A Broadband and ADSL forum. BroadbanterBanter

Welcome to BroadbanterBanter.

You are currently viewing as a guest which gives you limited access to view most discussions and other FREE features. By joining our free community you will have access to post topics, communicate privately with other members (PM), respond to polls, upload your own photos and access many other special features. Registration is fast, simple and absolutely free so please, join our community today.

Go Back   Home » BroadbanterBanter forum » Newsgroup Discussions » uk.telecom.broadband (UK broadband)
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

uk.telecom.broadband (UK broadband) (uk.telecom.broadband) Discussion of broadband services, technology and equipment as provided in the UK. Discussions of specific services based on ADSL, cable modems or other broadband technology are also on-topic. Advertising is not allowed.

Estimating ADSL & VDSL Speeds From Given Distances



 
 
Thread Tools Display Modes
  #1  
Old November 16th 16, 03:33 PM posted to uk.telecom.broadband
Java Jive
external usenet poster
 
Posts: 378
Default Estimating ADSL & VDSL Speeds From Given Distances

I have been informed that our locality is not going to get FTTC during
the imminent local exchange rollout of it, so I am putting together a
case for bringing FTTC to us.

We have 59 premises, about 12% of the total for the local exchange, in
a reasonably small geographical area. Currently, we are all connected
by exchange only lines, which would have to be rerouted via one or
more cabinets. I have drawn up a map and a spreadsheet table of all
the addresses and calculated their roadside, cabled distances from the
exchange and proposed cabinet site(s).

The average distance from the exchange is 8.81km, the furthest is
12.2km. All online ADSL checkers I've seen give either rather vague
estimates of 1Mbps speed for all the properties, or no result at all,
whereas in reality the best speed I've heard of is 2.5Mbps, while most
seem to get less than 1Mbps, and the furthest houses cannot get
landline broadband at all.

With one cabinet, the average distance from the cabinet would be
2.22km, the furthest would be 4.28km. With three cabinets - which
sounds unrealistic but would save BT relaying much of the local
cabling, and therefore might actually be cheaper - the average would
be 1.17km and the furthest 3.08km.

Does anyone know of formulae that I can use to give more useful
estimates of speeds for ADSL, and estimate the speeds available to
each property via VDSL - preferably in each case something I can put
in a spreadsheet formula?
--
================================================== ======
Please always reply to ng as the email in this post's
header does not exist. Or use a contact address at:
http://www.macfh.co.uk/JavaJive/JavaJive.html
http://www.macfh.co.uk/Macfarlane/Macfarlane.html
  #2  
Old November 16th 16, 04:03 PM posted to uk.telecom.broadband
Graham J[_2_]
external usenet poster
 
Posts: 723
Default Estimating ADSL & VDSL Speeds From Given Distances

Java Jive wrote:
I have been informed that our locality is not going to get FTTC during
the imminent local exchange rollout of it, so I am putting together a
case for bringing FTTC to us.

We have 59 premises, about 12% of the total for the local exchange, in
a reasonably small geographical area. Currently, we are all connected
by exchange only lines, which would have to be rerouted via one or
more cabinets. I have drawn up a map and a spreadsheet table of all
the addresses and calculated their roadside, cabled distances from the
exchange and proposed cabinet site(s).

The average distance from the exchange is 8.81km, the furthest is
12.2km. All online ADSL checkers I've seen give either rather vague
estimates of 1Mbps speed for all the properties, or no result at all,
whereas in reality the best speed I've heard of is 2.5Mbps, while most
seem to get less than 1Mbps, and the furthest houses cannot get
landline broadband at all.


That's fair comment ...

With one cabinet, the average distance from the cabinet would be
2.22km, the furthest would be 4.28km. With three cabinets - which
sounds unrealistic but would save BT relaying much of the local
cabling, and therefore might actually be cheaper - the average would
be 1.17km and the furthest 3.08km.

Does anyone know of formulae that I can use to give more useful
estimates of speeds for ADSL, and estimate the speeds available to
each property via VDSL - preferably in each case something I can put
in a spreadsheet formula?


Google for vdsl speed vs distance ...

http://www.increasebroadbandspeed.co...ainst-distance
- has a nice graph.

