality control of this network in fact is, however, he does not state and this is not a criticism of that which he has published, how and what if any calibrated test equipment was used.
-
You know, I'm beginning to get more than a little irritated by your patronising remarks, and your repeated vague references to unidentified sources. Who are these 'qualified professionals' who you associate yourself with? Where is the link to yarwell's posting to which you refer, so that we can read it for ourselves and not have to rely on your translation of it?
-
You state I fail to supply any references or link, as you claim to be an oracle on the subject you clearly should be fully apraised of all facts without any assistance before profering advice to others. We believe all can determine the facts by reading this thread. The fact remains that customers who have a line attenuation greater than 10db per kilometer have a line requiring maintenance or an "uplift". QED
-
I really do believe this has got very much out of hand.
This all started as a result of this thread here
http://forum.kitz.co.uk/index.php/topic,1331.msg46375.html#msg46375
Because I dared to state a reality "The 10dB figure is a "pluck out of the air" figure and bases no reality on the true situation of the vast majority of lines in the UK".
Because I stand by that fact, you seem to be taking every opportunity you can to rant and discredit me. Spreading rumours and implying that because I stand by that I and others are in the pocket of BTw. I have heard this fact from several sources and this is downright libellous.
You have a serious issue with BT, I don't know the exact cause of it, but to contact people and spread malicious rumours is something I do take big issue with.
Believe it or not I do actually believe that BT can and should do more for some lines, but Im a realist and know the huge cost of investment. I still stand by the fact that the vast majority of lines in the UK are more than 10dB. That would appear to be a fact and until you come up with evidence to the contrary, I will let others make up their own mind.
In your vendetta you have obviously called Phil into this.. but I bet you havent told Phil the whole story have you. You had already been advised earlier this month that it would be better to bring your rants about BT into a separate thread in the tech chat section. Continuing to rant about BT and the 10dB situation in totally irrelevant thread in which are discussing other issues and taking the subject off topic does nothing to help the original poster as to why they come here in the first place.
I have been fair and given you free rein to rant at BT, the only times I took exception is when it was totally irrelevant to the topic involved and that is why I brought it into this thread.
The fact that I have had several enquiries from users of this forum as to why you are PMing people giving them your phone number and asking them to call you actually has disturbed several people who come to these forums asking for help.
I will not stand by if users who come to this forum are concerned and find it a little strange.. but the audacity to also imply that I and other members of this site are in BTs pockets is libellous.
You think Im pro-BT... then why would I actually suggest to someone specific that they contact you and see if there is anything that can be done about your "cause", because I cant go down to London. If that person has contacted you or not I don't know.. I just made the suggestion that maybe someone should to see if it could help further your cause.
The basics of this forum are that we will try to help anyone - who ever their ISP is, but ranting at BT rather than try to investigate some of the basics has taken several threads off topic.. and that is what you have been politely requested not to do. If you dont agree to the rules then I suggest you find elsewhere to take your rants.
>> Now with reference to the brightly coloured graph in this thread it I believe has little practical relevance to the practical situation in the Uk but is an interesting guide to the overall picture and is of course drawn up by whatever means by a group in Australia.
Please read it again.. I didn't hide the fact that it was done for an Australian company.. but adsl technology is the same world-wide and runs on the same frequencies. I also stated that when I had coded the checker that I ran it against every UK line stat I could lay my hands on to see if it also came within acceptable parameters - which it did. There may be some differences in the amount of copper used.. just like many different areas of the UK uses different amounts of copper in the line too.
Offensive and incorrect posts such as some of yours do nothing to give you credence.
No-one on this site has ever pretended to be a competent engineer, if I dont know I will say so, and I wont blindly make facts up for the sake of it.
This site aims to give information to help people diagnose problems and I am not about to let it be destroyed by someone with a huge chip on his shoulder about BT, and then attempts to spread malicious and untrue rumours simply because I dared to state that from my observations the average line length is more like 14dB.
I am entitled to have my own opinions of the running of BT... and just because it may differ from yours doesn't mean that I am wrong... nor does it mean that I am in BTs pockets (incidentally an insult you would appear to dish out to anyone who disagrees with you).. nor does it give you the right to make up things as you go along.
>> it is not for anyone here or anywhere else to condone a substandard network or suggest to customers that they put up with 15db or 20db,
Please stop twisting facts - what we have done is said that 14-15db for km it appears to be the average and far more realistic than your figure of 10db per mile or km.
I have still yet to see some official evidence to support your claims that adsl attenuation should run at 10dB per mile or km. I'm aware there are regulations for voice... but afaik nothing for adsl.
-
/snip /
- is this at 300 kHz or at a weighted average of the frequencies in use or.....
