Print

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

[limbo]

Hi everyone,

Thanks for your feedback.

I popped over to my next door neighbours and they let me test their broadband speed. For some reason the BT Speedtester was not working (and in fact on this connection it is just very slow). I ended up using speedtest.net on the same server as I do- they got 2.42 Mbps down and 0.37 Mbps up, with a 50ms ping. That sounds more reasonable!

I've attached yet another SNR graph in case you didn't have enough already! This captures a nice drop and then increase again.

Also, does anyone know why my IP profile for uploading is faster than for downloading (this does seem to be a strange side effect of the problem)?

Cheers,

Thomas

[attachment deleted by admin]

[burakkucat]

I don't know what's inside that grey box on the outside wall, but the wiring to it looks very amateurish, and I can't believe that BTOR would have left it in such a state. Perhaps someone else can identify it for us. If it's just an external junction box, then I think you need to get BT to regularise the wiring from there to the NTE5, and if it was the work of an unofficial person then I guess you'll be charged for that job.

The grey box is a standard Cover 101A and below it is a length of Capping 25. Inside the grey box should be gel-crimps connecting the active pair of the service cable (the black cable) to the feed-in cable (white).

In theory there should be a Connector Bend at the top of the service cable duct (visible in the paving slabs), then a length of Capping 25, all topped off by a Cover 101A. The hole should have been drilled through the wall from under the Cover 101A. As it currently exists, everything that could be wrong with that installation actually is!

[limbo]

As it currently exists, everything that could be wrong with that installation actually is!

Hi b*cat,

What I find surprising (not unhappy with him- I guess he was focussed with the attenuation data which he took) is that when the OR engineer came a year or two ago, he said that he was only looking at the test point to the exchange (no internal wiring- which is how OR operate). But he didn't spot any of these errors which you guys pointed out straight away.

I guess it's up to me to get the OR engineer on Monday on my side (with deployment of tea/coffee, biscuits and bacon butties!) to try and "clean up" the external point.

Stock up with chocolate biscuits, bacon (sandwiches, for the use of), tea and coffee. Then as soon as the engineer has arrived, show her/him the outside "folly" and say that in your opinion it does not look correct -- what does (s)he advise?

Hopefully the engineer won't feel like I pounced on them when they arrive! 

Does anyone know what the chances of them helping out are (guess as you say, it totally depends on the engineer)?

Cheers,

Thomas

[burakkucat]

Chances:

50%     50%

If I was the engineer tasked with your FTTC installation, I would:

(1) Externally -- Strip off the Cover & Capping, identify the three whitish cables present, remove the existing feed-in (white cable) replacing it with new, silicone seal it where it enters through the brickwork, make a nice u-bend drip loop, re-make the gel-crimped connections from the service cable to the new feed-in cable and then finally fit the correct Connector Bend, Capping & Cover working from the service cable duct, upwards.

(2) Internally -- Replace your existing NTE5/A with a new OR banded one. Check for POTS service and sync to the DSLAM at the test socket with my JDSU HST-3000C. Eat bacon sandwich. Drink mug of coffee or tea (depending if a morning or afternoon appointment). Discuss with you where you would like to have the modem sited, where you would prefer the router and where you have your computer. The modem would be wall mounted (highly recommended, to ensure efficient cooling), a SSFP would be fitted to the NTE5 and any necessary data cable deployed. The current 2x6 wires attached to your existing face plate would be left disconnected and tidily coiled at the bottom of the NTE5 backing box. The new face plate would be re-affixed to the SSFP. I would then check, again, the POTS service and the DSL signal with my JDSU. Assuming all is well, the modem would be connected and powered up. Presuming that you had tempted me with a chocolate biscuit, I would suggest that you now connect your ISP / CP supplied router, power it up and see if you can obtain access to the Internet.

Well, you did ask . . .

[limbo]

Hi,

Thank you b*cat, I hope that the engineer has a soft spot for biscuits and bacon sandwiches!

