Frequent dropouts 25th Jan 2025

[#1]

Yes running straight to the Openreach Master TEST socket would be the ultimate solution.

.

[#2]

If it helps, I think this is where I bought my short RJ11 to BT431a CAT5e cables from - eBay : COMMS4u.NET 

I'm currently using the "Solid copper white CAT5e 0.5m" (as the cable is fixed to a wall), but also own the normal flexible CAT5e version - which work just as well.

.

[#3]

@Dan_the_Van and @outcast 

Since I have a SoGEA connection the idea to remove the faceplate permanently and plug directly into the test socket sounds good. I may try that idea. It won't look neat but it is hidden behind the desk and protected by the wooden foot bar I installed so not a big problem.

I don't shop on eBay and I can't get a 1m long BT431A to RJ11 cable at a reasonable price so I have ordered a BT431A to RJ11 adapter at a quarter of the cost. I'll plug this adapter into the test socket and I will plug my new DSL cable into that: -

https://www.kenable.co.uk/en/home-cctv/telephone-cables/phone-adapters-ends-tools/6037-bt-431a-plug-...

[#4]

Whereas I have always run my BB and landline phone using a BB dongle filter from an LJ10 extension socket which is itself a master socket - complete with voltage surge arrester - connected to the actual Openreach 5C Master socket, which is never directly used, unless for test purposes if a fault is present. i.e. I am running with two unfiltered 'master sockets' in series. 😊

It's never done me no 'arm!

To be fair, I should point out I am on the old restricted 10/40 VDSL BB, so have plenty of SNRM headroom and no speed issue with the above setup. Possibly it might have some effect if I were on full unrestricted VDSL BB, which would be around +70Mbps according to my local cabinet.

[#5]

@outcast and @Dan_the_Van 

I finally got the BT to RJ11 cable, removed the face plate and plugged directly into the test socket. It was much worse so I put the face plate back on and speeds are okay again. This brings my experimenting to a conclusion so I will be leaving the face plate on.

Thanks for all help but FYI here are some before and after photos.

Before (plugged into filter plate)

Plugged into test socket

After face plate refitted

[#6]

@Madeleyite  It's pretty obvious that the BT to RJ11 cable does not have the correct connections, the attenuation figures are way higher than the face plate connection so the speeds are very low.  I would guess that only one of the incoming pair of wires is connecting from the test socket to the router.

Your decision to keep the face plate on is probably the best choice, I don't think the direct connection from the test socket (if it was properly connected) would have given any significant improvement to your speeds.

[#7]

Thanks for that information. I'll unsuscribe from this thread now. If I have instability issues in the future I'll create a new thread.

[#8]

That is interesting.

It is one thing to have a piece of wire with the right end connectors. but that wire also needs to be suitable for the intended purpose.  Looking at the marked differences in the router stats, it would be reasonable to conclude that (unlike the supplied RJ11-RJ11 cable) the BT to RJ11 cable is not twisted pair, being intended solely for use as a telephone cable, not a data cable.

You clearly appear to have environmental RFI (aka REIN) which to some extent you have mitigated (but not eliminated) by creating some distance between electrical components and the router cable.  You might see further improvement if you can eliminate the source of the RFI (replace a marginal PSU) or by obtaining a high quality shielded RJ11-RJ11 such as the one mentioned earlier.  There is a page on kitz.co.uk which might be helpful - but right now I cannot access it: looks like the domain was renewed on the 11th Feb but has dropped out of Plusnet's DNS.

NOTE: The "This fixed it" button is not intended to indicate that the issue is resolved (as some posters do by marking the last post) but rather the post which helped the most to reach a resolution.  The intention is that other users encountering similar issues can readily find the advice which lead to the solution, without needing to read thorough the whole thread.  Was there a particular post which helped you arrive at a satisfactory conclusion?

You can change your mind by going to the "This fixed it" post and using post options to select "This is not the fix".

[#9]

@Townman  I don't think RFI can be blamed for a such a huge change in attenuation, in my opinion the cable isn't connecting the correct pins together. Since the faceplate filter is a straight through connection for the VDSL2 signal there is very little to be gained by using a connection direct to the test socket, whatever cable is used.

[#10]

Here is the link - ::. Kitz - REIN, SHINE & RFI .::

[#11]

I agree with the previous few posts as to why using the test socket is apparently worse than the filtered faceplate.

Just another hypothesis -

When the faceplate is in circuit, the upload sync speed of 10Mbps is unusually low for a line with a 66Mbps download sync, therefore there is potentially either something wrong with the faceplate or the incoming line.

Having removed the faceplate and the sync speeds got WORSE !, then that would indicate either the BT-RJ11 cable is faulty, a test socket contact is corroded, or the incoming line is unbalanced in some way.

I'm wondering whether fitting the filtered faceplate is masking an unbalanced line, with the voice filter's impedance altering the line termination characteristics, and 'pulling' the copper pair signals 'together' and lessening the 'unbalance' ?

An unbalanced landline cable with only one wire working properly, might also explain why this connection is so susceptible to local EMI/RFI.

.

[#12]

---

@outcast wrote:

Just another hypothesis -

... or the incoming line is unbalanced in some way.

I'm wondering whether fitting the filtered faceplate is masking an unbalanced line, with the voice filter's impedance altering the line termination characteristics, and 'pulling' the copper pair signals 'together' and lessening the 'unbalance' ?

.

---

... and of note, an unbalanced line is more susceptible to REIN, which would fit with the hypothesis.

[#13]

@Townman @outcast 

Can I remind you that @Madeleyite has stated

"This brings my experimenting to a conclusion so I will be leaving the face plate on."

"Thanks for that information. I'll unsubscribe from this thread now. If I have instability issues in the future I'll create a new thread."

Maybe time to respect the OP wishes?

[#14]

@Townman  Yes we've all seen that and it's irrelevant because the problem is an incorrectly wired or faulty cable. Please can you explain how the downstream line attenuation changes from 18.7dB with the RJ11 to RJ11 cable to 57.3dB with the BT to RJ11 cable. It's nothing to do with Rein or Shine etc, it's a cable problem. That change in attenuation is equivalent to a significant length of cable. Maybe a mile or more. 🙄.  You would probably get a better connection with a couple of lengths of wet string as demonstrated by RevK at A&A in the past.  ADSL over Wet String  😀 

Edit -  OK @Dan_the_Van  I'll shut up now.