If longer lines have fatter copper to compensate (on voice) then the dB/km will be lower in any case, I've seen numbers like 62 dB over 10 km on a thick overhead line.
This is something that did actually strike me a couple of days ago.. I still dont know where grapevine has got the figure from that lines should be no more than 10dB per mile... but something that did occur to me is that as we all know voice is transmitted at a lower frequency.. and as such attenuation for the voice frequencies would be much lower than they would be for adsl.
I believe that BT do have regulations as for the provision of voice, but afaik there inst anything regarding the specifics of adsl.
Therefore if this document of "proof" exists somewhere and does relate to voice frequencies, then surely we cant use figures obtained from an adsl modem/router as a comparison measure.
It is a possibility that many,many years ago when laying lines for voice only that they did perhaps use higher gauge copper wire to compensate very long lines and to bring things up to a standard for voice.
I also have grave concerns about people using adsl routers to measure attenuation.. different routers report different figures and its something Ive been stating elsewhere for several years.
Fore eg
A sample of stats on my own line using different routers.
As the crow flies 160m
Actual cable length 600m
7dB - BT voyager 2100
8dB - Speedtouch 576
9dB - SAR110 (VIK-2.1.040311a)
12dB - Netgear DG834G
21dB - Solwise SAR110 (original firmware)
-
How could there be regulations for adsl? The wires were designed for voice and adsl over them is a by-product. You can't put regulations on wires that relate to a system that they were never designed for. That just means replacing the old system with a completely new system designed for adsl (ie thicker wires) to meet the new regulations.
-
That just means replacing the old system with a completely new system designed for adsl (ie thicker wires) to meet the new regulations.
I seem to remember when BT was privatised that the biggest asset they owned, by far, was the copper cable network. Anyone with an ounce of commonsense would realise that it is not a practical proposition to replace that network over a short timescale.
My feeling is that another transmission technology will be developed long before the copper wire network gets replaced.
-
Stetecio has got it spot on. the cost of providing new cableing for several customers be it for 10,20,50+ plus lines is very expensive, lately much more so with the incredible increase in costs of raw materials. The BT group has to work with a net work originaly designed for voice not data. Doubleing up on pairs in therory can improve a signal, this cannot be done in the network for a number of good reasons. A cable directly feeding one property (the wire from a pole/distribution point to the property) this may be done, but has as much chance or creating problems for broadband (and other issues too long to go into)
The last time I looked from memory the OfCom licence requirement was the ability to provide a data connction to at least 14.5kbps. 64kbps with a mid band connection if avalable on a relavent contract with your telecom provider. Now with the advent of broadband 0.5mb is the minimum reasonable speed, that is assuming that the total cableing is capable of supporting this (which sometimes on the odd occasion isn't possible within reason). I've never heard of any target db loss per set distance so stating any more than 10db/km is substandard at 300khz is a fiction in my opinion.
From my experience any bt broadband engineer would do their best to get the optimum service to any customer. Long line issues you do your best to look at the optimium cableing, sockets etc to get a fair and reliable service.
I thought this string started as a forum to gauge a rough ball park guide. I've never worked to an expected figure per kilometer because of the variables I come across. Just I know typicaly beyond 7-7.5km the line lenght "may" become a real issue. this string is becoming a bit ugly me thinks
-
A thought has just occurred to me; with the current astronomical price of scrap copper would it not make economic sense to to replace it with fibre optic over a large part of the network?
-
Something called 21st century network (or 21CN)
It's happening now & the budget is massive. although there's a lot of network to get through so it will take time
-
>Something called 21st century network (or 21CN)<
There you are, graevine1; BT is addressing the problem, so we won't need to hear anything more from you on this subject.
-
Doubling-up wires is just as likely to degrade the ADSL signal as to improve it, because the two pairs will have slightly different transmission delays. In the worst case this difference would be such that the two signals would arrive at the end in opposite phase and cancel each other out.
-
Doubling-up wires is just as likely to degrade the ADSL signal as to improve it, because the two pairs will have slightly different transmission delays. In the worst case this difference would be such that the two signals would arrive at the end in opposite phase and cancel each other out.
Thanks roseway and Ezzer for clarifying that. It was just something I have heard done and wondered if it actually did make any improvement.
With regards to 21CN, the first stage is only a converged core network to replace existing seperate platforms.
Unfortunately 21CN in its first phase will not be FTTC or FTTP, although I agree, with the costs of copper and the theft of copper becoming such a cost, fibre suddenly looks like an attractive option.
Apparently Openreach are carrying out consultations and have conducted trials and all that so we will see what happens.
