• 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

bottletop said:
Maybe you've got gremlins? Something that I hadn't thought of for a while is the ole bell wire interference on ADSL line. Would accidentally removing that wire help your horrible connection? (You're not supposed to mess with the BT master socket and there is a risk involved if you try etc)
Click to expand...
The Openreach socket front can be replaced by the customer, I believe, and the latest front panels have extra sockets and connections, and the filter built in.

I could never understand why Openreach are allowed to charge £150 to move a 2 wire low voltage connection elsewhere. However, you are allowed to take a feed from the terminals provided.
 
prpr said:
What's the type and state of your extension wiring? And how long?
Click to expand...
At the moment it's just a spool of Cat5, one pair in which is testing the run I propose to use when I get around to it, maybe 15m. It's been in circuit from the terminals in the master socket to an RJ11 faceplate ever since I found it had no effect on my data rate (as you said - it's nothing compared with the 2.5km to the exchange).

In any case, it doesn't matter how long it is or what state - first thing this morning it was working (and has been for many months), but by 11 it had just dropped out. I've not looked into it yet.
 
Have you tried taking the feed directly from the terminals where the line enters the building? You only need 2 wires, anyway. You can run that to a second openreach box.
 
The router is currently plugged directly into the ADSL socket on the master box. The trial cable is wired to the pick-off inside the master box, which is near enough what you are talking about (when I install all this properly, I intend to move the master socket - but that is only a matter of a few feet.

The only other "improvement" I could make would be to wire into the external junction box at the end of the underground cable from the street cabinet - but then that would be obviously DIY, and I very much doubt it would have any effect on my data rate.
 
Trev said:
Is BH lucky enough to have one of those?
Click to expand...
Quickly found where a long alarm cable had been damaged on installation in the place where I was working back in the late 70s using an improvised TDR. Consisted of a function generator and a scope, calibrated for velocity factor against a reel of the same cable.
 
Whatever, it doesn't matter now I know there is something amiss (and bypassed it). Proper installation is pending (indefinitely), but the concept is proven for when I get a round tuit.
 
ConnectionSpeed-2021Feb27-0656.jpg

OMG!!! What is going on??? :frantic: :dance:

SpeedSmart is telling me 4.98Mbps (but it has been up as high as 5.66Mbps in the past).

Normally my sync rate (connection speed) chugs along around the 3,000 mark, and SpeedSmart reports 2.5-ish.

Somebody's bound to ask, so:

NoiseMargin-2021Feb27-0656.jpg

No, the speed hasn't gone up because the noise margin went down - the noise margin went down because the sync rate has gone up. The question is why the line decided to accommodate a higher speed all of a sudden.
 
Looking back through the (intermittent) record, the receive noise margin has been over 10dB without triggering a renegotiation. I missed the jump from 3000 to 4500, but before the margin was around 7-8dB, and after 8-9dB (the graph shows what look like diurnal variations, presumably better when noise generators - ie people - are asleep). Weather might have a bearing on my line quality too.
 
Black Hole said:
No, the speed hasn't gone up because the noise margin went down
Click to expand...
Oh yes it has.
the noise margin went down because the sync rate has gone up.
Click to expand...
No. Speed is dependent on noise margin setting, not the other way round, although obviously they are interlinked, and I guess is based on the error statistics which drive the whole process.
The question is why the line decided to accommodate a higher speed all of a sudden.
Click to expand...
That's the DLM working.
 
prpr said:
Oh yes it has.
Click to expand...
Oh no it's not. We've been through all this before.
prpr said:
Speed is dependent on noise margin setting
Click to expand...
Agreed. But it's not the setting which is plotted in the graphs, its the measured noise margin (while the sync rate remains constant). When the measured noise margin goes up, there is opportunity to negotiate a greater signal bandwidth... and then the noise margin goes down when a greater bandwidth is in use. If you don't agree with that we'll have to agree to differ.
 
Scrat said:
Have you re-synced the router?
Click to expand...
If I did, the speed would go down. My noise margin has gone down to less than 3dB.

The point is: how can Openreach be testing my line speed via a website and a browser and coming up with a figure of 11Mbps when my current sync rate is only 6.2Mbps? Implausible. labs.thinkbroadband.com measured 5.8Mbps. Even that is at the top end of predicted speed on my line length (1.2km), twice what I was getting a week ago (no, I haven't disturbed anything), and about what prpr thought I should be getting. No, I think a miracle has occured and a drop is imminent (I have seen these rates before, briefly), and that Openreach test site is untrustworthy (useful if you're trying to sell your house).
 
Openreach tests the speed up to the router / Home Hub. Thinkbroadband tests the speed up to the computer including the wi-fi or ethernet connection.
 
You must log in or register to reply here.
Back
Top