Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
   Glossary   Glossary
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Author Topic: G.Fast - the next generation of broadband  (Read 5801 times)

c6em

  • Reg Member
  • ***
  • Posts: 504
G.Fast - the next generation of broadband
« on: May 11, 2013, 09:28:19 PM »

I raise this subject as one for discussion.

As I understand it the major cost in FTTH is claimed to be the final 50 yards from the manhole/joint box to the property.
Well, I might agree in an urban environment but I’d rather take issue with it in a rural environment where the last 50 yards is trivial compared to the previous 1000 yards from the FTTC cabinet

Anyway the solution proposed is being touted under the fancy name of “G.Fast”.
All in long term development at the moment (5+ years away) however it seems to be fibre to the distribution point or node (FTTN), a Mini-Dslam installed there and hence by conventional phone wire to the house.  This would allow the use of both a huge increase in the max frequency of the DMT spectrum (talking possibly up to 200MHz) and maybe an increase in the max bit loadings above 15 used at present.
All leading to achievable speeds in excess of 500Mbps at 100 yards or so – ie available at all the properties served in almost all cases.

How to get the power to these mini-slams?  Well the proposal is that the end user should do this by back feeding power from the property to the Dslam.
Presumably this is down the phone line – otherwise if you have to put in a power cable from the house you might as well put in the full fibre FTTH anyway.
Quite how the various end users are going to work out and share out their electrical contribution to the overall running costs of the Dslam I dread to think. (what happens if a number go on holiday and switch off – does the Dslam stop working?)

 
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: G.Fast - the next generation of broadband
« Reply #1 on: May 11, 2013, 09:44:33 PM »

Do you have any links to references, please?  ???
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

ColinS

  • Reg Member
  • ***
  • Posts: 529
Re: G.Fast - the next generation of broadband
« Reply #3 on: May 11, 2013, 10:29:48 PM »

Do you have any links to references, please?  ???
Hi B*Cat,

AIUI, G.Fast is the transmission standard required for what OpenReach refers to as FTTdp. A quick Google on the conjunction of those 3 terms should provide some info about their thinking on that.

Two ETSI TR references for FTTdp Reverse Power
http://www.etsi.org/deliver/etsi_tr/102600_102699/102629/01.01.01_60/tr_102629v010101p.pdf
http://www.etsi.org/deliver/etsi_tr/102600_102699/102614/01.01.01_60/tr_102614v010101p.pdf
« Last Edit: May 11, 2013, 10:47:32 PM by ColinS »
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: G.Fast - the next generation of broadband
« Reply #4 on: May 11, 2013, 11:20:58 PM »

Thank you, both, for the links.

I now have something to read!  ;)
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

Black Sheep

  • Helpful
  • Addicted Kitizen
  • *
  • Posts: 5722
Re: G.Fast - the next generation of broadband
« Reply #5 on: May 12, 2013, 09:47:54 AM »

This sounds interesting. I haven't got the brain-power to read through the links, (that's why we have a cat and an asbo), but when you talk about power the DSLAM being back-fed from the premises, would this be 240Vac, or a stepped down voltage, which is stepped back up at the Cabinet ?? I'm just pondering over the existing BT cable feeds into EU's premises, and the fact they'd be woefully inadequate to transmit 240V over them ??
Logged

c6em

  • Reg Member
  • ***
  • Posts: 504
Re: G.Fast - the next generation of broadband
« Reply #6 on: May 12, 2013, 10:04:24 AM »

No, this would be at "low voltage"  - I'd guess at no more than then current ringing voltage max (?50V?).
You'd want to keep it reasonably high (ie not 12V) due to wanting to keep down the current (milliamp) requirements otherwise at too low a voltage (hence higher current) the volts drop along the dropwire back to the mini-slam become a problem.

The mini-slam would then use this voltage directly or convert it directly into the voltage required by the electronics.  So it would not convert back up to 240V.
I think the idea is that these mini-slams are designed as self contained totally sealed vulcanised whatever units.  So they become just another electronic component which you fit and forget and when it goes wrong you chuck it in the bin.

The other thing you COULD do is to provide an additional low voltage power source to the mini-slam by way of a spare line pair fed from the powered FTTC cabinet - assuming the distribution point has spare line pairs which would now be re-designated as 'power pairs': but voltage drop on longer length pairs would be a problem here.
« Last Edit: May 12, 2013, 10:06:31 AM by c6em »
Logged

Black Sheep

  • Helpful
  • Addicted Kitizen
  • *
  • Posts: 5722
Re: G.Fast - the next generation of broadband
« Reply #7 on: May 12, 2013, 10:54:25 AM »

Thanks for replying. This is all very interesting and expanding my knowledge. Another thought, how could the power be transmitted in two directions ?? IE- 50Vdc (approx. 100Vac when ring current is applied), going into the EU's premises, and the required power for the mini-slams coming back from the EU's premises ??

