Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[NewtronStar]

John I don't know why you think zxcvbnm can't go any faster than 24Mbps as a cleanish line @ 1000 meters can get close to 34Mbps and add a 8800NL or Zyxel into the mix and that may gain another 3-4 Mbps.

The Internal wiring would need to be in good nick and hopefully the D-SIDE is mostly all copper to the PCP cabinet, A good OR broadband engineer visit should be able to help zxcvbnm find what is holding back his line performance.   

[j0hn]

John I don't know why you think zxcvbnm can't go any faster than 24Mbps as a cleanish line @ 1000 meters can get close to 34Mbps and add a 8800NL or Zyxel into the mix and that may gain another 3-4 Mbps.

The Internal wiring would need to be in good nick and hopefully the D-SIDE is mostly all copper to the PCP cabinet, A good OR broadband engineer visit should be able to help zxcvbnm find what is holding back his line performance. 

No idea where that comes from. Line length was only disclosed a couple posts ago and I've given him sensible advise throughout this thread.

Thank you to burakkucat and J0hn and the other helpful suggestions. It is very nice to have hope there may be a definite findable and fixable problem

You're welcome. As I previously said there's definitely a problem showing on the Hlog and any decent engineer should be able to see it.

[zxcvbnm]

Oh Sod, I have rather failed to check the obvious, sorry, unhelpful.

Some of my extension sockets are still live with a dial tone, a handful of unused ones behind beds upstairs. I really thought they were all off, the main used ones were, but I should have checked more thoroughly.

Not to mention where are they connected? My original master socket is definitely dead. The current new master socket has two wires coming into the back from one cable and one cable with three disconnected wires. Nothing else I can see.

The wires coming into the house have lots of jelly crimps but only two wires in each crimp so presumably are straight through?

So presumably somewhere under the floor boards between my so called master socket and where the phone comes into the house must be some sort of spur or connection to other cables..

You would think the fibre install man would be able to detect this sort of thing. Perhaps not. I have some sympathy its hard to know what bt put under the floor decades ago.  But it does mean I now have a problem that I'm technically not allowed to touch presumably as its their side of the master socket...

Ah well on to tomorrow to see if I can get floorboards up with a torch. With my luck there's probably a 30 year old bt junction box sealed under cement somewhere.

[zxcvbnm]

Yup it would seem I have led you all up the garden path. Ah well. After starting where the cable comes into the house, goes under the floor and then vanishes through a joist, we found, on lifting the fifth random floorboard, a six screw Ashley electrical junction box. A round brown circle. With Four wires into it, one of which on tugging appears to be the main cable into the house.

So I guess this is the star wiring people talk about. Everyone denies knowing it was there. Some happy electrician must have upgraded bt's work something in the 80's or early 90's I would guess. There must be other junctions on the wire but we have ripped up enough of the floor for now. Some wires head off toward areas with no sockets suggesting I am my own bridging tap.

I guess I will have to convince plus.net they should send someone back to rectify the mistake the installer made in the first place. If not the strong temptation is just to try and quietly move the master myself, how hard can it be. But I must of course resist such perfidy.

Anyway sorry for somewhat wasting everyone time, though still I now know what a bridging tap is and a lot more about telephone wiring than I did before.

[burakkucat]

Oops! And yuck! At least we can now see the problem.

Please update this thread, as and when adjustments occur. 

[NewtronStar]

If not the strong temptation is just to try and quietly move the master myself, how hard can it be. But I must of course resist such perfidy.

Using CW1308 cable from where the BT pairs enters the premises to your designated Master Socket can be straight forward or difficult in your case with cable being under the floorboards it's just not easily accessible to an OR engineer and you don't need those extensions around the house as DECT wireless phones do the same job

[zxcvbnm]

Well plus.net suddenly called up and agreed to send an engineer! Lots of warnings how I could end up being charged for my indulgence but fingers crossed.

He turned up today and was actually very helpful. In hindsight I wonder if he would have diagnosed the problem if I just said my broadband was slow. But taking him to the master socket and saying look its clearly in the middle of my extensions why not move it to where the cable enters the house, he agreed completely. He even put an old extension socket of  ours back into the empty hole.

Fortunately where the cable enters the house there is about a foot of exposed cable where the outside cable is joined to the inside in a little box so he was able to just join onto that and put the master socket in the window. Nice to have it out of a bedroom really.

Didn't seem to have to go to exchange for dlm problem? Just pressed buttons on some gadget that then said dlm reset.

So far speed is up from 18  to 38! From 1.2 up to 6 up! So twice the down speed and five times the up speed! Snr margins are down to close to 6 and power has dropped dramatically from 5.6 to 2.6.