I found out which PCP I am on, and where it was- it is the one which I thought was connected to the adjacent exchange- apparently not! If the cables go via the main road, I estimate a cable length of 750m. It's a rather long route, so if they cut across roads (the way the footpaths do), I reckon it's nearer 600m. If it took an even longer way round, we would be talking 800m. Thinking about it, I think the cable must come down the road, in order for it to be a sensible route. In a straight line it's 400m.

Assuming the line is in good working order, what speeds should I expect from (what I believe ought to be the length of the line) 750m?

Thanks,

Thomas

[Bald_Eagle1]

Assuming the line is in good working order, what speeds should I expect from (what I believe ought to be the length of the line) 750m?

Hi Thomas,

I don't wish to step on b*cat's paws, but if I may be so bold, I will just step in for a moment.

That looks like such a simple question that should have an even simpler answer.

20 pages into the other thread, we still do not have a definitive answer, or even a decent guesstimate.

Your pre-installation speed estimate is/was 11 Mb.
God only knows what it was based upon, & I would put a month's salary on him not being 100% sure.

BT can't/won't explain.
Plusnet say it is based upon line length, but then quote the length right from the exchange & then clarify it by saying it also takes other factors into account, but admit they don't know what the other factors are.
Near the end of May, my estimate from BT was that FTTC is not planned for my area within the next 6 months!
Plusnet's estimate was 14.6Mb.
At first I achieved actual downloads of 33Mb, probably syncing at around 35Mb.
My download speeds have been up & down like a yoyo for 2 months, currently around 20Mb, & for a while as low as 7 Mb.
Plusnet have claimed more than once that as current speeds are more than the originally estimated speed they can't/won't do anything about it, but please see my latest update in the other thread.
In other forums, some users have achieved 37-38Mb downloads over 600-650m.
You already appear to have some line issues that "hopefully" will become irrelevant once your broadband line comes from the nearby cabinet rather than all the way from the exchange.

Does that clearly answer your question? No, I thought not.

If you ever do manage to find a proper answer, please let us know & pass us the 6 winning numbers for the lottery while you are at it.

The only answer available at this time is - "it depends".

So, good luck with the installation, record the stats from the engineer's JDSU or other device & please post them on here for others to absorb.


Paul.

[burakkucat]

Assuming the line is in good working order, what speeds should I expect from (what I believe ought to be the length of the line) 750m?

Having dispatched a rat ( nasty creatures), I drew its entrails with my especially sharp disembowelling claw and threw them up into the air. Once landed, I then studied them intently for 19.7 seconds and now I am able to make this prediction --

40 Mbps > downstream sync speed > 20 Mbps

I leave the rat remains for a passing bird of prey to tidy-up.

The best advice I can give, Thomas, is hope for the best but be prepared for the worst.

With regard to the length of your D-side copper, I would be tempted to assume twice the straight line length from your home to the fibre cabinet -- hence I'd work on the assumption of 800m (until told otherwise).

[waltergmw]

@ Paul,

Were you of a naval disposition the words following "depends" are usually "on the angle at the dangle". However as this is a family-friendly site so I shall not continue!

One of our major problems is that we are aware of a single constant "uplift" figure for all services from a particular PCP. It seems quite preposterous that such an approach is suggested when we all know that different D side lengths and line conditions can and do cause significant variations. Quad erat demonstrandum !

Kind regards,
Walter

[limbo]

Hello,

At Walter's suggestion that my old DG834Gv1 might be knackered, I've switched over to using the 2Wire.

Interestingly the DSL upload rate has increased (although it may have been the cut which caused the DSLAM to start syncing at a higher rate), but the download rate is staying where it is.

The Downstream SNR looks much lower than with the DG834G- at around 15dB, yet the speed has yet to increase. I notice that it is running on ADSL2 (I think)- does that have any adverse effects, or should I stick with it?

Thanks,

Thomas

[attachment deleted by admin]

[waltergmw]

Hello Limbo,

It does appear that your bRAS has stuck at a very low level as your maximum rates are actually above what you might normally expect according to the Kitz estimator.
Your 2Wire data does indeed show you are on an ADSL2+ service which, at your line distance, is probably unwise.

