Fibre 2000m from Exchange - 80/20 or 40/10

Hi,

We've just gone live with Fibre in my village. Unfortunately I'm 2096m from the exchange and therefore only expecting speeds of around 10-15Mbps, even though my line can support 36ish Mbps. First off, is this assumption correct?

Secondly, would 80/20 increase my possible download speeds in this situation, or would I just be better off going with the cheaper 40/10

Any help would be great! Many thanks.

Edited by deleted (Thu 02-Jun-16 19:17:42)

Exchange distance is irrelevant. Distance to the street cabinet is key.

40/10 is your max according to that estimate.

Edited by deleted (Thu 02-Jun-16 19:21:54)

Sorry, yes, I meant 2096m from the cabinet.

Thanks for the help, appreciate it.

Why are you only expecting 10-15mbps if your line can support 36?

What results do you get for telephone number/address from this checker http://www.dslchecker.bt.com ?

In reply to a post by lee111s:

Why are you only expecting 10-15mbps if your line can support 36?

What results do you get for telephone number/address from this checker http://www.dslchecker.bt.com ?

No way would you get 36Mbps at 2km from the cabinet. Agreed, we need to see the results from the checker.

Why are you only expecting 10-15mbps if your line can support 36?

err ... "sorry, yes, I meant 2096m from the cabinet. "

Thanks for the responses all, appreciate it.

I've run the line checker:

Downstream Line Rate Upstream Line Rate
High Low High Low

VDSL RANGE A (Clean) 47.6 ; 35 | 10 ; 6.9
VDSL RANGE B (Impacted) 35 ; 18 | 8.4 ; 4.4

Is 18Mbps referring to the lowest I could expect before the distance from the cabinet takes effect? Or is that suggesting the overall lowest I should expect, regardless of distance, is 18Mbps? I would imagine the former.

Edited by deleted (Thu 02-Jun-16 22:19:18)

I would expect about 30 mbps by those results but you could be lucky and get the 40+ estimate.

Edited by epyon (Thu 02-Jun-16 22:20:05)

Even at 2096m from the cabinet?

In reply to a post by soziblewuup:

Even at 2096m from the cabinet?

You aren't.

The line above the estimate table gives you a cabinet number. If you post that number and the exchange, there are people here that will find it on google maps and give us the link.

Ah, perhaps I've mixed up the exchange and cabinet after all. I was using CodeLook earlier.

I'm served by the Woodseaves Exchange Cabinet 2

I am about 2500m from the cabinet and get 5.9Mb, a neighbour is closer, so at about 2000m from the cabinet and they are on 12Mb.

So your expectations are realistic, you can be happy if you get anything faster. It will certainly be faster than the ADSL connection you had before. There is no point in signing up for 80/20 at all.

HTTPS://goo.gl/maps/PWNHjZZvwF42

Great stuff Ribble! I know where that is, and it's no where near 2000m. More like half that distance. Expecting something in the 20s now if I'm lucky!

Thanks for your help.

I am at over 1300m from the PCP and I get around 24Mbps download just to give you an idea.

Remember it is line length which is not direct straight line and not always te road route. Your line may go off in various directions before getting to you. As an example friends are under 1km from their exchange - the line in 9km.

Where did the original 2096 come from? was it from a BT Tech?

Thanks for the response guys. It's starting to look like I'm just going to have to wait and see what happens. There seems to be no hard and fast rule. I've ordered 40/10, with my activation date being 20 days away. Excited. Mind, hoping I lean towards the mid 20's rather than the low teens.

In reply to a post by soziblewuup:

Is 18Mbps referring to the lowest I could expect before the distance from the cabinet takes effect? Or is that suggesting the overall lowest I should expect, regardless of distance, is 18Mbps? I would imagine the former.

The two ranges already take account of your distance (strictly, other factors influence speeds too, such as thickness and material).

You can expect your speed to be somewhere in the A range if your line is in decent shape - no dodgy joints, no extra branches of cable left attached, no insulation worn away etc.

You can expect the B range to apply if there are faults in the wiring.

ISPs will decide which range to use based on your installation choice; go for an engineer installation, and they should quote range A speeds ... and jump into action if reality falls well below these higher figures. Go for a self-install, and they'll use range B instead.

Whichever range is chosen, your actual speed is more likely to start out at the higher end of the range (or above it) if you are an early subscriber on the cabinet. It is likely to drop as time goes by, and more subscribers get added.

