» ISP News » 

BT Take Flak for Missed Appointments, FTTP and UK Broadband Maps

Wednesday, March 16th, 2016 (9:39 am) - Score 1,913

The on-going cross party Culture, Media and Sport Committee inquiry into UK digital infrastructure produced another interesting evidence session yesterday after BT and sibling Openreach were both grilled over the country’s broadband quality, 1,000 missed appointments per week and USO plans etc.

Much of the debate was familiar fare, although the session did produce a few interesting bits of information. For example, BTOpenreach’s director, Kim Mears, admitted that their “biggest failing” was missing 1,000 appointments every week (here): “When we get it [installs etc.] wrong, we repeat the problem by missing appointments. We repeat the problem due to systems, process, ways of working – it could be a multitude of issues,” said Mears.

Mears was flanked by BT Group’s CEO, Gavin Patterson, who chimed in to add, “We’ve got to be faster in repairing service. BT Consumer, Sky or TalkTalk have to work together because the problem could be in the home, in their systems, or down to us … we will focus more on service going forward.”

However ISPs sometimes complain that they are often forced to foot the bill for Openreach to find and repair faults in its own network (e.g. disputes over SFI charges). Similarly if a problem on OR’s network causes an ISP’s customers to be angry and leave or complain (e.g. loss of service due to OR fault) then the ISP takes the hit or possible ADR complaint handler fees, while Openreach as the supplier avoids this.

Similarly Ofcom’s Broadband Speed Code of Practice and the ASA’s linked advertising rules also only hit the ISP, not Openreach and the latter may sometimes be to blame for related performance woes. In other cases you can end up with a lot of oddball mistakes, such as how BT recently insisted that AAISP book an SFI engineer to test for a copper line fault, even though the faulty service was a pure fibre option FTTP line (here). Certainly some room for improvement.

Elsewhere BT suggested that the proposed 10Mbps Universal Service Obligation (USO) could be delivered by a mix of existing “fibre broadband” (FTTC / VDSL, G.Fast and FTTP) solutions to reach 99% of the UK for just under £2 billion (alternative methods would be required for the final 1%). But this still needs more explanation as it’s unclear whether they’ve included the existing BDUK “superfast” (24Mbps+) coverage expectation of 95-96% by 2017/18.

The alternative for BT would involve a much cheaper approach, possibly by using an extended range enhancement of VDSL (FTTC) broadband (details), fixed wireless networks (e.g. 4G) and or ADSL2+ from the street cabinet (here) etc. But the details remain unclear and are subject to the outcome of future consultation and debate. BT talked a bit more about all this last year (here).

Meanwhile the Connecting Devon and Somerset (CDS) project, which last year saw plans for a contract extension with BT collapse due to disagreements over coverage and time-scale (here), also submitted some new written evidence to the DCMS inquiry and used the opportunity to criticise BTOpenreach’s lack of transparency over coverage mapping and confusion in its local FTTP roll-out.

CDS Statement – Mapping

Since the start of the contract, BT have been reluctant for us, as a partnership, to provide information to communities about when they can expect to see superfast or improved broadband. The position they have taken is that communities should only be informed after structures have gone live.

In July 2014, The Secretary of State for Culture Media and Sport wrote to all local bodies, instructing them to publish rollout maps. Following this, Connecting Devon and Somerset published its own map and postcode checker in order to better inform communities of the intended rollout. The map was driven by data designated as CSI [Confidential Sensitive Information] in the contract and at the time this was published at our own risk.

We also do not have agreement to show Superfast areas on our deployment map, only fibre delivery or NGA coverage. This means that homes able to connect to a fibre enabled cabinet are shown as receiving services – ‘green’ regardless of whether they do receive any form of speed improvement or superfast capability. Needless to say this causes confusion and anger to many residents. We are in the process of reviewing how to improve this information on our map, but any revised mapping that shows superfast rather than fibre connectivity would be at our own risk given that this information is also regarded as CSI by BT.

CDS Statement – FTTP Deployment

Openreach have encountered a number of engineering problems when deploying Fibre to the Premise (FttP) in rural areas, leading to cost overruns in some cases. As a result of this CDS and BT agreed to put the FttP programme on hold pending a review to examine the common issues encountered by Openreach and where possible to discuss alternative, more cost effective solutions. This is still ongoing.

Unfortunately the build of FttP is quite visible in many rural areas and this has caused uncertainty and frustration amongst those communities who can see fibre coming towards them but have no information on completion dates. In some cases Openreach engineers have spoken to individuals and told them they would be getting FttP services.

BT’s unwillingness, however, to let communities know which technology they are expecting has made it impossible to be open and transparent with those who could be in scope for FttP. Whilst we have provided open information in response to MP letters and FoI, this has been at risk of BT claiming that we are in breach of our contract with them.

The mapping problem is a long running one that dates back to the early Phase 1 Broadband Delivery UK contracts and still has not been entirely resolved (this caused a lot of altnet schemes to stall because they couldn’t be sure where Openreach’s network would deploy or not), although it should be said that some local authorities are notably better than others at both communicating their deployment plans and through the delivery of superior maps.