Please don't think I'm trying to be awkward here, far from it, I'm genuinely interested.  :)
Logged

smurfuk

  • Member
  • **
  • Posts: 26
Re: G.Fast - the next generation of broadband
« Reply #8 on: May 12, 2013, 11:53:14 AM »

... Another thought, how could the power be transmitted in two directions ?? IE- 50Vdc (approx. 100Vac when ring current is applied), going into the EU's premises, and the required power for the mini-slams coming back from the EU's premises ??

Please don't think I'm trying to be awkward here, far from it, I'm genuinely interested.  :)

Does this help? http://www.etsi.org/deliver/etsi_tr/102600_102699/102629/02.01.02_60/tr_102629v020102p.pdf

In simplistic terms a development of PoE, which is used to power the "equivalent" receiver for my fixed wireless connection.
Logged

ColinS

  • Reg Member
  • ***
  • Posts: 529
Re: G.Fast - the next generation of broadband
« Reply #9 on: May 12, 2013, 03:10:32 PM »

Thanks for replying. This is all very interesting and expanding my knowledge. Another thought, how could the power be transmitted in two directions ?? IE- 50Vdc (approx. 100Vac when ring current is applied), going into the EU's premises, and the required power for the mini-slams coming back from the EU's premises ??

Please don't think I'm trying to be awkward here, far from it, I'm genuinely interested.  :)
BS, the ETSI documents are not exactly bed-time reading, but I think the sort of nominal line power levels they are talking about are 60vDC @ 300mA, total power < 20W, and each EU pair provides its own power.  It does indeed cause problems for POTS - both ringing current and off-hook indications, so, while the line is back-powered, the FTTdp/'back-powered' faceplate combination (AC supplied) in co-operation intercept those signals at their respective ends, and pass them down the line signalling in supervisory signals instead.  If/when the AC power to the back-powering faceplate fails, then everything reverts to POTS signalling. But note, this is only a conceptual plug-in faceplate; I doubt the final design(s) are actually worked out yet.
One of the design challenges is to operate at such lower power levels; but there have already been sucessful demonstrations I believe.  There are graphs in the documents showing power fall-off with drop-wire length, but they have in mind generally average short drops.
HTH  ;D
« Last Edit: May 12, 2013, 03:20:01 PM by ColinS »
Logged

Chrysalis

  • Content Team
  • Addicted Kitizen
  • *
  • Posts: 7402
  • VM Gig1 - AAISP CF
Re: G.Fast - the next generation of broadband
« Reply #10 on: May 13, 2013, 07:45:10 AM »

One hand it is not a great idea as it keeps copper going longer with its crosstalk issues, joint issues, corrosion issues, distance dependency and so on.

However the main way this spec achieves it speeds in reality is just shortening the copper length (like FTTC does) so its a good thing as it will remove more copper from the loop used on broadband and that makes FTTP closer as the less copper there is the more viable a move to FTTP is.
Logged

renluop

  • Kitizen
  • ****
  • Posts: 3326
Re: G.Fast - the next generation of broadband
« Reply #11 on: September 26, 2014, 10:46:05 PM »

I know this is an old thread but in today's Telegraph business section is an article about
  G-Fast.
Logged

guest

  • Guest
Re: G.Fast - the next generation of broadband
« Reply #12 on: October 13, 2014, 10:50:32 AM »

That's more or less what I was working on earlier in the year for Alcatel-Lucent (Bell Labs) - ie reducing power consumption & increasing vectoring resources.

I'm laughing at 20W consumption - its going to have to be a LOT less than that to support the required number of ports/module. Hell with 20W/port we could have TRIPLED vectoring resources but some of the modules would have required active cooling which is a big no-no for pole mounting due to maintenance costs.

Can't give you any figures but its going to be a lot less than 20W.

Edit - and it may indeed be 5 years away in the UK but it certainly isn't in other European countries. Austria has a trial which has been running for a while now, I'd expect a full product rollout within 2 years max, especially now the EU regulator has moved more towards a wholesale/virtual unbundling model.
« Last Edit: October 13, 2014, 10:54:36 AM by rizla »
Logged
 

anything