I think a simple way to look at things at present is that 21CN is BT wholesale's realm, with most of the work being done by BT Operate, which makes it a separate infrastructure to the access network, which is down to Openreach to develop as they see fit.
-
The fact remains that customers who have a line attenuation greater than 10db per kilometer have a line requiring maintenance or an "uplift".
Another BT official "price point" was the original 3.5 km limit for 2M at 41 dB = 11.7 dB/km.
There is a mechanism built into MaxDSL to counter deterioration in the local loop - during the 10 day "training" period about the only thing that happens different to any other time is that the lowest sync rate achieved is clocked as the "Maximum Stable Rate" or MSR.
If the sync speed subsequently deteriorates below 70% of the MSR (the 70% being the Fault Threshold Rate or FTR) then the ISP can raise a fault due to reduced service speed.
So there is a mechanism in place to counter any tendency to long term neglect of the loop. This wouldn't address any "Day 1" high attenuation issues though.
FWIW the 21CN project doesn't address the local loop or its quality, it starts at the exchange and simplifies systems upstream of there.
Phil
-
Some info :-
1. Enhanced spec Openreach tie cable for LLU to approach Cat 5 performance - "Attenuation : 14 dB/km @ 300 kHz, 31 dB/km @ 1 MHz, 103 dB/km @ 10 MHz" from here. (http://www.openreach.co.uk/orpg/products/llu/planbuild/downloads/comingling/NEXT_Tie_Cable_product_description_issue5%20_R_.pdf)
2. CW1308 doesn't define the attenuation spec, but a cable supplier (http://www.nexans.sk/pdf/nex/koax/nfmult/nex_cw1308.pdf) quotes typical values of 45 dB/km at 1MHz which is 45% more loss than the above spec.
If the 45% factor also applies at 300 kHz (bit of a guess, admittedly) then CW1308 would be 10 dB/km. Back to the "5.5 km = 55 dB" BT figure.
[Admin edit: "If" the 45% factor applied at 300 kHz it would be 21 dB/km*]
If, as I suspect, routers report a weighted average attenuation across the frequency band then the fact that attenuation doubles over the ADSL downstream frequency range is important. A higher speed line using all the frequencies will show higher attenuation per km than a long low speed line struggling with only low frequencies. The bits/bin allocation of routers can also vary, some filling up from the bottom of the frequency spectrum and others spreading across the whole available spectrum.
Phil
[Admin Note]
It has come to our attention persons have been pointing to this post as supposed proof that attenuation on CW1308 cable should be 10dB/km. Hence the need for the edit.
Since CW1308 has more loss than Enhanced spec, this additional loss should be added on to the original figure, not subtracted.
The rest of the post is not disagreed with and is something already stated/discussed elsewhere.
-
Thank you Phil, the weakest link in the cable feed is the customers end ie the CW1308 which would be approx 10 dB/km. That cable being the lowest cross-section of copper.
I received confirmation of the 10db per kilometer from the Chairmans office just over a year ago. So I and all others should hold them to that.!!!!! Anyone who says any figure greater is not in need of maintenance or "uplift" is simply missleading the service receiving customer
If any contract has caused a lower grade cable -v- distance to be inserted into a length then the words are -tough- it just requires an "uplift" ie replacing so as to return an acceptable service to the customers. The problem that is occuring is short lengths being 'let in' to repair faulty runs in cable lengths, hence adding extra joints etc with their attendant losses. Another is "E" side cables (those from the exchange to the CAB) having had water ingress from a faulty joint or for whatever reason not being able to be ever returned to original specification by the passing of the compressed drying air into the length from the exchange.
I get extremly annoyed when individuals profer their opinions (without any formal training or qualifications) and appear to say accept 15 or even 20db per kilometer line attenuation, with its intendant lower SNR/ crosstalk,'cutting into busy' liability. It does however require certified, calibrated test equipment to establish the figures as routers only give a guide figure.
Its true openreach inherited a "national asset" and as such must maintain it to an acceptable standard
Astral states "I seem to remember when BT was privatised that the biggest asset they owned, by far, was the copper cable network." Is anyone now suggesting we should accept a degrading local network.?
Astral then states "Anyone with an ounce of commonsense would realise that it is not a practical proposition to replace that network over a short timescale." But we the customers should ensure that no one allows the twisted copper pair network to deteriorate, and that in many instances is the situation today.
We may see FTTC fiber to the cab, but I doubt we will see for many years, other than to 'new build' and then not all, FTTH fiber to the home, IF EVER !!!!! to other than a very few!
-
I get extremly annoyed when individuals profer their opinions (without any formal training or qualifications)
And I get extremely annoyed when you make such unsupported assumptions about people who disagree with you.
This thread has become no more than a slanging match and is going round in circles now. I'm locking it.