However as you are about to obtain a FTTC service it's probably better to wait for that installation rather than complicate matters with further enquiries.

Kind regards,
Walter

[limbo]

Hello Limbo,

It does appear that your bRAS has stuck at a very low level as your maximum rates are actually above what you might normally expect according to the Kitz estimator.
Your 2Wire data does indeed show you are on an ADSL2+ service which, at your line distance, is probably unwise.

However as you are about to obtain a FTTC service it's probably better to wait for that installation rather than complicate matters with further enquiries.

Kind regards,
Walter

Hi Walter,

Thank you for confirming this. I'm glad that there is potential on this line!

Does the modem calculate the "maximum" speed? It seems to have been changing from 4.2Mbps to 3.6Mbps. Currently it's sitting at 4Mbps.

I just power cycled the modem, and the upload speed has marginally increased to 1Mbps (from 976Mbps), but the download speed is still at 575Kbps. I wonder whether the ip profile might be "stuck" on this low speed, but seeing as there's less than 24 hours on this ADSL connection left, I don't mind!


When the engineer comes over tomorrow, I'll see how obliging he is with our external wiring! I'll take as much data as I can from him, and post them here once it's up!

Many thanks,

Thomas

[waltergmw]

Hi Limbo,

A good plan. If you are able to capture the full diagnostic data complete with the bit loading diagram that would be be useful ammunition as well. (Just in case you get into Bald Eagle mode ! )

Kind regards,
Walter

[limbo]

Hi Limbo,

A good plan. If you are able to capture the full diagnostic data complete with the bit loading diagram that would be be useful ammunition as well. (Just in case you get into Bald Eagle mode ! )

Kind regards,
Walter

Good morning everyone,

Well BT Openreach have now been and gone. The engineer (in his Superfast Fibre van!) wasn't an ex-army chap, but he specialises in installing and fault-finding FTTC.

The engineer turned me down on the offer of a bacon buttie  ! I explained to him the problems we had been having. He then went off to the PCP to wire in the DSLAM. When he got back he said that in the cabinet one of the wires of our pair was broken and held together by the insulation, which when he touched snapped in half. He said that it may well have manifested itself as a HR fault. He seemed surprised that we hadn't heard any crackling on the line, but said that might explain a bit of the "white noise" which we had been hearing. He didn't seem to know the length of the line to the PCP, but said "I think you'll get faster speeds than 11Mbps- that's very conservative".

He took a look at the external wiring, and took the front face off. Seems that the two wires to the right simply pass through, and do not touch the incoming wire. He re-crimped all of the incoming wires- I think there used to be another line, as there were 4 incoming wires (did it so quickly that I didn't ask him to just remove the seemingly unused pair!). He didn't replace the wiring to the NTE, but tucked it all in to the Cover101A, so there are no bare pairs showing now. He also replaced the NTE with a new OR branded plate. He looked at the internal wiring and said "well that's all wrong!" and redid it properly for the new plate. Interestingly, he did wire the bell wire in.

He then got his JDSU out, and said that everything looked OK. He then let me take a picture (see attached). The JDSU reckons a line length of 0.9km (how does it estimate that?). Apparently PN don't pay OR for a 'managed install', so technically all he was meant to do was plug the modem in and run, but he wanted to check the line speeds as well.

All of the diagnostic data he took was on the JDSU- he got it out and checked that there was a reasonable speed. When I asked about bit loading (I may well have explained it wrong), he didn't seem to follow me. Lastly he got a LTOK, and that was that.

So this is what I am currently getting:

I would attach some BT Speedtester results, but I did one a little earlier (stupidly didn't take a SS of it) and apparently I need to wait an hour to do another one.

How come the IP profile is at 27Mbps (looking at the JDSU)- should it start at 40 and work its way down to a stable level, or am I misunderstanding something?

Sorry I couldn't get more data, I think he saw this as a very straight forward install and so didn't do much diagnostic stuff. I do hope the JDSU data is useful.