On top of that not even Openreach can be 100% sure of the final deployment areas until their engineers actually get on the ground and start work, which is where costly problems with blocked ducts or other issues may finally become apparent (this could change the roll-out plan). On the flip side they are sometimes able to cover more areas than expected.

As for the FTTP challenge, it’s well known that BT scaled back their original commercial deployment plans in order to save money and focus on expanding the reach of cheaper / slower speed FTTC services that were faster to roll-out and install into homes. However it’s interesting to see it confirmed that some of the same cost issues have also impacted BDUK contracts like CDS.

The inquiry is expected to publish its outcome within the next few months.

Share with Twitter
Share with Linkedin
Share with Facebook
Share with Reddit
Share with Pinterest
Tags: , , , , , ,
By Mark Jackson
Mark is a professional technology writer, IT consultant and computer engineer from Dorset (England), he also founded ISPreview in 1999 and enjoys analysing the latest telecoms and broadband developments. Find me on Twitter, , Facebook and Linkedin.
Leave a Comment
45 Responses
  1. Kits says:

    I do think it is time that if someones copper cable is over 10/15 years old that BT cannot charge for SFI appointments. The joints become corroded and all BT do is lower the speeds they say you can get instead of renewing the cable.

    The copper lines are not something that can go for ever I have watched my speed drop slowly over the years from max fttc connection to below 60. I have lived in this home for 30 years and the copper down the front of my home has never been changed they didn’t do it when I paid them for a new line in couldn’t even get the engineer to put a new master socket in he never touched it just new line from pole to my home because BT had removed it 5 years earlier.

    This is not acceptable when the cable is what brings the broadband into the home. Regular replacement due to the way copper can break is the only way if BT wish to stay with copper.

    1. Ignition says:

      Sure your speeds haven’t dropped just from crosstalk going up?

  2. GNewton says:

    “When we get it [installs etc.] wrong, we repeat the problem by missing appointments. We repeat the problem due to systems, process, ways of working – it could be a multitude of issues,” said Mears.”

    It’s good to see that even BT now admits to its failure here. This is the reason why it’s one of the worst rated companies in the UK. On the plus side, Ofcom in its review already has stricter targets for BT on this. This, along with stricter regulation and improved network infrastructure competition, should help to improve the situation.

    As regards the mapping of VDSL or fibre deployment areas: You can’t blame BT here, it was mainly up to the councils and the BDUK projects who failed to ensure a better transparency here, quite often hiding behind obscure commercial confidentiality clauses.

  3. chris conder says:

    Shame they weren’t forced to submit to the OMR. Everyone would know where they stood then. I did try very hard to watch a replay of this discussion, but couldn’t bear the snake oil hype. I just hope the committee saw through it all. For £7million a year I guess any of us could sell snake oil or what?

    1. Gadget says:

      Not sure what you mean by “shame they weren’t forced to submit to the OMR” – BT did input to the various Open Market Reviews.

  4. Perhaps CDS should talk to someone not bound by a CSI by BT e.g. https://labs.thinkbroadband.com/local/index.php?area=E07000040 and happily talks about difference fibre availability and superfast availability.

    1. gerarda says:

      Andrew Why don’t you publish a list of all the postcodes that without the installation of a new cabinet would not get superfast speeds? If that could be subdivided into ones that would still get below the proposed USO and the current USC that would be brilliant.

      It wont show those whose cabinet is not proposed to be enabled and live close to the cabinet but would be a hugely useful tool for the rest still awaiting BDUK improvements

      I would also like to compare it to the current Suffolk Phase 3 OMR which appears to be consigning many thousands of people to the digital dustbin.

    2. There are the maps, bulk data requests are a different matter and need commercial agreements in place and explanations of what the data does and does not tell you.

      Would love to be able to share every bit of data for free but that would be a fast path to being popular but not being able to pay a mortgage.

    3. MikeW says:

      “Something is rotten in the state of Devon”

      Whatever is happening in Devon, it isn’t repeated in every county. Like the DCMS committee, I wonder why Devon has such problems? Progress seems to happen just like other counties, but the public aggravation is orders of magnitude worse. Certainly the local MPs and councillors appear not just negative, but aggressively so. Does CDS fall between a rock and a hard place?

      There’s a description of the way some of the teams work, with Openreach people working side-by-side with LA people – but no word on what happens in CDS. Does the political shoving up above make for a fractious atmosphere? Or even an impossible one?

      The mapping point is, nowadays, a strange one. BT appeared to say that LAs can publish information at the 7-digit postcode level, as it is no longer subject to NDA. Which surely means they don’t need to report a full exchange or cab level, even if a premises level would be even better.

    4. DTMark says:

      “that would be a fast path to being popular but not being able to pay a mortgage.”

      “Welcome to thinkbroadband.com We are an **independent** broadband news and information site”

      I’m not sure how to reconcile that.

