@John: I picked 250m because these were the average measurements that adslmax mentioned from his post.

Higher noise margin tends to suggest lower line noise! If a 300 meter line cannot achieve higher than 60Mbps then there is something wrong either with wiring or it is using aluminum instead of copper or very high crosstalk.

Lower noise margin like 3dB for example may achieve higher sync speeds but has a higher risk of internet connection drops and sync errors unless there is absolutely no line noise! But there comes a point where you can't achieve 80/20 even with 3dB and that is when line is too long like in your case at 1km.

I compare with ADSL figures because ultimately FTTC is not pure Fibre, it is still very much dependent on copper length, quality, crosstalk, etc. Until we do not have FTTP in this country, FTTC will very much be reliant on our copper length and quality to our cabinet.

You have a noise margin of 6.3dB at 1km, but is it syncing at the top speeds? I think you previously said you were getting 40Mbps a few months ago when we had this discussion on line length.

If you are taking lower FTTC package I would expect that naturally you will have higher noise margins at your line length compared to a package where the router is attempting to sync at higher speeds.

Here's what happened when I had 3dB with ADSL, was getting 18Mbps but the internet connection would drop out after a few hours. At 6dB I was getting 15-17Mbps and it would still drop out every 2-3 days, but the noise margin continued to default at 6dB. This vicious cycle continued for years, I didn't even come on these forums as I knew it was pointless consulting for help. No BT Openreach engineer was able to solve this problem all these years!

Finally a manual cap of 9dB profile made my connection stable but it was syncing at only 11-12Mbps connection. And guess what? At certain times of the day (particularly mornings) it would drop down to as low as 4dB noise margin due to line noise, but after half an hour or so the noise margin recovered back to 9dB this is why my connection did not disconnect.

And yes there were noise problems that could never be resolved last 15+ years in this property. Actually I live last 30 years in this flat, but ADSL only around 17 years or so, before that we didn't have ADSL broadband except dial-up.

This was a real problem. Because as I describe a drop from 9dB to 4dB suggested that the problem was still there, except that it stopped causing my connection to drop out. At 6dB or 3dB default profile in my case was absolutely impossible to achieve a stable connection regardless on what kind of router I used! It was like that with Tiscali, Be Unlimited*, Sky, Plusnet last 17 years or so since having ADSL even when plugging via test socket.

Briefly I lived in one of our other leasehold property for 4 years during my college days between 2005-2009 and I was getting 7dB SNR and 16Mbps in these 4 years I never had problems with ADSL drop-outs at all. Returned back to our old property here and the disconnection problems returned yet again.

In certain situations like in my case, there was simply no cure. 1300 meter to BT exchange only line, but lots of noise and drops outs.

FTTC came as a lifesaver, too little, too late though. This is the first time I'm observing stable internet connection ever! Now even if theoretically my line wasn't at 320 meter to cabinet but let's just say it is 380 meter to cabinet due to extra wiring, still it's going to eliminate most of the copper line where most likely the line noise is stripped away due to shorter copper length.

I very much doubt I will be on 500 meter length, though. I don't think I would be syncing at 80/20 Mbps and the BT Wholesale checker would not show 68.2Mbps to 80Mbps on a 500 meter line, prove me wrong here!

My parents are nearly 500m away and his DSL Checker estimated for their line is between 79.2 and 55.8 and his line sync are 53.8 with a SNR of 4.7dB with line attenuation of 16.1dB

Exactly! As we can see the low estimate of 55.8Mbps is an accurate indicator that the line length in their case is around 500 meters with a max of 79.2Mbps.

Now I don't know why in my case the checker would show 68.2Mbps-80Mbps but my line attenuation shows 17.9dB. That's the only thing I don't quite understand.

The engineer showed his device 70Mbps minimum 83Mbps max for my line. Your parents getting 53.8 Mbps with line attenuation of 16.1dB. Line attenuation info in router stats may not always be accurate indicator of distance.

With a telephone number the checker will be more accurate. But I get this "There is no data available for this number. This could be either because it is not a BT line or it is a new BT number that has just been provided. Most new numbers will appear on the checker 24 hours after BT has installed the line "

This is ever since I changed my telephone number for the new FTTC service, because I used to get spam phone calls and voice mails in my old phone. So I decided to sign up with a new number with TalkTalk but the checker no longer works, so I can't test the latest result with phone.

With the old number only difference I saw 65.1Mbps for Impacted line vs 64.8Mbps with the address checker. The rest is the same 80Mbps max. Briefly the checker was slightly lower and then returned back to how it was before.

But my estimates definitely don't reflect that of a 500m line. Even if there were slight variations in the way the copper travels from the cabinet to the property, I can only see it add a further 50-60 meters extra from the 0.2 mile that google map shows. On google maps it gives you the option to drag the yellow character and drop to exactly the location of the street cabinet.

Here's a chart. https://www.increasebroadbandspeed.co.uk/2013/chart-...

Broadband users within about 300 m of the street cabinet can expect to achieve about the maximum possible downlink connection speed (currently 80 Mbps). Speeds fall to about 60 Mbps when 500 m away from the street cabinet

Now maybe this is a little bit pessimistic estimation. But I believe with 500 m 80Mbps is not attainable. You need to be at least 450 meter to cabinet for max speeds.

There is no way getting 80Mbps if the street cabinet are 500m away.