Or: vdsl speed calculator ...

http://www.speedguide.net/dsl_speed_calc.php

Trying a few values in this suggests the speed vs distance graph is not
a straight line

http://forum.kitz.co.uk/index.php?topic=10566.0 - this gives the basis
for calculation.

But for your 59 possible users you could plug in some estimates based on
approsimate distance.

But this is all probably of no real relevance.

-------------

To whom will you be submitting your case?

In a region I know of (near Holbeach St Matthew, nort-heast of Spalding)
BT are rolling out FTTP. There's no commercial need - probably there
are well under 60 houses, and it's not being extended to nearby farms
where it might be of some real use. So the only reasons I can think
it's being done is that:

1) it will demonstrate there's no commercial case, so BT will not push
it into other areas;

2) it allows BT to test all aspects of the roll-out in an area where
failure won't be noticed.

I can see that there's some point in the latter reason. So you might
adapt that to your case.

We know what you want is technically feasible. Evidently BT now have
"Remote nodes" which can be installed on eexisting poles. This avoids
rerouting exising cables through newly installed green cabinets. See:

http://www.ispreview.co.uk/index.php...d-network.html

http://www.ispreview.co.uk/index.php...t-q4-2014.html

Do let us know how you get on.

--
Graham J







  #3  
Old November 16th 16, 04:25 PM posted to uk.telecom.broadband
Andy Burns[_5_]
external usenet poster
 
Posts: 315
Default Estimating ADSL & VDSL Speeds From Given Distances

Java Jive wrote:

I have been informed that our locality is not going to get FTTC during
the imminent local exchange rollout of it, so I am putting together a
case for bringing FTTC to us.


Not to discourage you but ...

I was speaking to friend last night who left school at 16 to apprentice
with BT and 35 years later still works "on the tools" for them. He says
he often encounters situations where e.g. there's a 100 or 200 pair
cable into a village that gets the VDSL treatment, and then an extra 20
pair cable, serving a handful of houses that is left behind on ADSL,
this can be awkward when neighbours are on different cables.

He's been told off for explaining to frustrated users that all it would
take is the will to install an extra few feet of cable from a pole
carrying the smaller cable to the VDSL cabinet and back, and encouraging
the locals to encourage their county broadband scheme to fund this work.

The planners have planned it and they don't want "helpful" suggestions
like that after it's been installed!



As for your spreadsheet, this page has general graphs and tables of
speeds by distance, obviously it's all variable and depends on cable
quality as well as distance.

http://increasebroadbandspeed.co.uk/2013/chart-bt-fttc-vdsl2-speed-against-distance

  #4  
Old November 16th 16, 08:42 PM posted to uk.telecom.broadband
Java Jive
external usenet poster
 
Posts: 378
Default Estimating ADSL & VDSL Speeds From Given Distances

Thanks to both for your replies, but ...

On Wed, 16 Nov 2016 15:03:57 +0000, Graham J
wrote:

Google for vdsl speed vs distance ...

http://www.increasebroadbandspeed.co...ainst-distance
- has a nice graph.


Yes, I'd found that already ...

Or: vdsl speed calculator ...

http://www.speedguide.net/dsl_speed_calc.php


.... and kitz has something pretty similar. The problem with all of
them is that they are designed for use by people with a connection who
can read their Attenuation directly from their router, whereas I need
to calculate from distance alone, and that is where the
errors/uncertainty are worst. For example, my attenuation as measured
by my router is 63dB, which leads such calculators, working at a rate
of 13.81dB/km, to assume I'm about 4.5km from the exchange, whereas
I'm just under twice that!

Trying a few values in this suggests the speed vs distance graph is not
a straight line


No, I wouldn't expect to be, even as it tails off. Although I haven't
the background knowledge to do a detailed analysis, based on more
general scientific knowledge, my first guess would be that the
relationship is hyperbolic, where speed is proportional to 1/distance.

http://forum.kitz.co.uk/index.php?topic=10566.0 - this gives the basis
for calculation.