    5. gerarda says:


      There is anger in all the projects – its just that the CDS project is the only one honest enough to give a public voice to this (except when they appear in front of a Select Committee). Most of the others seem to work on the basis they have to make the most of the mess that the DCMS/BDUK have put them in.

  5. dragoneast says:

    I know BT inherited a mess with the local loops, and have never managed to sort it out, not least because the game is constantly changing. But is it, as seems to be the case, that no two groups of exchange-based engineers seem to operate in the same way? Historically we had loads of problems, but for the last 5 years or so, the local loop seems to be remotely monitored somehow, and most DSL faults on the network seem to be picked up and remedied by the local engineers with no customer or ISP input, and often when we are not even aware of a fault. How do they manage this, and why, apparently, not elsewhere? Is it some secret investment because, presumably, of our history of problems, or due to the underground nature of the relatively compact local network?

  6. ken says:

    All BT have ever done was point and blame now there taking blame? Ok yes we are sorry that you took close to £10Bn in returns and public funding for making Fibre Optic networking a UK-NATIONWIDE option. But failed at just 150,000 homes… for FTTP. So when we talk energy consumption why build 55,000 VDSL cabinets “that have an infinite fibre capacity line (Direct Node)” for just a mere 305/1 ratio? On 1gb “each cabinets, thus a cabinet uses more than 10kwh a day more in the “100-1000″. Say if it was 200 kwh x that by 55,0000=11,000,000 now by 365 =4,015,000,000 but even if it was a 50 kwh a day still be 1,003,750,000. Or 10Kwh we are still at 200,750,000- (£30,374,532.77) And yea 50 watts per line x 305=15,250! That is per CAB! 55,000×15,250=838,750,000Watts(Per day Fully occupied 305 brick dslam). So now the energy is out of the way let’s talk reliability yes copper last long because of its low carbon and iron content compared to steel so yes it lasts a long time in the ground but it is prone to EMP & other electronic interference. Copper is only acceptable in short runs!10-85meters. The upside of modern fibre-optics”fftp” is that it is MUCH FASTER AND CHEAPER TO Deploy on scale than copper infrastructure. Copper was before easy to deploy but due to everything else HSE and risk aassessments make copper the highest vulnerable cause of injury, but not to talk to much tech. once a fibre optic network is run and in place it is future proof can be upgraded at anytime with little to no disruption in service! Same cannot be said about copper due to bridge tapping or wire tracers. We could have spent a £20Bn on bringing the WHOLE OF the UK to full fibre within 10 years! But nope bt had there way… like I say if you are going to do a job do it first time correctly! Now we will only hand BT more money to put out FTTP.

    1. AndyH says:

      The latest FTTP rollout with BT stands at over 220,000 homes as of the end of Feb 2016.

      Uptake remains disappointing in some areas, primarily due to the lack of major providers wanting to supply a FTTP service.

    2. TheFacts says:

      @Ken – how did BT stop £20B being spent on FTTP?

    3. TheManStan says:


      Please refer to this charging proposal document for total power consumption of cabinets.

    4. MikeW says:

      The UMSUG document was my first thought as a response.

      That tells us the power demand is never higher than 500W, is less than 300W in 99% of cases, and less than 200W in 75% of cases. IIRC, my very rough rule-of-thumb was 100W base load, plus 1W per active port.

      Right now, that means the average cabinet consumes around 150W, or 3.6 kWh per day. Not close to the 200 kWh that @ken believes. A couple of orders of magnitude out. Oops.

      Deploying fibre isn’t cheaper than deploying copper. Not when the copper is already in the ground and up the poles.

    5. Oggy says:

      Poor ken.

      So wrong in everything he has posted.

    6. GNewton says:

      @Oggy: Your comment is inappropriate here. Ridiculing other posters here, however wrong they might be, does not contribute to this discussion here.

      Besides, though ken may have got some points wrong here, as explained by MikeW, he’s right in some points: There is indeed a danger that BT might receive more of taxpayer’s money in the future, which cannot be desirable. Also, for new deployments, e.g. on newly built estates, it often makes more sense to build FTTP rather than copper. And there are even cases where it is more economical to deploy fibre rather than VDSL in some rural locations. A ‘One size fit all’ approach doesn’t always work.