Kind Regards,

Thomas

[attachment deleted by admin]

[limbo]

Oh dear that photo of the JDSU is terrible! Here's an uncompressed version which hopefully will look a bit better:

[Bald_Eagle1]

I explained to him the problems we had been having. He then went off to the PCP to wire in the DSLAM. When he got back he said that in the cabinet one of the wires of our pair was broken and held together by the insulation, which when he touched snapped in half. He said that it may well have manifested itself as a HR fault. He seemed surprised that we hadn't heard any crackling on the line, but said that might explain a bit of the "white noise" which we had been hearing. He didn't seem to know the length of the line to the PCP, but said "I think you'll get faster speeds than 11Mbps- that's very conservative".

He then got his JDSU out, and said that everything looked OK. He then let me take a picture (see attached). The JDSU reckons a line length of 0.9km (how does it estimate that?). Apparently PN don't pay OR for a 'managed install', so technically all he was meant to do was plug the modem in and run, but he wanted to check the line speeds as well.


So this is what I am currently getting:

I would attach some BT Speedtester results, but I did one a little earlier (stupidly didn't take a SS of it) and apparently I need to wait an hour to do another one.

How come the IP profile is at 27Mbps (looking at the JDSU)- should it start at 40 and work its way down to a stable level, or am I misunderstanding something?

Sorry I couldn't get more data, I think he saw this as a very straight forward install and so didn't do much diagnostic stuff. I do hope the JDSU data is useful.

Kind Regards,

Thomas

Well Thomas, you're not stuck in limbo any more then 

The engineer seemed to be quite helpul & obliging & hopefully your various line issues have now been permanently fixed.
I'm not sure about the bell wire though. Someone else may advise you to remove that???

It should take around 10 days or so for your speeds to stabilise.
Are you happy with the current speeds? They are a lot higher that your "meaningless" estimated speeds.

I'm really glad that you managed to take a photo of the JDSU showing the estimated line length. None of my visiting engineers have shown that display to me. More than one visiting engineer told me that the JDSU could not display that sort of detail when I asked them for it

I will be using your photo as evidence (if you have no objection) when the next engineer vists me, hopefully during this week.

I believe that line length is estimated via the JDSU mainly based upon on the attenuation figure, but more knowledgeable people than I aren't really quite sure for VDSL2 purposes. It may well take other factors such as SNR levels into account.

VDSL2 as delivered by BT, known as Profile 8c, Annexe B has the downstream split across 2 frequency bands, the lower band is more or less the equivalent of ADSL2+ frequencies, & the higher band is used for VDSL2 higher frequencies.
It is thought that the attenuation as displayed on the engineers' JDSUs must be some sort of an average value for both bands.

Some other countries use higher frequencies, with more power, split across 3 bands.

Nobody appears to be quite sure what typical attenuations for a given line length for VDSL2 purposes should be yet.
It would probably depend whether it is all copper, or part aluminium cabling. Both have different attenuation values.

Due to your actual or estimated distance from the cabinet, it is very unlikely that 40 Mb would be delivered to your home.
Others with guesstimated line lengths of around 650m & a JDSU attenuation value of around 19dB or so have been achieving download speeds of around 37Mb, more or less the full whack available, less a bit for overheads.

The IP Profile & download speeds as shown in the BT speed tester & the engineer's JDSU apparently always appears a bit more pessimistic than actual download speeds as shown by other speed testers such as speedtest.net.

Your US & DS LIne Atten values are identical, which is interesting for me as there is a large difference between the two for my line, the US being much higher. The signal attenuation is probably the more important value though.

All I would suggest is that you regularly monitor your download speeds as reported by speedtest.net for consistency, along with regular BT speed test results (save the screenshots).
If nothing else, you will have a good record of your connection's capabilities if your speeds should start to plummet outside the 10 day training period. Your speeds may well even increase over the next few days.

EDIT: Just for curiosity, which version of the modem has BT supplied for you?
It should be displayed on a sticker underneath the modem, the firmware version ending in SP06 or SP10, & the hardware version may have a 2V or 2B sticker, also underneath.

Also please see the PM that I sent to you

Paul.