Yes, I'd found that already, and it works from attenuation as
described above. One of the pages linked thence gives more of the
background of the kitz calculator, and includes a spreadsheet which
does the same 'calculation'. However, examination of the spreadsheet
code shows that all it is doing is straight-line interpolating between
points on ...

http://www.internode.on.net/images/c...2-distance.jpg
... which now gives a 404, but originally was almost certainly the
graph at ...
http://www.internode.on.net/resident...d/performance/

The problem with all such sources is that I can't 'interpolate' beyond
the end of the original data!

To whom will you be submitting your case?


Probably to funding bodies such as Highlands & Islands Enterprise,
MPs, MSPs, etc.

In a region I know of (near Holbeach St Matthew, nort-heast of Spalding)
BT are rolling out FTTP. There's no commercial need - probably there
are well under 60 houses, and it's not being extended to nearby farms
where it might be of some real use. So the only reasons I can think
it's being done is that:

1) it will demonstrate there's no commercial case, so BT will not push
it into other areas;

2) it allows BT to test all aspects of the roll-out in an area where
failure won't be noticed.

I can see that there's some point in the latter reason. So you might
adapt that to your case.


That does seem rather strange.

We know what you want is technically feasible. Evidently BT now have
"Remote nodes" which can be installed on eexisting poles. This avoids
rerouting exising cables through newly installed green cabinets. See:

http://www.ispreview.co.uk/index.php...d-network.html

http://www.ispreview.co.uk/index.php...t-q4-2014.html

Do let us know how you get on.


Thanks for the links, and I'll keep you up-to-date

On Wed, 16 Nov 2016 15:25:57 +0000, Andy Burns
wrote:

As for your spreadsheet, this page has general graphs and tables of
speeds by distance, obviously it's all variable and depends on cable
quality as well as distance.

http://increasebroadbandspeed.co.uk/2013/chart-bt-fttc-vdsl2-speed-against-distance


Yes, thanks, as discussed above, I'd already found that page, but I
was hoping for something I could just put in a formula, rather than
manually reading 59 values off a graph!
--
================================================== ======
Please always reply to ng as the email in this post's
header does not exist. Or use a contact address at:
http://www.macfh.co.uk/JavaJive/JavaJive.html
http://www.macfh.co.uk/Macfarlane/Macfarlane.html
  #5  
Old November 17th 16, 06:11 PM posted to uk.telecom.broadband
Andrew Benham
external usenet poster
 
Posts: 284
Default Estimating ADSL & VDSL Speeds From Given Distances

On Wed, 16 Nov 2016 15:25:57 +0000, Andy Burns wrote:

Java Jive wrote:

I have been informed that our locality is not going to get FTTC during
the imminent local exchange rollout of it, so I am putting together a
case for bringing FTTC to us.


Not to discourage you but ...

I was speaking to friend last night who left school at 16 to apprentice
with BT and 35 years later still works "on the tools" for them. He says
he often encounters situations where e.g. there's a 100 or 200 pair
cable into a village that gets the VDSL treatment, and then an extra 20
pair cable, serving a handful of houses that is left behind on ADSL,
this can be awkward when neighbours are on different cables.

He's been told off for explaining to frustrated users that all it would
take is the will to install an extra few feet of cable from a pole
carrying the smaller cable to the VDSL cabinet and back, and encouraging
the locals to encourage their county broadband scheme to fund this work.

The planners have planned it and they don't want "helpful" suggestions
like that after it's been installed!


BT certainly don't want to alter their wiring just because developments
have taken place since installation.
At the top of my road is a parade of shops dating from the 1930s, with
a few older cottages opposite them. The shops are fed from underground,
the cottages are fed from a the last of a line of poles which are by the
service road to the front of the shops - from the same cabinet. I guess
the cottages were connected to the phone network before the shops were
built.
Someone drove their car into one of the poles and broke it off at bumper
height, so Openreach came to replace the pole and then rewire the line
of poles. Would it have been sensible to ditch the line of poles and
just feed the cottages from one pole wired to the same underground feed
as the shops ? Perhaps, but it would have meant changing a diagram, so
we'll just rewire it as it was before.