    7. ken says:

      Sorry for not putting evidence. But your document that you posted is out of date @thestanman . I am talking about Huawei cabs “MA5603T/-288′ I don’t like getting into detail as even the actual fibre engineers are baffled about code… when they tested my cab the average was 650watt 6 open bricks non vector. Underload they recently ran an update due to node update/repair exactly 2 weeks ago yet again this went on for a week or so testing, 1.2kw vectored. 6 brick 75%capacity “loading faze”. A loading faze exempts all issues and pushes max attenuation “line load at range/ distance to NTE”. each card is 100-200 watt.
      So please don’t be saying I don’t know my things about tech or information technology kinda grinds my gears as people get bored when they here tech talk and I find ignorance is a shame too Jelousy. Getting back to the cab stuff the exact loading faze we had was 75% loaded bricks one day was avenging 15Kwh where as vectored is around 27-40Kwh a day these are modern VDSL2 Cabs not the 09 input 7a locked cabs. And plus if you need to be technical against this ask openreach why must they source Hvdc’s engineers! Or better know as grid electric engineers a power cable to the cab is quite beefy. And if you would like to read my post I stated modern fibre install vs a copper install is much faster not over building. FTTC IS OVER BUILDING ON COPPER INFRASTRUCTURE TOWARDS THE PCP/CPU. If you start a fresh brand new infrastructure FIBRE OPTIC DEPLOYMENT IS FASTER!you can put nodes, splitters, splices fibre deployment housings anywhere!one 25mm fibre outbound can supply 300+ customers and you only need to supply one pair towards the one fibre housing.so you would be left with a couple free blown tubes. There is so many options in directions to supply fibre cabling example, dedicated, dark, pair, open, demand, infrastructure. One fibre cable can supply so much options! But like wise just money thrown at the wind. unlike copper there is only 2 direct one line or bride tapping.

    8. TheManStan says:


      Are you saying that vectoring more than doubles the power consumption of a cabinet?

      The more recent charge codes show peak 590W for fully populated 288 cabinets and the H806VPEA vector engine board for the MA5603T is 51W max with all 6 VCMM boards (line cards) connected…

    9. AndyH says:

      “one 25mm fibre outbound can supply 300+ customers”


      How many 25mm fibre cables have you seen?!

    10. ken says:

      25mm/1inch fibre cable and a lot. ..blown fibre can usually have less strands or pairs. But generic fibre cable is usually 1/6 the size of copper cable. Have you seen the size of PCP fed exchange copper cable? I think it’s around about 4 inches or so or 150mm. Sorry about the mistakes on spelling or what ever been using my phone to update. And yes I’d you read that sheet or pdf it dates 2009 to comply with the ef code . How can the 6 block ma5603t only use 51w?you realise to populate a line is 50+watts….

    11. TheManStan says:

      Design power max for a MA5603T non vectored is 700W fully populated with 6 line cards and 288 lines all connected, OR charge based on actual average draw for a cabinet in this config which is 590W.
      With the vector engine board in place and connected to all 6 line cards, this board uses an additional 51W max (21W static, i.e. if it was not connected to the line cards)

      So the maximum a vectored cabinet can be is 751W…

    12. ken says:

      @thestanman I am talking about real world data… not openreach to pass the un metered laws my cab is highly populated and is active G.inp we are not just taking about cards and racks here… we are talking the overall usage not just a open dslam. Fully populated dslams with 200+connections that require 50 watts!each pair!yes openreach hides these figures threw the old PCP but once you bond a like to the dslam/PCP crossover that is considered a active Fttc/VDSL line thus why I am making real world energy stats. Yes in the lab they average 604watts or so but as soon as you load faze or populate lines in the real world then the figure is much larger! 50wx288=1.44kwh that Is the max out share of a complete block in the PCP towards the dlsam. Even if we just say the dslam alone 0.6kwh is still 14.4kwh a day! (792,000kwh a day for 55,000dslam’s”once complete”) one year one cab 5256Kwh. 55,000 x 5256= 289,080,000Kwh that is all the DSLAM Active. Once you populate and add lines it is just a insanity number… but on the good side.. a Huawei Fttc cab cab have its dslam cards enhanced or chanced to accommodate fttp cards thus is why I’m not all against the Huawei cabs just the crappy FTTC bricks!

    13. TheManStan says:


      These are Huawei own specified MAXIMUM power draw, not OR…

    14. TheFacts says:

      @Ken – what’s the power consumption of a modem in the home?

      What’s ‘load faze’?

      50W x 200 = 10kW/cab. Really?

    15. MikeW says:

      “I am talking about real world data… not openreach to pass the un metered laws”

      What a load of tosh. This is nothing to do with “laws”, and everything with electricity supply companies wanting to be paid correctly. They want “real world” accuracy.

      Openreach’s original application for unmetered supplies was rejected because it didn’t include enough detail, enough real-world data, and enough fine-tuning of charge codes.

      As part of getting approved, they needed to supply proper “real world data”. In the submission, they supplied samples for 34,000 cabinets. More than half of their cabs at the time – all deployed and working. With real loads.

      I suspect that kind of sample might well class as “real world”.

      The numbers came out as:
      MA-5603: Base load 108W (0 lines); Full load 590W (288 lines)
      MA-5616: Base load 68W (0 lines); Full load 206W (128 lines)
      ECI M41: Base load 150W (1 line); Full load 614W (256 lines)

      Here’s the document with those 34,000 sampled cabinets, broken down into 44 groups with a different number of ports in use:

      Pages 19-21 show the results, but the rest of the document goes to explain the sampling process, and the statistical approach to showing “real world” predictability.

      Are you saying we should believe you in preference? Do you have an equivalent document explaining your sampling results in the real world?