In reply to a post by WWWombat:

ISPs will decide which range to use based on your installation choice; go for an engineer installation, and they should quote range A speeds ... and jump into action if reality falls well below these higher figures. Go for a self-install, and they'll use range B instead.

That doesn't really ring true because if your speed is outside the A range after a self-install and you report it, they will send an engineer.

Guess it depends on the ISP. There have been plenty of posts on self-install cases where range B is both quoted pre-order and used when reporting faults.

I can't speak from experience though. All my installations came with an engineer, and all my faults have been dealt with as verified by a test, not just with "low speed" as the only symptom.

IMO, it would be a good move for BT to do as you say.

Just out of interest, which ISP have you chosen?

It will not be low teens.

You will get probably 25Mbps with a chance of even getting up to 40.

I am sure it'll be great. Just ensure to use the master phone socket and ensure there are filters on all sockets if you opt for a self install.

Too right Batboy. After all, I'm over 7km from the exchange but around 250mtrs (if BT measuring stick can be believed) from the fibre cabinet. I get around 73-74mb/s

Hi RobertoS, I chose Zen in the end.

I've had fibre for 8 days now, and the results haven't been great. When I was first activated (8 days ago) I was achieving speeds of between 16 and 22Mb/s (constantly fluctuating). However, as time has gone on I'm now down to 10-12Mb/s. Everyday I seem to lose speed. When I was at 20Mb/s the connection didn't appear to drop at all. I also haven't rebooted the router in the 8 day period. My upload speed remains constant at roughly 8Mb/s.

Looking at my BT line stats:

VDSL Range A (Clean)

Downstream High = 44.7
Downstream Low = 32.4
Upstream High = 9
Upstream Low = 6.2

VDSL Range B (Impacted)

Downstream High = 35
Downstream Low = 18
Upstream High = 7.7
Upstream Low = 4.3

Surely this can't be right? Roughly 800m from the Cabinet on a 40/10 package and I'm only receiving 12Mb/s max (currently) downstream.

Any help, would be greatly appreciated!

Do you have line stats from your router?

I'm using the Zen standard ZyXEL router in bridge mode with a Netgear R7000 Nighthawk

From the ZyXEL:

VDSL Training Status: Showtime
Mode: VDSL2 Annex B
VDSL Profile: Profile 17a
Traffic Type: PTM Mode
Link Uptime: 5 days: 3 hours: 51 minutes
============================================================================
VDSL Port Details Upstream Downstream
Line Rate: 9.456 Mbps 13.030 Mbps
Actual Net Data Rate: 9.428 Mbps 12.980 Mbps
Trellis Coding: ON ON
SNR Margin: 5.9 dB 13.4 dB
Actual Delay: 0 ms 8 ms
Transmit Power: 6.1 dBm 11.5 dBm
Receive Power: -9.1 dBm -20.8 dBm
Actual INP: 0.0 symbols 3.0 symbols
Total Attenuation: 15.2 dB 32.4 dB
Attainable Net Data Rate: 9.432 Mbps 25.776 Mbps
============================================================================
VDSL Band Status U0 U1 U2 U3 D1 D2 D3
Line Attenuation(dB): 9.3 42.2 61.4 N/A 19.0 53.8 77.8
Signal Attenuation(dB): 9.3 41.3 63.8 N/A 30.3 53.1 N/A
SNR Margin(dB): 6.0 5.9 5.9 N/A 13.4 13.2 N/A
Transmit Power(dBm): 0.1 - 1.1 3.5 N/A 9.4 7.4 N/A
============================================================================

VDSL Counters

Downstream Upstream
Since Link time = 51 min 49 sec
FEC: 15135432 11910
CRC: 8 1042
ES: 4 830
SES: 0 47
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Retr: 0
Latest 15 minutes time = 14 min 2 sec
FEC: 17767 3
CRC: 0 1
ES: 0 1
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Retr: 0
Previous 15 minutes time = 15 min 0 sec
FEC: 17758 1
CRC: 0 0
ES: 0 0
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Retr: N/A
Latest 1 day time = 20 hours 59 min 2 sec
FEC: 1583668 6715
CRC: 0 103
ES: 0 84
SES: 0 11
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Retr: 0
Previous 1 day time = 24 hours 0 sec
FEC: 3987865 1101
CRC: 4 185
ES: 2 139
SES: 0 10
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Retr: 0
Total time = 1 days 20 hours 59 min 2 sec
FEC: 15135432 11910
CRC: 8 1042
ES: 85406 1150
SES: 34 144
UAS: 164 132
LOS: 3 0
LOF: 15 0
LOM: 0 0
Retr: 3
============================================================================

That doesn't look right, but there are a couple of things to do. The first it to find the line stats on your router and post those (things like sync speed, noise margin and so on). Also, are those WiFi or direct-wired speeds? In some areas with lots of contention WiFi can be very poor.