My parents line: SNR 3dB sync at 58Mbps, SNR 4dB sync at 56Mbps, SNR 5dB at 54Mbps and 6dB sync at 52Mbps with line attenuation of 16.1dB (report by PN Hub One) as BTw Checker say their handback would be 50Mbps for their line due to crosstalk. Estimated between 79.2 and 55.8.

Downstream Observed Speed (Mbps) 53.83 (higher)
Downstream Observed Speed (Mbps) 51.83 (lower)
Upstream Observed Speed (Mbps) 13.94 (higher)
Upstream Observed Speed (Mbps) 11.94 (lower)

PN GEA Test say their line length estimated are: 453.7m

Your many assumptions are wrong.

Your attenuation suggests your line is considerably longer than 300m.

The engineer connected an expensive, specialised piece of testing equipment and that suggested a line length of 500m.
That roughly matches your attenuation. From the attenuation alone I would have guessed 450-500m, perhaps even a little longer.
80Mb on a 500m long line is very achievable.

You also had a EO line which historically were made up of thicker gauge copper which has less signal lots over longer distances.
Your also 1 is the 1st on your cabinet so there will be very little crosstalk present.

That chart you have posted is VERY pessimistic.

Have a look at this chart showing a wide range of observed speeds at different distances.

https://www.researchgate.net/profile/George_Ginis/pu...

Crosstalk is very very variable. The is no fixed rate over distance, every line is different.

I trust the engineers equipment and the reported attenuation from your modem much more than your guess from Google Maps and your incorrect assumptions on what sync speed is possible.

j0hn83

Do u ever work for Openreach? It's seem that u know everything about Fibre, DSL, DLM, Openreach network etc

@John: I have done several measurements and I know my area very well, I live here all my life. I've went to visit the cabinet and I can see it from the window of my flat.

The Bishopsgate Cabinet 20 is located in Old Montague Street and it travels in a straight line all the way to the end of Wentworth Street would be 482 meters to Petticoat Lane Market. But I live in Commercial Street which situates between Wentworth Street which is around 200 meters closer.

So unless the copper cable is zig-zagging artificially underground there is no way 300 meters worth of straight-lined copper can turn into 500 meters. And that I include the height of the residential building approximately another 20 meters which should total around 320 meters. The measurement is actually less than 300 meters if height wasn't included.

Your chart shows that with 500 meter line (non-vectored) does not achieve 80Mbps. It needs to be vectored. But with vectored line it shows close to 120Mbps, how is it that the router shows Maximum Rate [Kbps] 80543 if vectoring was really applied to the line?

If you think my line is 500 meter long, why is the minimum estimate 68.2Mbps for me and for the parents of adslmax being 55.8Mbps. This is a big difference, isn't it?

Also if my EO line was historically made out of thicker gauge of copper then I shouldn't have been having a ton of internet disconnections before on ADSL that required 9dB SNR cap.

It would be interesting to know from other people's experiences what speeds they get on average with a 500 meter cabinet.

That chart also shows 650 meter line vectored can achieve 80Mbps and with 800 meter line vectored 60Mbps, these seem too good to be true. I have yet to see anyone on the internet claiming to achieve 80Mbps with a 650 meter cabinet distance or 60Mbps with 800 meter cabinet distance. Yet at the same time I have seen BT Wholesale post code checkers show estimates sometimes with only 40Mbps max that aren't reflective of this chart.

In reply to a post by BLaZiNgSPEED:

Also if my EO line was historically made out of thicker gauge of copper then I shouldn't have been having a ton of internet disconnections before on ADSL that required 9dB SNR cap

Any broadband problem on your "EO" line between the exchange and the cabinet is eliminated on FTTC. It is filtered out before being merged with the VDSL2 at the DSLAM.

In reply to a post by BLaZiNgSPEED:

I know my area very well, I live here all my life.

You may have lived in the area all your life but do you know the telephone cabling infrastructure? I worked out of Bishopsgate and Wapping telephone exchanges for several years back in the day and I wouldn't be surprised if a few hundred metres of copper was added to your line length.

Can't say I miss working that patch as I always had to keep my eye's in the back of my head and had a few near scraps that could have cost me my life. Although I did enjoy a curry in Brick Lane and going to Truman's brewery.

Edited by deleted (Sun 26-Apr-20 10:25:26)

In reply to a post by gary333:

You need to get over this 30 day nonsense. It's happening, you cannot stop it, and no, no consumer based company will ever tell you about as it's not a fault.

You are living in cuckoo land with this obsession of no disconnects, and I can demonstrate this with numbers. Having the router restart and taking say 1:45 seconds per month to comeback to life is the equivalent of 99.99% uptime! As far as I am concerned this is a superb up time for something you are only paying £20 per month.

Spending 1 hour on to customer services costs them around £15 (average outsourcing cost for this industry for UK staff) so I am pretty sure they don't want people wasting their time with nonsense of this nature and thus if it was an issue to more than a sample of 1 in a million they would document it.

Well actually TalkTalk Support forum have offered me a new router replacement which I have received 2 days ago! It is a Huawei Router DG8041W-2.T5. The packaging and router design looks 100% identical to the Sagemcom FAST 5364, I have no idea about the specification differences between these 2 routers.

I haven't connected it yet, but will do within a few days. I'm on 25 days connection up-time and will see if the old router reboots again for the 3rd time after 5 days.

But anyway, it will be interesting to find out if the new router will remain stable after 30 days, hopefully it will stop the issue from occurring.

I'm pretty sure if this was a nonsensical issue then TalkTalk would not have voluntarily offered to send me the new router replacement which they have done under their own good will!