My hlog is looking smoother, still an odd squiggle and the end but there you go.

Oddly I can't get a download speed test to go faster than about 16mb/s? But I am mostly in it for the much better upload speed so I will let that lie for a while I think.

So all a great success I think. And now I have some confidence my phone line is working correctly I can finally feel freer to move to someone cheaper (: Maybe I will splash out on one of those cat5 modem cables.

Otherwise I'm going to try and not fiddle with it for a week and hope it adjust to the max possible speed if I leave it alone.

Thanks for all the thoughts. I can't believe after two years I have actually got somewhere...

[j0hn]

Fantastic result.  The Hlog looks perfect now. The "squiggle" at the end is normal.

The engineer will either perform the DLM reset using his iPhone, or by calling the DCoE. There's no need to go to the exchange.

Regarding the slow throughput, is that in a wired state, or wireless? Always check in a wired state (via Ethernet) if possible. My 1st step would be to check your I.P profile from Plusnet/BT. The banding is definitely removed and the line should give close to full sync speed.

[burakkucat]

I'll echo j0hn's congratulations. 

So, looking at the new Hlog plot, we can now see that the US0, DS1, US1 & DS2 bands are all present with the typical curve v frequency shape. US2 is absent and DS3 will be assisting the cumulative DS to some extent.

When the QLN plot is examined, the absence of US2 and partial DS3 in the Hlog plot is not too surprising.

Just to clarify, the service feed enters the building and is now directly connected to the NTE5/C (fitted with a SSFP)? No extension wiring is connected to the NTE5/C? 

[NewtronStar]

Hlog looks sweet could even get confused that I am looking at my own 1000 meter hlog, ok the errored seconds have increased with more sync and as your on a ECI cabinet without G.INP

Your line is using the DLM profile of fast allowing you up to 2880 ES per 24 hours so keep your stats going as long as you can, yet it looks like 300 ES per day as a prediction since you last resync   

[burakkucat]

Hmm . . . I see that there was some disturbance in the SNRM plot, earlier today.

I'll leave the analysis of that plot to N*Star, j0hn or A.N.Other. (I suspect that is evidence of a cross-talker powering down her/his modem -- between 2108 hours, Wed 16 Nov and 0728 hours, Thu 17 Nov.)

[NewtronStar]

Can see two crosstalkers in that SNRM plot the first one changes the SNRM by 0.9dB the second disturber is much stronger seen as a large upward spike which changed the SNRM by 3.0dB for a minute.

And would be happy with your errored seconds on this line of 1030 meters which is using fastpath I couldn't even hold on to fastpath on my line on the DLM standard profile as it was hitting just over 2000 ES per day.

[zxcvbnm]

Thank you it is rather pleasing.

J0hn I did do speedtests connected with ethernet and wifi turned off. As I said I'm not too worried, if its still stuck in a month I might summon more energy to pursue it. 16mbps is enough for anyone...

burakkucat I'm afraid my extensions are all turned on again. But now the mk4 master socket is at the very beginning of the line I hope they are all kept separate by the filter this time. I did try connecting the modem directly to the test socket for a few hours midday today but I think that was a mistake so I put it back. Still I was able to check the ring wire was happily missing as I had asked. The installer had a little trouble connecting the extension wires in the push down crimp socket so I am going to avoid touching them again incase the flipper is getting a little worn out... I appreciate someone suggested you just need dect handsets now but even with a full set of six we somehow still need a few extra sockets... Maybe someday.

NewtronStar my line is a km, well estimated. But I like to think that being a rural km the lack of people connected to it will give me a slight advantage over an urban line. Hundreds of meters of the cable is over fields with nobody connected to it.

Interestingly having avoided having a ring wire our big button bt phone would not ring which was a surprise as it is fairly modern? Anyway easily fixed with an old microfilter but does show ringing is not as ubiquitous as you might expect.

Anyway fingers crossed if I can force myself to leave it alone it may tick up a little more. Mind you I'm always suspicious, with not hard evidence, that our phones are worse in heavy wind and its blowing a gale at the moment.

Thanks again for all the support.

[burakkucat]

Regarding the IDCs . . . They are designed to take a maximum of two wires only.

So in the case where more than two need to be connected, the usual advice is to as many as necessary three way gel-crimps and house them in an empty case of a BT80. Basically one reduces the number of wires per pair down to one and then just have one cable taking the pair from the gel-crimps in the BT80 to the SSFP of the NTE5/C.

I've re-read the above and can see it is not that clear . . . If necessary, I will try to create a diagram to illustrate the idea.

[NewtronStar]

What happened in the wee small hours zxcvbnm your line had over 7000 ES and your Interleaved