The other is a detailed look at your domestic wiring setup. VDSL is very sensitive to how that is configured, and it's easy to lose a huge amount of your speed potential. Unless you have only a master socket in your house with no extensions (used or not), then I always recommend base-lining your performance by getting stats after eliminating all the effects of the extension wiring.

Assuming that you have an NTE5 master, this is easliy done, but if it's not the normal location of your router, you will have to (temporarily) relocate it near the master socket and, if necessary, use a mains extension lead to provide a temporary power supply.

To test with all the extensions disconnected, simply unscrew the faceplate from the NTE5 master and pull it away. That action should disconnect the extensions and, underneath, there will be a test phone socket. That connects directly to the incoming line. Connect the router to that (using a microfilter if necessary), wait for it to sync and then copy the line stats. Also perform a few speed tests (using a cabled connection). That should be the best performance that your line is capable of (at least without fixing any line faults) and should act as the performance baseline. If you compare that with the stats with your normal set-up it will tell us what degradation is being caused by the domestic wiring.

If you don't have and NTE5 master then you'd have to post back details about your configuration.

If it turns out that domestic wiring is causing issues, there are a number of ways of resolving those which range from some very simple things to the fitting of a filtered faceplate and re-jigging the wiring.

Those stats look odd. Upstream attenuation is reasonably low whilst downstream is higher than I'd expect. See my other post on baseline testing from the master socket (downstream performance is often impacted by wiring issues much more than upstream),

On the "good" news front, I'd expect a line capable of syncing at around 10mbps to be much better downstream than that.

Appreciate the response Euler, I'll get testing right away.

I think you may have found the route of the issue!

Wireless Speed Test: D:12 U:8.7
Wired Speed Test: D:12 U:8.7
Test Socket - Wired: D: 33.5 U:9.3

Stats from test socket:
============================================================================
VDSL Training Status: Showtime
Mode: VDSL2 Annex B
VDSL Profile: Profile 17a
Traffic Type: PTM Mode
Link Uptime: 0 day: 0 hour: 3 minutes
============================================================================
VDSL Port Details Upstream Downstream
Line Rate: 10.057 Mbps 36.155 Mbps
Actual Net Data Rate: 9.999 Mbps 36.090 Mbps
Trellis Coding: ON ON
SNR Margin: 7.3 dB 6.3 dB
Actual Delay: 0 ms 8 ms
Transmit Power: 7.1 dBm 11.5 dBm
Receive Power: -8.3 dBm -16.7 dBm
Actual INP: 0.0 symbols 3.0 symbols
Total Attenuation: 15.5 dB 28.3 dB
Attainable Net Data Rate: 11.224 Mbps 43.447 Mbps
============================================================================
VDSL Band Status U0 U1 U2 U3 D1 D2 D3
Line Attenuation(dB): 8.9 40.0 58.9 N/A 17.3 46.3 70.7
Signal Attenuation(dB): 8.9 39.7 58.7 N/A 26.0 46.0 N/A
SNR Margin(dB): 7.4 7.3 7.2 N/A 6.3 6.1 N/A
Transmit Power(dBm): 0.5 - 3.6 5.6 N/A 9.2 7.8 N/A
============================================================================

VDSL Counters

Downstream Upstream
Since Link time = 3 min 39 sec
FEC: 89150 0
CRC: 0 0
ES: 0 0
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Retr: 0
Latest 15 minutes time = 12 min 35 sec
FEC: 89150 0
CRC: 0 0
ES: 0 0
SES: 0 0
UAS: 536 536
LOS: 0 0
LOF: 0 0
LOM: 0 0
Retr: 0
Previous 15 minutes time = 15 min 0 sec
FEC: 34472 1658
CRC: 2406 136
ES: 23 19
SES: 16 4
UAS: 50 34
LOS: 0 0
LOF: 9 0
LOM: 13 0
Retr: N/A
Latest 1 day time = 21 hours 42 min 35 sec
FEC: 1725906 8374
CRC: 2406 241
ES: 23 105
SES: 16 15
UAS: 586 570
LOS: 0 0
LOF: 9 0
LOM: 13 0
Retr: 1
Previous 1 day time = 24 hours 0 sec
FEC: 3987865 1101
CRC: 4 185
ES: 2 139
SES: 0 10
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Retr: 0
Total time = 1 days 21 hours 42 min 35 sec
FEC: 15277670 13569
CRC: 2414 1180
ES: 85429 1171
SES: 50 148
UAS: 750 702
LOS: 3 0
LOF: 24 0
LOM: 13 0
Retr: 4
============================================================================

I'm assuming this is being caused by extensions?