    16. MikeW says:

      On your other point … Yes, DSLAMs use power. For everyone that has moved to a DSLAM port in the FTTC cabinet, they’ve moved away from a DSLAM port in the exchange, so reducing power use there.

      A full fibre rollout does indeed, ultimately, end up with a reduction in energy usage. That means it is part of the financial calculation to decide whether to roll out fibre. But it is a small amount of money, in the grand scheme of things, and so is a small consideration.

      My line, connected to a Huawei 5603, causes it to use 2.1W. Over a year, that amounts to 18 kWh. That means Openreach are spending £2 of my annual rental on DSLAM energy: £2 out of £143, or 1.4% of their income.

      The standard study for the cost to install fibre throughout the country comes up with an average of £1,000 per property.

      If “DSLAM energy” were the only benefit from an upgrade to fibre, it would take 500 years to recover the installation costs. I think you need to seek out some reasons that contribute bigger savings before BT will agree it is worth doing.

    17. MikeW says:

      Finally, consider G.Fast.

      In 2014, Sckipio reckoned that 1W per port was a feasible target. GPON backhaul seemed to use more power than plain GbE; it looked to me like a target of 1.25W per line was more reasonable if GPON were used.

    18. Steve Jones says:


      If you do want to make statements about power, then please source your stats, and also don’t confuse KW (power) with KWh (energy). As you are so keen of real world stats, then have a look at this as it is based on actual metered figures from actual installed cabinets in operation. At the time the report was done, average per-cabinet power consumption was around 205W, with only 0.2% of cabinets consuming more than 500W. Not doubt consumption has increased somewhat with higher uptake by customers, but the power per line usage will drop (and power requirements are no higher than ADSL).

      In reality, power consumption in the VDSL cabinets amounts to about £2 per year, not all of which would be saved with FTTC as that still requires active elements at each end (GPON being a bit more efficient than PTP fibre). Even if all that power was to be saved, it wouldn’t be more than a tiny fraction of the build costs. Most of the FTTP costs are in the “final mile” connection. Probably 80% or more. Vectoring does not involve a significant increase in power. Indeed the power that can be transmitted onto the line is closely controlled by the ANFP, and all the cancelling out of echoes is done using DSP in the digital domain using very efficient chipsets. To state that cabinet power consumption will double is laughable.

      G.FAST has a lot of power saving features built in so it will use notably less than VDSL or ADSL.


    19. Steve Jones says:

      Of course that should have read “not all of which would be saved with FTTP”, not FTTC.

    20. ken says:

      Ugh it’s like talking to walls… if you don’t want to even do basic maths don’t even judge my stats or knowledge, simply saying that a cab would only use £2 of electric a year is absolutely insane… and power (KW) equates to Energy (kwh) expenditure of time thus having an formular. I mean if you know basic electronics you would know what’s more dence Watts or voltage 50watt/1000volts or 50watt/200volts

    21. MikeW says:

      Do you read any of the posts? Properly?

      £2 per year per line.

      I don’t know if English is your native language or not, but your posts certainly make it hard to understand your knowledge, let alone judge the quality of that knowledge.

      However, on the evidence presented so far, I can certainly judge your stats. You started with a sample of 1. Plenty of links have been provided with samples of thousands (34,000 in 2014, somewhat fewer but still thousands in the 2013 submission). I don’t know about you, but the statistics portion of my maths education helped me understand that more samples are usually better. Ever heard of statistical significance?

      Or is that a little too far beyond “basic maths”?

      You’ll find that plenty of posters on here have both the education and experience to talk about electronics and electricity. And telecoms hardware, software and protocols for that matter. And plenty more aspects of networking, computing and IT. Even if you get bored in dealing with details, there are plenty who will do so instead.

      Please show us you’ve read the UMSUG documents, by telling us where their catastrophic mistake is.

    22. TheFacts says:

      200W = 0.2kWhr = 1752/yr x eg. £0.15 = £262.80.

    23. ken says:

      Ugh ok we will go ahead and use the aged graph and even Point out the wattage “Huawei MA5603T” Better know as a 288, so 6 cards, 288 lines, designed Total power non vector 729 watts, count of cabinets 13, average watts an hour (kwh 0.378) so we are at that cross road once again these stats are 4 years out of date wales is in the higher tier for 288 block cabs with over 0.5 million homes and businesses ready for fttc. Even if we go on a median basis 3 line cards 259 watts. .. so like wise we will do our energy consumption one cab first 3 line cards 259 watts. E (kWh/day)=P(w)xt (h/day)/1000 (W/kW). Consumption 259watts (W),Hours of use a day 24 (h/day), Energy consumed a day 6.216kWh/day, Energy consumed 186.48 kWh/Month, Energy consumed a year 2268.84 kwh/year. Now this is a median on smallest usage per median count of cabinets. 55,000 cabinets x per single median cabinet (3 line cards) consumption per year = 124,786,200 kWh/year. So if you would like to add the additional line consumption even if you say it’s 2 watts “clearly not” (0.03amps) because each line is 48-56volts, but anyway 2.1x 288=604.8 so if you go about and add that to overall cabinet consumption 604.8+259=1.566kW So again Power consumption 1.566(kW),Hours a day 24 (h/day),Energy consumed 37.584 (kWh/day), Energy consumed 1127.52 (kWh/month),Energy consumed 13718.2 (kWh/year). So know you have the full down consumption on your data provided you can see how much consumption one cab or different cabs use it made a median so it’s not the minimum or maximum but in the middle less bs and just more facts. But all in was really debating was wasted energy and money, we will sooner more than later have to replace the copper with Fibre yes bt played there cards by making more jobs by spending money to make money, yes there plan worked because one we now are the most viable for fibre tech, 2 we have heat fibre back bone open air/ dark fibre( free fibres) 3 bt has created 10s of thousands of jobs in this process many ex armed forces. Every area with a fibre fttc cabinet is already SET FOR FTTP all they have to do is over build one a fttp displacement cabinet next to the dslam or take the dark fibres from the fttc node to open splitter direct to the main exchange what ever is cheaper and easier. It’s basically how FTTPOD works takes a pair of the aggregation node or by adding a add in card in the occupied fttc cab. Which saves money and build. So like I said again the fttc cabs ain’t all loss just a much harder way of dealing GEA FIBRE!

    24. themanstan says:


      look at the Huawei systems soec link i´ve put below and figures provided.
      Those are Huawei complete power numbers for their cabinets.
      And those are the figures for are total max power draw for a 288 cabinet…. they are 751W vectored.

      You can´t take max power draw and add more power from nothing. If you want line consumption, you take the max power draw and divide by the 288 lines… 751/288= 2.6W

    25. Steve Jones says:


      You do realise that the line voltage of 48V has precisely nothing to do with xDSL services don’t you? It’s for the phone service (and always has been) and it’s DC coupled. VDSL & ADSL work perfectly happily in the absence of that DC voltage, and it’s a complete irrelevance.

      The broadband service is a modulated radio frequency signal using the pair as a transmission line. It’s capacitively coupled with the phone line offering a characteristic impedance of roughly 600 ohms. The total amount of power that the ANFP allows onto the line is less than a watt (to avoid cross-talk and other RFI issues). Due to the power masking (to protect things like useful exchange-based ADSL frequencies) the actual power allowed onto the line is lower where the cabinet is close to the exchange.

      Here’s a document on line driver efficiency, where you’ll see that they are talking of about 700mw. Of course it’s not possible to achieve 100% efficiency in a line driver, and there are also other components which use power (although the line driver dominates).


      As for the figures being out of date, do you think that these cabinets have suddenly got less efficient? Yes, there will be a higher line occupancy with more take-up, but as only 0.2% were over 500W it’s quite reasonable to assume there were a few that were highly occupied.

      You seem to think that network engineers are fools, that they don’t consider issues such as power consumption and that they are incapable of understanding the issues involved. They are most certainly not, and neither are the accountants that look at total operating costs.

      My rough calculations are that total power consumed by VDSL cabinets with approximately 7m lines enabled (around 30% penetration) will be of the order of 20MW, or less than 200GWh a year. Total UK power generation capacity was of the order of 75TWh in 2015. So that’s rather less than 0.3% of the whole. Significant, but small compared to what is used by industry, for lighting, cooking and so on. It does, of course, omit all the consumption of modem/routers in the customer premises, but FTTP still requires active terminating equipment, routers and so on.

      This is the ANFP document. There may be later versions, but I don’t have any inside information so this is what I’ve found published by the NICC, but the principles won’t change.


      I should declare my qualifications, which are degree in Physics from Imperial College London (graduated in 1976). They’ll have the records if you wish to check it.

    26. ken says:

      @steve Jones I have never said any one has or had been wrong…I never added 48v line capacity to the the figure. I am just giving stats on cabs and yes myself Energy Trade Certified 4 years ago used to do domestic analysis on Electric, Gas, natural heat loss and consumption. So I’d say I know what I’m talking about plus it’s not hard to give a median on energy consumption. I mean my electric consumption was above 6000kwh now I’ll be looking at a overall of roughly 4000 or less with the changes I’ve made. But it doesn’t take a genius to work out consumption. Plus we are not looking at power generation. Consumption plus yes it’s easy to use gwh As a reference of ease for big credits on numbers.

    27. ken says:

      But if you wanted to just keep going in circles then we can just do the complete total power consumption of all 55,000 cabinets that will be deployed in total (361.8318GWh) just easy maths… 6,578.76×55,000=361,831,800 or (361.8318GWh)

    28. Steve Jones says:


      Do you read what you write? You now claim that you never raised the issue of the line voltage having anything to do with VDSL power consumption, so how do you explain writing this?

