• The forum software that supports hummy.tv has been upgraded to XenForo 2.3!

    Please bear with us as we continue to tweak things, and feel free to post any questions, issues or suggestions in the upgrade thread.

Options for Domestic Wired Networking / Broadband

I can't vouch for the microfilter, but there's not a lot to go wrong with a master socket! However, what I've read about the new spec for VDSL satisfies me that the latest design should have benefits (otherwise I wouldn't have gone to the bother).
 
Doubled my bandwidth
So I was right then!
Even with 43dB loss, the speed is about half what I would expect.

You might get a further improvement by not connecting any internal wiring to pin 3, this (bell) wire is unlikely to be needed with most modern telephone equipment, if everthing in your house 'rings' without this wire you are much better off not connecting it, as it an aerial you don't want
I said that in post #47...
Having the ring wire connected (pin 3) also gives a single unbalanced 'aerial', which picks up even more 'everything'. This is what decent sockets and i-Plates get rid of.

Are you sure it's RJ-45? I thought these were usually RJ-11
Edit: It is RJ-45 (although obviously an RJ-11 plug as used with most ADSL routers will fit)
Are you sure? Post #59 seems right to me...
In this picture RJ-11 (6way) is on the left, BT631A (BS6312 6 Way) in the Centre and RJ-45 (8way) is on the right
 
Now what...

4137

The router stats aren't showing a problem, and browsing feels responsive, so I guess this is a SpeedSmart server issue.
 
Now perhaps I can think about moving it somewhere more convenient.
Is there a "better" cable I can use from the junction box to the master socket? At the moment, I've got standard 6-core single strand twisted pair in telephone colours.
 
1.9 download rate this morning (confirmed with a different speed checker), 6.2 line rate. What's going on?
 
A mate of mine reckoned it might train up!

The line rate reported in the modem stats is still > 6000kbps, so something else is throttling it. That leaves the ISP and the WiFi.
 
Hmm.... go into range of my 5GHz WiFi and the rates go north of 5Mbps again. I checked for channel contention on the 2.4GHz band and moved to an empty one with no significant benefit. Maybe there is non-WiFi 2.4GHz interference slugging things.
 
This is a post I made over on SNB forums earlier this month after I was experiencing 2.4GHz issues.

'I had fitted a combined USB and mains socket in my office. I thought it was great, for a short while. Then I noticed my IoT devices on the 2.4 band had become very flakey. It was really bad and frustrating. In the end, I used an old trick.
Using an old AM radio tuned to 615MHz, I literally when round the house listening for interference.

Turns out that this USB/mains socket had gone faulty and was generating lots of RF interference. As it was hard wired into our mains supply, the whole house wiring had become a transmitter.

Ironically the USB unit was functioning perfectly and gave no other indication of the problem.

I was blaming neighbours!

It could be a ‘wall wort’ PSU, faulty LED light bulb etc.

I’m just posting this so that you can realise that things can change without you being aware.
'
 
USB mains sockets are horrible things

There's not enough room in there for decent filtering on the mains side so they pump crap into the wiring. One a neighbour has made the use of Homeplug devices impossible but they wouldn't take it out.
 
I know, but the speed test apps run on the iPad...

It's only now I have improved my line rate that the bottleneck has transferred to WiFi!
 
I know, but the speed test apps run on the iPad...
:o_O: You can run non-app speed tests from your PC.

(I've used thinkbroadband's as I've been a member for years and it keeps the history. When we moved to Zen a few years ago I discovered that it's the one they use/recommend too.)
 
NTEs have had a bell wire filter in them for some time to eliminate this problem. I've not pulled one apart but there's clearly a bobbin inductor in there.
Yes they do, but it doesn't eliminate the problem, it is an attempt to reduce rf problems when using the 'bell wire', it will never be as good as disconnecting this redundant wire completely
 
You should be testing with a cable, not wireless. Unless you are trying to test the wireless itself. :rolleyes:
Just done a test here using Ookla.

On a reported VSDL link speed of 51Mbps and a maximum data rate of 50.3Mbps, a laptop came in at around 42Mbps on both 2.4GHz and 5GHz 802.11n and a gigabit-wired PC came in at 48Mbps.
 
You can run non-app speed tests from your PC.
...which is also connected by WiFi and doesn't have 5GHz. Yes, I know I can wire it up, but like I said until now this hasn't been a problem, I'm not on my PC most of the time, and actually all I need to do is look at the router line stats.

I left the PC on overnight, collecting a history of line stats. The 3Mbps never really bothered me too much for what I do, what really annoyed me were the drop-outs. Waiting to see if I still get them.
 
Last edited:
Back
Top