Quite possible, perhaps it would be best if he posted pictures of them.

In reply to a post by R0NSKI:

If you have no true master socket, isn't it free to get one fitted?

What was installed was old style star wiring which was never remedied (it looks like at one point it was free to have this remedied by BT but this is no longer the case). Because they are star wired, both sockets are identical so there's no master/slave relationship.

At some point the individual sockets have been modernised (from the looks of things sometime in the 80s), the style of wiring and type of sockets used seem to indicate that this was change performed by a BT engineer.

I'll post pictures of the junction box this weekend.

Edited by deleted (Thu 21-Mar-19 12:16:43)

An Openreach engineer would not have left two master sockets, even of an old type, on the same line. Nor would they have installed such a setup in the first place.

In reply to a post by RobertoS:

An Openreach engineer would not have left two master sockets, even of an old type, on the same line. Nor would they have installed such a setup in the first place.

I suspect this pre-dates openreach. I also suspect this was the standard manner in which star wiring (which implies multiple sockets) would have been connected, and presumably each socket (at the end of an independent spur) would have required a capacitor so that it would work with old style phones.

Edited by deleted (Thu 21-Mar-19 12:41:26)

A BT engineer in the days before Openreach existed became an Openreach engineer. Leavers and deaths excepted. (Oh - and promotions out of harm's way! )

Assuming you are talking about LJ-type sockets, no. AIUI there should still be only one master on the line.

In reply to a post by RobertoS:

Assuming you are talking about LJ-type sockets, no. AIUI there should still be only one master on the line.

Okay, I assumed the capacitor was there to pass the variable ring current up to the connected phone, while blocking the dc current.

In this case there are two independent spurs off the junction box - so would have assumed the socket on each spur would require a capacitor - anyway, will try and take photos of the insides of both sockets and the external junction box and post them here.

It probably depends on how many wires go from the junction box to each socket. Two would I suppose work in the way you say. Three should have T3 > junction box > T3 at the other socket, and I think that would work my way.

So this is a picture of the junction box outside the house:

https://ibb.co/KxX7GN4

The two greyish wires lead are the ones that lead to the sockets - they are of identical vintage, and are stapled to the wall, the one leading upstairs goes up the outside the house before going directly through the wall to a LJU. Both LJUs look like this:

https://ibb.co/HN6MJ6q

and like this inside:

https://ibb.co/GxM76cb

The only modification's I've made was to disconnect pins 3 and 4 (orange and green respectively).

Incidentally, I've now had VDSL activated on the line, and am currently getting 15.358 Mb/s / 5.075 Mb/s

I'm sure I'd get better rates with modern wiring, but suspect most of the benefit would be from having proper twisted pair rather than old style solid core, and the elimination of various extraneous bits of wiring between the filters and the socket, with the inbuilt filter playing a relatively minor role (the price point of an NTE5 is in the same ballpark as a decent standalone filter -- and so I don't assume the quality of the electronics is going to be a whole lot different).

In reply to a post by chrisstiles:

In this case there are two independent spurs off the junction box - so would have assumed the socket on each spur would require a capacitor - anyway, will try and take photos of the insides of both sockets and the external junction box and post them here.

I think that was how my house was wired. Creates an interesting situation if you have a fault as you can argue both legs are BTs. The correct way is to use 3 wires between the master and the slave sockets. Most modern phones do not use this third wire.

In reply to a post by chrisstiles:

The only modification's I've made was to disconnect pins 3 and 4 (orange and green respectively).

Incidentally, I've now had VDSL activated on the line, and am currently getting 15.358 Mb/s / 5.075 Mb/s

I'm sure I'd get better rates with modern wiring, but suspect most of the benefit would be from having proper twisted pair rather than old style solid core, and the elimination of various extraneous bits of wiring between the filters and the socket, with the inbuilt filter playing a relatively minor role (the price point of an NTE5 is in the same ballpark as a decent standalone filter -- and so I don't assume the quality of the electronics is going to be a whole lot different).

Having a spur before the VDSL modem is not good practice as you get reflections which will interfere with the signal. If you disconnect the other spur from the junction box you could find out what improvement that gives. You could then decide what to do next!

Edited by Michael_Chare (Fri 22-Mar-19 16:47:02)