Connecting Wire to Master Socket Causes Fall in Broadband Speed

[AABBCC_CCBBAA]

Hope someone can shed some light on this curious issue for me.

I wanted to connect my router to a slave socket in an understairs cupboard.

The house was a new build 20 years ago and I ran CAT5 cabling throughout including a length of CAT5 from the back of the Master socket to the understairs cupboard.

Download speed with router connected to the Master socket test socket is 23 Mbs. (Faceplate removed)

Speed with router connected via faceplate connection is 15 Mbs.

The blue/white and White/blue of CAT5 cable are connected to terminals 2 and 5 respectively. In the understairs cupboard they are not connected to anything!

If I disconnect the 2 blue wires from the master socket and plug the router into the faceplate download speed is about 20Mbs.

If I connect a single wire to terminal 2 (only one wire connected)  the download speed falls to 18Mbs.

If I connect the second wire to terminal 5 the download speed falls to 15Mbs.

If I use any other pair of wires in the CAT5 cable the speed via the faceplate is 15Mbs.

In short, if I connect any wire to terminals 2 and 5, even if the cores have no connection at the other end under the stairs, the download speed is reduced from about 23Mbs to 15Mbs

How can this be rectified if I want a slave socket for my router to be connected to somewhere else in the house?

Thanks

AJ

[Devon_Dave]

Suggest you check the integrity of the cable, sounds like there maybe an earth contact and/or a short circuit of each individual wire.

Maybe worth put a test meter across each pair. 

[DaveS1234]

Hi,

Which type of Master Socket do you have and do you wish to have a phone connected at the master socket in your desired arrangement?

Master Socket Pictures in

https://community.bt.com/t5/ADSL-Copper-broadband/Master-socket/m-p/2138143

Message 6

If you are connecting your 2 cores to the unfiltered signal at the Master Socket ( which you need to to extend the A/VSDL ) then you are effectively connecting an un-terminated stub that can act as an aerial. Have you tried  connecting your router to the extension socket with nothing (except for your extension pair) connected at the master socket?

I've seen unfiltered extensions (with nothing connected) knock ADSL2 down from a solid 16Mbs to a flaky 10Mbs at the Master Socket.

Dave

[AABBCC_CCBBAA]

Hi Dave,

Thank you for taking the time to reply. 

Firstly in answer to your questions.

The Master socket type is numbered 2 in the link you gave. It only has the telephone jack connector so I assume that any connections inside which I make are unfiltered. There are terminals 2,3,4 and 5 and I am only using 2 and 5 to one pair of twisted pair in my CAT5 cable. All my tests are using the RJ11 connector from the router to an ASDL filter which is then plugged into the telephone jack of the socket I am using.

I took your advice and connected a Slave face plate to the other end of the cable (under the stairs) rather than leaving the connections open. Interestingly this is what I found:-

1. With the slave faceplate connected with terminals 2 and 5 under the stairs (so effectively the cable is connected to the Master socket) and with the router plugged into the faceplate at the Master socket, I get download speeds of around 15Mbs. i.e. slower than using the Test socket.

2. If I move the router and connect it to the Slave faceplate (so using the extension cable) and obviously with the faceplate on the Master socket in place, I get download speeds of around 20Mbs!

How does that work!?

So the follow up question is this. If someone has a daisy chained phone extension cable in their house serving several slave faceplates, but their router is connected to the Master socket (with the Master faceplate in place and so not using the BT test socket) their internet speed is seriously compromised by the mere fact that they have redundant cable in their house?

Is there a way around this?

Thanks again,

Andrew

[AABBCC_CCBBAA]

Hi,

I did test each pair using a meter. Also tried different pairs. But see my latest post. It does appear that the other Dave is right in saying that the redundant cable is acting like an aerial perhaps.

Andrew

[imjolly]

if the extension sockets are correctly wired in the master to terminals 2&5 (nothing connected to terminal 3) then that should not affect you connection speed at the master when using a filter.  The speed at extension sockets would normally be lower than that achieved at master socket

[DaveS1234]

Hi,

The speed is faster at the extension socket with nothing else at the master as you have am unfiltered  "straight" electrical run from the incoming line to the router i.e. no spurs.  (Note: you don't need a filter at the extension end if you don't need a phone connection there. Just uses a RT11/RT45 socket to suit your lead - or a router lead with a BT plug)

Re. your follow up question.

If there are extensions wired for possible phone usage then a filtered faceplate should be used with the extensions connected to the filtered punchdowns  and the router connected to the RJ11/RJ45 on the faceplate. This helps prevents interference back onto the unfiltered connections ( and removes the high frequency signals from the extensions)

With the "new" NTE5C MK2 and MK4 VSDL filtered faceplate the filtered connections are on the body of the NTE and unfiltered connections (for one run to a socket for a router) on the removable faceplace - if required)

Hope this helps,

Dave

[AABBCC_CCBBAA]

I am still curious as to why on earth having an extension cable running to a slave faceplate from the master socket faceplate causes reduced download speeds when these are measured at the Master socket faceplate.

Whereas, download speeds measured directly from the slave faceplate are at least 5Mbs better than they are when the same router is connected to the Master socket faceplate?

AJ

[Keith_Beddoe · Accepted solution]

---

@AABBCC_CCBBAA wrote:

I am still curious as to why on earth having an extension cable running to a slave faceplate from the master socket faceplate causes reduced download speeds when these are measured at the Master socket faceplate.

Whereas, download speeds measured directly from the slave faceplate are at least 5Mbs better than they are when the same router is connected to the Master socket faceplate?

AJ

---

Its because you have an unterminated transmission line. Without going into too much detail, standing waves are generated which cause signal to be reflected back out of phase and cancel part of the broadband frequency range.

If you are connected at the end of the transmission line, those standing waves are not generated as the line is terminated.

For reference and bedtime reading.

https://www.allaboutcircuits.com/textbook/alternating-current/chpt-14/standing-waves-and-resonance/

[DavidM]

Hi @AABBCC_CCBBAA and welcome back.

I've had a read through this and there have been some really helpful replies, thanks to all who have posted. Hopefully this has answered your query. Can you get back and let us know how you got on?

Cheers

David