      “So if you would like to add the additional line consumption even if you say it’s 2 watts “clearly not” (0.03amps) because each line is 48-56volts”

      In any event, at least your latest number is rather more credible than the earlier ones, such as this where you were claiming 50W per line, or 15,250W on a 305 line cabinet.

      “And yea 50 watts per line x 305=15,250! That is per CAB! 55,000×15,250=838,750,000Watts(Per day Fully occupied 305 brick dslam).”

      Of course, if that had been 50 watt-hours (or about 2 watts x 25 hours), then it might have looked a bit more reasonable and rather more in line with what I (and others) have used as an estimate.

      Of course many of the newer cabinets under BDUK are smaller, so have lower consumption. In any event, as the great majority of VDSL enablements are migrations from ADSL, any net increase in power consumption is very small as an idle line driver amplifier on the losing ADSL port will reduce power consumption at the exchange by a similar amount.

      Whatever the total, £20 or even £30m a year makes no material difference to the economics of retrofitting FTTP (new build areas are a different matter, and the latest announcement is OR will build those for free in the great majority of new developments with more than 250 properties).

    29. MikeW says:

      Ah. An electrician, who undoubtedly understands the concept of electricity, power, energy and costs. You are good at multiplying the power figures to create energy, and multiplying up to give annual figures, and the “always interesting” multiplication by 55,000 (you’re out of date on that, BTW).

      But you aren’t good at figuring out the input to that multiplication figure: the concept of what is happening within one DSLAM or on one line, and what the power consumption is before you start multiplying. Don’t both multiplying until you get the base figure right.

      Unfortunately, if you think the DC line voltage (-50V) has something to do with the power consumed by a DSLAM, then you certainly don’t know anything about the design of telco electronics. You don’t understand what power is being consumed, where, and supplied from which source.

      If you think that you should take the “median power for a 3-card DSLAM” (as 259W), and then add a spurious “2.1 * 288 = 604.8W” on top, then you show both a failure in maths (if it is a “3-card median” cab, you would have 144 lines, and be adding “2.1 * 144” instead), but a failure in understanding the “power per line” calculation (because you shouldn’t be adding “2.1 * anything” at all).

      That tells me you don’t understand how to apply your knowledge of electrical engineering into the world of electronic engineering, and specifically into telecom electronic engineering.

      UMSUG tells us the power consumption of that “3-card median” cabinet is 350W (I’m not sure where you got the 259W from). That’s it. Period. No need to add anything “per line”. Consumption is just 350W.

      The 350W of power consumed by the DSLAM is provided by the 240V supplied into the base of the cabinet, from a supply hookup that costs minimum £1,000 per cabinet. The power isn’t supplied by the phone lines, or their -50V DC voltage.

      The calculation of “power per line” is then made from taking the total consumption, 350W, and dividing it by the number of lines it supports; 3 cards will support 144 lines. 350W / 144 lines = 2.4W per line.

      If you want to multiply up into a per-village, per-county, per-nation, per-annum energy and cost figure, go ahead from there. But there a problems with doing that too, as you need to work from the right figures.

      Your use of “3-card median” show a lack of knowledge of statistics, as you’ve created a “median” from half of the size of a 288-line cabinet. That isn’t the way to calculate the median.

      The right way to find the median-sized cabinet is to know how all the cabinets are provisioned, with how many cards and how many ports, right now; then use the sample of the 50th percentile cabinet. UMSUG numbers (2014) told us there were 14,527 Huawei-288’s in the sample; the 50th percentile would be the 7,264th cabinet. From UMSUG’s sample sizes, that comes out in charge code 4 – with between 49 and 64 ports. Just starting to use a second card. UMSUG charge code 4 (Huawei-288, two cards, 49-64 ports) has power consumption of 217W. The “power per line” figure would be 3.4W (217 / 64).

      We cannot calculate an up-to-date median, because we don’t know up-to-date sample sizes. But we can calculate an up-to-date average, using the number of cabinets (currently 74,000 not 55,000), and the number of subscribers (last reported as 5 million).

      If we wanted to be properly accurate, we’d split everything down into the 3 different types of cabinet (they’re not all Huawei-288 types), and apply individual calculations. But let’s not bother being *that* accurate. Lets assume they are all based on the Huawei 288.

      The average-sized cabinet is therefore around (5 million / 74,000), or 67 subscribers per cabinet, at the start of 2016. That’s comfortably in the middle of 2-card usage, but with slightly more ports than the 2014 figure. Let’s apply the UMSUG charge-band 5 (Huawei-288, 65-80 ports, 237W power). Average power consumption of a cabinet is likely to be 237W, supplying 67 lines. Average power per cabinet in 2016 is 237W.

      Average “power per line” is, with 2016 deployment figures, currently (237 / 67) = 3.5W per line.

      Until you can understand how these two basic numbers are derived, either from thousands of samples or from the electrical specifications, there is no point multiplying anything up into nationwide figures per annum, and getting emotional about the cost.