  #6  
Old November 17th 16, 08:53 PM posted to uk.telecom.broadband
Dick
external usenet poster
 
Posts: 104
Default Estimating ADSL & VDSL Speeds From Given Distances

On 16-Nov-16 7:42 PM, Java Jive wrote:


errors/uncertainty are worst. For example, my attenuation as measured
by my router is 63dB, which leads such calculators, working at a rate
of 13.81dB/km, to assume I'm about 4.5km from the exchange, whereas
I'm just under twice that!

Many routers won't measure below 63dB and just report 63dB regardless of
actual value.
  #7  
Old November 17th 16, 10:00 PM posted to uk.telecom.broadband
Java Jive
external usenet poster
 
Posts: 378
Default Estimating ADSL & VDSL Speeds From Given Distances

On Thu, 17 Nov 2016 19:53:42 +0000, Dick wrote:

On 16-Nov-16 7:42 PM, Java Jive wrote:


errors/uncertainty are worst. For example, my attenuation as measured
by my router is 63dB, which leads such calculators, working at a rate
of 13.81dB/km, to assume I'm about 4.5km from the exchange, whereas
I'm just under twice that!


Many routers won't measure below 63dB and just report 63dB regardless of
actual value.


If that's true, then that makes all these so-called 'calculators' even
more worthless!

As far as the ADSL calculations go, I've given up on them, because I
get the sense that there is insufficient accuracy to justify further
effort. For the projected VDSL figures, I've made some
reasonable-looking extrapolations of the curves based on a hyperbolic
law.
--
================================================== ======
Please always reply to ng as the email in this post's
header does not exist. Or use a contact address at:
http://www.macfh.co.uk/JavaJive/JavaJive.html
http://www.macfh.co.uk/Macfarlane/Macfarlane.html
  #8  
Old November 18th 16, 10:05 AM posted to uk.telecom.broadband
Martin Brown
external usenet poster
 
Posts: 343
Default Estimating ADSL & VDSL Speeds From Given Distances

On 17/11/2016 21:00, Java Jive wrote:
On Thu, 17 Nov 2016 19:53:42 +0000, Dick wrote:

On 16-Nov-16 7:42 PM, Java Jive wrote:


errors/uncertainty are worst. For example, my attenuation as measured
by my router is 63dB, which leads such calculators, working at a rate
of 13.81dB/km, to assume I'm about 4.5km from the exchange, whereas
I'm just under twice that!


Many routers won't measure below 63dB and just report 63dB regardless of
actual value.


If that's true, then that makes all these so-called 'calculators' even
more worthless!


The problem is more with real world old cables being less than ideal. In
particular the number of junctions between you and the exchange and the
state of corrosion affecting them. If there is any aluminium in the path
at all then all bets are off since you get rectification of the ADSL
signal and chaos ensues. BT insist that no such cables exist ;-)

Overhead cables sometimes develop fractures too.

As far as the ADSL calculations go, I've given up on them, because I
get the sense that there is insufficient accuracy to justify further
effort. For the projected VDSL figures, I've made some
reasonable-looking extrapolations of the curves based on a hyperbolic
law.


This page details the attenuation calculation which roughly for ADSL is
~4MHz and an attenuation rate of 14dB/km with ranges upto 5km and for
VDSL ~30MHz attenuation rate 38dB/km with a useful range of =2km. No
point in paying for VDSL and getting a performance that is worse than
cheaper ADSL because of line attenutation at the higher frequency!

This Kitz thread deals with what you are trying to compute:
http://forum.kitz.co.uk/index.php?topic=10566.0

Has a better set of references
http://forum.kitz.co.uk/index.php/to...html#msg209077

It isn't a law so much as a rough heuristic that is approximately right
down to around the 1Mbps and 65dB mark and cable lengths 5miles. Beyond
that and you are in the lap of the Gods. The maximum length for getting
a benefit from VDSL is even shorter. My ADSL service is faster than a
VDSL service would be over the same line (if one were offered).

Some routers can be confirgured to work more reliably with really bad
lines - the worst rural lines I have ever seen included a segment of
aluminium and could only support 256kbps sync despite plausible
attenuation and noise margins. The joints act like poor diodes.

Basically if the line is too long you are stuffed. The only serious
option is one of the rural microwave link companies or a 3G Mifi with a
serious directional antenna attached.

