How does it work

I changed to BT Infinity on a Friday but due to a cock up I had no broadband, didn't get broadband until following Tuesday.
This was due to someone making the wrong connection in the cabinet.
Just so I can understand, can someone explain how the phone was changed to BT and worked o.k. but not the broadband when only one pair of wires deliver both.

Because beforehand the phone pair went from the exchange to the phone cabinet and from there to you.

With FTTC it goes from the exchange to the phone cabinet just the same, but is re-routed in there to the fibre cabinet, where the fibre signal is merged onto it. (The FTTC signal doesn't necessarily even come from your exchange). The result is routed back to the phone cabinet and in there is connected to the pair to your house.

What exactly happened I don't know, but maybe someone else's line was connected to your fibre and your line wasn't touched. The ADSLx would normally have continued working in that case so that probably isn't it. I'd need to think more, but that's how the circuit is constructed.

Edit - who was the line with before BT? That may give us a clue.

Edited by RobertoS (Tue 24-Mar-15 16:05:10)

Or to put it a little more simply - the phone connects in one cabinet, the broadband in another - someone messed up the broadband bit

Uh uh.

Nobody goes near the FTTC cabinet for FTTC installs. It's all done using the "patch panel"* installed in the PCP. He was told it was in the cabinet.

Perhaps the engineer connected to the wrong FTTC to/from coupling, instead of the one that would have the OP's feed at the other end. That would keep the phone working but with no fibre input.

* I don't know if that's what it is called in this scenario, but that's the function.

Agreed generally with what you have stated.

The question that arises in my mind - Is there a specific link/VDSL filter point (pre-) allocated remotely within the FTTC?

Presuming that some form of "multiplexing" is carried out in the FTTC (DSLAM?) to get all of the (present) VDSL subscribers connected both ways to the fibre run, does it matter which of the available, linked and filtered FTTC terminations that anyone is connected to?

-------------
Edit
I wonder if the installer forgot the final action of cutting the two, original D-side lines in the PCP, thus short-circuiting the (new) links to and from the FTTC?

I wonder what effect that would have?

Edited by deleted (Tue 24-Mar-15 19:22:20)

There are individual ports on the DSLAM line cards per PCP connection. I seem to remember an engineer post on here that the installer is told which PCP port to use in order to link to the subscriber's FTTC headend connection. The implication being that the provision from the headend is to a specific port on a specific card.

These ports being hard-wired to the PCP at card installation time.

I'm not sure if that technically covers your question, but AIUI it is not up to the engineer just to connect using any he wishes.

In reply to a post by eckiedoo:

Edit
I wonder if the installer forgot the final action of cutting the two, original D-side lines in the PCP, thus short-circuiting the (new) links to and from the FTTC?

I wonder what effect that would have?

(This wasn't present when I started my previous post.)

I don't think that holds water. To connect to the patch panel there would need to be a link from both the exchange and premises pairs. If not the OP's then whose?

"I wonder if the installer forgot the final action of cutting the two, original D-side lines in the PCP, thus short-circuiting the (new) links to and from the FTTC?

I wonder what effect that would have?"

Assuming that the installer had actually carried out almost all of the work correctly, from locating the subscriber's correct D-line (17070), through connecting that pair correctly to the correct port in the FTTC - but simply forgot to do the "final snips", to ensure that the POTS is properly diverted to the FTTC and back - or to view it another way, that the POTS can still flow directly through the PCP to the subscriber, instead of being forced by the "final snips" to go via via the FTTC.

--------

My suspicions are that although the VDSL signal would still get to the subscriber, it would be attenuated to the degree that it would not be detected by the subscriber's modem.

In that situation, one pair of the FTTC links would be delivering the VDSL signal effectively without rejection filtering, to the E-side pair (including a very short length of the D-side in the PCP) to the Exchange.and the implied losses of the long length, particularly at the higher operational frequencies of VDSL.

Only experimentation or if someone has actually encountered the situation, might provide the answer.

Interesting.

Sorry, I don't understand (on a quickish read) that theory, and I don't know what this final snip is that seems to be the basis of it.

The existing E-side and D-side pairs start off connected. To connect either to the FTTC panel requires disconnecting them from each other.

Edited by RobertoS (Tue 24-Mar-15 20:46:24)

I am not an OR or BT engineer, but am trained in microelectronics.
"multiplexing" is used in the FTTC cab and then the resulting signal is sent back to the phone pair, and then to the home, but there might have been a fault in the multiplexing chip [on an input or output gate] that needed checking.?
Hence the delay.
The VDSL signal would not be getting to the home [at least not fully], because if the multiplexer was faulty it would be bottlenecking the circuit until fixed.
The delay was probably due to a circuit fault somewhere in the FTTC box even if not at the main multiplexer chip [or chips]

Edited by deleted (Wed 25-Mar-15 00:38:32)