      However, you do like to trumpet the big numbers, so let’s humour you:

      237W per cabinet
      @ 24 hrs/day
      = 5.7kWh per day

      @ 365 days/year
      = 2.1 MWh per annum

      With 74,000 cabinets
      = 155 GWh per annum

      Cost to BT:
      @0.12p per kWh
      = £18m per annum

      Cost per FTTC subscriber
      @ 5 million subscribers
      = £3.70 per subscriber, per annum.

      After that, we can start on your mistaken belief that it is cheap to install fibre from the aggregation node to the home. But let’s leave that for another day. The maths and statistics are even more complicated.

  7. Al says:

    Yet another committee report which is basically saying the same thing as previous ones, whilst the BDUK project has had some success it has also failed in areas, I and no doubt many others are only interested in one thing the date for which we can expect a fibre based service. Given that I’m in a FTTP planned area (I only know this because I can see work being done every now and then and I messaged BT to enquire) as the when and where site gives seemingly gives no information on planned FTTP deployments.

    But I guess they are between a rock and a hard place, give a date and it slips people complain, don’t give a date and people complain.

  8. TheManStan says:

    The devil is in the detail

    VDSL2 17A 288 port (6 x 48 channel) MA5603T typical setup:

    H801SCUN*2 62W each 104W Control Board
    H801X2CS*2 18W each 36W 10GE uplink board
    H801CITD 4.55W 4.55W Interface board
    H801PRTE*2 1.5W each 3W Connect Power Board
    H801FCBI 72W 72W Fan Tray
    H805VDMF(17a)*6 82W each 480W 48 channel VDSL2 over POTS (line cards)

    Non Vectored Grand Total 699.55W

    H806VPEA 51W 51W Vector engine board

    Vectored Grand Total 750.55W

    Infromation source:

    Huawei SmartAX MA5600T/MA5603T/MA5608T Multiservice Access Module Hardware Description Issue 8

Comments are closed.

Comments RSS Feed

Javascript must be enabled to post (most browsers do this automatically)

Privacy Notice: Please note that news comments are anonymous, which means that we do NOT require you to enter any real personal details to post a message. By clicking to submit a post you agree to storing your comment content, display name, IP, email and / or website details in our database, for as long as the post remains live.

Only the submitted name and comment will be displayed in public, while the rest will be kept private (we will never share this outside of ISPreview, regardless of whether the data is real or fake). This comment system uses submitted IP, email and website address data to spot abuse and spammers. All data is transferred via an encrypted (https secure) session.

NOTE 1: Sometimes your comment might not appear immediately due to site cache (this is cleared every few hours) or it may be caught by automated moderation / anti-spam.

NOTE 2: Comments that break our rules, spam, troll or post via known fake IP/proxy servers may be blocked or removed.
Cheapest Superfast ISPs
  • Hyperoptic £15.00 (*25.00)
    Speed 50Mbps, Unlimited
    Gift: None
  • Vodafone £19.50 (*22.50)
    Speed 35Mbps, Unlimited
    Gift: None
  • NOW £20.00 (*32.00)
    Speed 36Mbps, Unlimited
    Gift: None
  • Shell Energy £21.99 (*30.99)
    Speed 35Mbps, Unlimited
    Gift: None
  • Plusnet £22.99 (*38.20)
    Speed 36Mbps, Unlimited
    Gift: £70 Reward Card
Large Availability | View All
New Forum Topics
Two ISPs on one line?
Author: unhappyguy1673
Brawband in Glasgow
Author: Simmie
Contract coming to an end.
Author: LeeWillRockYou
B535-232 + Poynting vs B818
Author: Conspiracy
Cheapest Ultrafast ISPs
  • Hyperoptic £20.00 (*35.00)
    Speed: 150Mbps, Unlimited
    Gift: None
  • Vodafone £24.00 (*27.00)
    Speed: 100Mbps, Unlimited
    Gift: None
  • Community Fibre £25.00 (*29.50)
    Speed: 300Mbps, Unlimited
    Gift: None
  • Gigaclear £27.00 (*59.00)
    Speed: 500Mbps, Unlimited
    Gift: None
  • Virgin Media £27.00 (*51.00)
    Speed: 108Mbps, Unlimited
    Gift: None
Large Availability | View All
The Top 20 Category Tags
  1. FTTP (3505)
  2. BT (3010)
  3. Politics (1925)
  4. Building Digital UK (1919)
  5. FTTC (1882)
  6. Openreach (1822)
  7. Business (1680)
  8. Mobile Broadband (1469)
  9. Statistics (1406)
  10. FTTH (1364)
  11. 4G (1271)
  12. Fibre Optic (1167)
  13. Virgin Media (1159)
  14. Wireless Internet (1153)
  15. Ofcom Regulation (1139)
  16. Vodafone (836)
  17. EE (830)
  18. TalkTalk (761)
  19. 5G (760)
  20. Sky Broadband (744)
Helpful ISP Guides and Tips

Copyright © 1999 to Present - ISPreview.co.uk - All Rights Reserved - Terms , Privacy and Cookie Policy , Links , Website Rules , Contact