From what I can tell round here fibre to remote node has been kicked
into touch since there is no way to power the remote node cost
effectively so without huge subsidies it doesn't get considered. There
are a couple of experimental sites like Ulshaw in North Yorkshire.

http://www.ispreview.co.uk/index.php...yorkshire.html

Nice technology shame about the unrealistic cost per household.

--
Regards,
Martin Brown
  #9  
Old November 18th 16, 02:06 PM posted to uk.telecom.broadband
Java Jive
external usenet poster
 
Posts: 378
Default Estimating ADSL & VDSL Speeds From Given Distances

On Fri, 18 Nov 2016 09:05:15 +0000, Martin Brown
wrote:

On 17/11/2016 21:00, Java Jive wrote:
On Thu, 17 Nov 2016 19:53:42 +0000, Dick wrote:

On 16-Nov-16 7:42 PM, Java Jive wrote:

errors/uncertainty are worst. For example, my attenuation as measured
by my router is 63dB, which leads such calculators, working at a rate
of 13.81dB/km, to assume I'm about 4.5km from the exchange, whereas
I'm just under twice that!

Many routers won't measure below 63dB and just report 63dB regardless of
actual value.


If that's true, then that makes all these so-called 'calculators' even
more worthless!


The problem is more with real world old cables being less than ideal.


That is certainly a factor, but if what Dick says is true, that seems
to me to be a far greater problem.

In
particular the number of junctions between you and the exchange and the
state of corrosion affecting them. If there is any aluminium in the path
at all then all bets are off since you get rectification of the ADSL
signal and chaos ensues. BT insist that no such cables exist ;-)

Overhead cables sometimes develop fractures too.


A few drop wires remain, but most of the cables around here are
buried, albeit sometimes not very well, as I've posted before.

As far as the ADSL calculations go, I've given up on them, because I
get the sense that there is insufficient accuracy to justify further
effort. For the projected VDSL figures, I've made some
reasonable-looking extrapolations of the curves based on a hyperbolic
law.


This page details the attenuation calculation which roughly for ADSL is
~4MHz and an attenuation rate of 14dB/km with ranges upto 5km and for
VDSL ~30MHz attenuation rate 38dB/km with a useful range of =2km. No
point in paying for VDSL and getting a performance that is worse than
cheaper ADSL because of line attenutation at the higher frequency!

This Kitz thread deals with what you are trying to compute:
http://forum.kitz.co.uk/index.php?topic=10566.0


Yes, as indicated elsewhere in the thread, I've already analysed that
page. In that same a respondent claims an error of 42%, for me it's
50%. Elsewhere I've seen a Canadian discussion where the first poster
pulls it to shreds, saying that it seemed to have no basis in either
theory or experiment, and inviting people to give evidence supporting
it. In support of it, there was no shortage of verbage, but no actual
evidence.

Has a better set of references
http://forum.kitz.co.uk/index.php/to...html#msg209077


Again, I'd already found this post, and the first few references now
lead nowhere, and the remainder are either irrelevant to my needs or
too technical to be practically useful.

It isn't a law so much as a rough heuristic that is approximately right
down to around the 1Mbps and 65dB mark and cable lengths 5miles.


The trouble is, it's based on Australian data where the cabling may
have different characteristics, the climate is radically different,
and the standard of workmanship may be completely different - I
would hope it would be better, but the fact that the calculator under
rather than over estimates my line length suggests that things are
even worse down under than here!

Beyond
that and you are in the lap of the Gods.


Yet practical information must exist somewhere, we are not the only
community to be based 6 or more km from an exchange.

The maximum length for getting
a benefit from VDSL is even shorter. My ADSL service is faster than a
VDSL service would be over the same line (if one were offered).


Yes, I know. It's a question of managing expectations. My intention
is that locals should not think of FTTC as supplying the sort of
speeds it would in a town, but rather the sort of speeds that ADSL
would in a town. It's a means of getting us speeds that we can
actually use in everyday life, rather than waiting for a minute or so
for the home page of a big firm to load.

Basically if the line is too long you are stuffed. The only serious
option is one of the rural microwave link companies


Yes there are examples here ...

http://www.hie.co.uk/community-suppo...d/default.html

.... but the blurb is high on vacuous news spin and low on important
factual detail, like:

:-{ How much are ongoing costs met by grant funding that could
disappear if the political climate changes?

:-{ How much does each subscriber pay:
Up front?
Monthly?

:-{ Are there any download caps, or other like restrictions, and
if so, what are they?

or a 3G Mifi with a
serious directional antenna attached.


Of the two local mobile networks available, one is still on 2G, never
mind 3G, and the EE one which is on 3G is even worse than landline
broadband. To use my mobile, I have to stand at a particular window
facing the mast, and even then, if it is raining heavily, the call may
drop every few minutes or so, necessitating redialling, so, if you are
dialling a helpdesk to report a fault, you get just to the point where
you have authenticated yourself to them and are explaining the fault,
when the call drops and you have to go to the back of the call queue
and start all over again. Using mobile data, a speed check app
registers the speed as 0 and the John Lewis home page takes five to
ten minutes to load.

From what I can tell round here fibre to remote node has been kicked
into touch since there is no way to power the remote node cost
effectively so without huge subsidies it doesn't get considered. There
are a couple of experimental sites like Ulshaw in North Yorkshire.

http://www.ispreview.co.uk/index.php...yorkshire.html

Nice technology shame about the unrealistic cost per household.


How much power does it require to power each customer's part of a
remote node. Surely, with modern miniaturisation etc, it should be
possible to power it from the normal operating voltage of a POTS line?
--
================================================== ======
Please always reply to ng as the email in this post's
header does not exist. Or use a contact address at:
http://www.macfh.co.uk/JavaJive/JavaJive.html
http://www.macfh.co.uk/Macfarlane/Macfarlane.html
  #10  
Old November 18th 16, 04:12 PM posted to uk.telecom.broadband
Graham J[_2_]
external usenet poster
 
Posts: 723
Default Estimating ADSL & VDSL Speeds From Given Distances

Java Jive wrote:
[snip]

I've no confidence that your sterling efforts will make one iota of
difference to BT.

What will work is point-to-multipoint Wifi - but there are several
practical difficulties to overcome.

1) You need a trustworthy householder living close to the exchange, who
gets reliable FTTC; and who is prepared to share his connection or allow
you to install another in his premises.

2) All your 59 users need a line-of-sight view of that householder's
property.

You then put a 5G WiFi access point on the first householder's roof,
aimed in the general direction of your 59 users. You might require more
than one such access point, aimed in different directions.

Each of your 59 householders requires a similar access point, configured
as a client. These then connect to the user's computers via Ethernet.
The Ethernet connection could support another access point to serve
iThings or other Wifi-only devices inside the property.

I've used Engenius access points for this in the past, over distances up
to about 5km - but their specs say that 30km is achievable. However
this is in the fenland area north of Spalding, so the civil engineering
in getting the access points high enough to be visible to each other is
compatible with what can be achived by a satellite dish contractor. The
access points themselves usually cost under 100 - but of course this us
the least of the possible expenses - installation could be much more.

This may all look very difficult - but you can manage it all yourself
(or within the local community) so you are not beholden to BT !!!

--
Graham J
 




Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Do slow VDSL lines use ADSL profiles ? Andrew Benham uk.telecom.broadband (UK broadband) 2 October 25th 16 01:53 AM
Upgrade from ADSL to VDSL - how long until download speed improves? NY uk.telecom.broadband (UK broadband) 6 March 29th 16 12:42 PM
Ceasing VDSL and providing ADSL instead - what's involved by BT? NY uk.telecom.broadband (UK broadband) 35 July 7th 14 06:04 PM
ADSL MAX Speeds Paul uk.telecom.broadband (UK broadband) 24 January 26th 07 01:09 PM
ADSL speeds Grumps uk.telecom.broadband (UK broadband) 10 January 14th 06 09:20 PM


All times are GMT +1. The time now is 08:59 PM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.Content Relevant URLs by vBSEO 2.4.0
Copyright 2004-2018 BroadbanterBanter.
The comments are property of their posters.