I also believe I already have a filtered faceplate (I think).

Have a look at the Attenuation by Band as posted by the OP:

============================================================================
VDSL Band Status U0 U1 U2 U3 D1 D2 D3
Line Attenuation(dB): 9.3 42.2 61.4 N/A 19.0 53.8 77.8
Signal Attenuation(dB): 9.3 41.3 63.8 N/A 30.3 53.1 N/A
SNR Margin(dB): 6.0 5.9 5.9 N/A 13.4 13.2 N/A
Transmit Power(dBm): 0.1 - 1.1 3.5 N/A 9.4 7.4 N/A
============================================================================

The line attenuations both Up and Down do look reasonable for an 800m line.

However looking at signal attenuation, Up is reasonable and is broadly in line with Line (as expected) but D1 is significantly different, 30.3 compared to 19.0. D2 is again reasonable. That is showing the result of the problem - what is causing it? Possible a dodgy filter? Maybe a tuned/resonant tap or unterminated wire? Or a modem from end problem at the lower frequencies.

Do you agree?

A DSL stats plot showing the the D1 tones with Bit Loading and SNR might shed a little light on what is affected - the whole band or just a block.

Those show a marked improvement, however there is still a mismatch of line and signal attenuations in the D1 band.

For now, is there any reason why I couldn't leave my filter in the test socket? My digital phones and internet are currently fully operational using the test socket, and I use PowerLine adapters for everything else (Smart TV, Sky, PC Ethernet, Gaming etc.). I've just had a quick scoot around the house and not one of the 4 or so extensions are in use. Or I suppose I could just disconnect the wires from the rear of the Faceplate and reattach?

However, if I ever wanted to start using the extensions again, for whatever reason, I'd have to get this sorted. It seems as though it could be any one of a number of things.

Either would do it.

I would disconnect the extension wiring and try in the faceplate to prove it is wiring related.

That's the sort of effect you get with a "bridged tap". That is length of cable where the signal is largely reflected off the end of the transmission line cancelling out much of its strength. I have my router on the end of about 8 metres of extension and it normally syncs at about 55mbps (I'm 700m from the cabinet). If I attach the router to the master socket then I lose 40% of that sync speed. If I detach the extension when connected to the master I get it all back.

My general rule is that there should be no unfiltered extensions other than the one your router sits on. All other extensions should be filtered at source.

Glad it worked. There's no problem with continuing to use the test socket. If you want to reinstate the extensions, then if they are being used only for voice purposes, then simply buy an Interstitial plate. See link. It fits between the faceplate and the master and filters all the extensions. There isn't even any rewiring to do. However, your router will only work at the master socket.

https://www.amazon.co.uk/GENUINE-BT-Openreach-Inters...

If you want to put your router on an extension, then it would require a bit of work as that extension (and that extension alone) would have to be unfiltered.

I didn't see that on my last post. Do you mean your current faceplate is a filtered one, or that you have one available? If the former, and the extensions are connected to the filtered output of the faceplate, then it ought to behave as if connected to the test socket so that would be puzzling. A photo showing the faceplate and how the extension wiring is attached would be good.

I know, I have also seen similar with a bad filter hence the list.. It is still there with the new stats which suggests it could be outside the property.

I'm not actually sure if it's filtered. Here's an image:
https://s32.postimg.org/uf1pvtl3p/IMG_0835.jpg

This is an image of how the wiring looks. There appears to be two white wires going into the same place, as well as two blue wires on the opposite side:
https://s32.postimg.org/caypb6nf9/IMG_0832.jpg

I was also previously using a broadband filter with surge protection. This didn't appear to work in the test socket, so I swapped it out for an older bog standard filter.
https://s32.postimg.org/ei2xz3sph/IMG_0836.jpg

There certainly could be something outside the house, which I suppose makes this an "impacted" line. At least the sync speed is rather closer to what might be expected for a line of that length.

It certainly does seem anomalous. Normally signal attenuation isn't so much different to line attenuation, but I'm no expert on that or why it might be.

That is NOT a filtered faceplate. You would normally have TWO outlet s - on RJ for DSL and one BT plug for phones. The reason for the loss of speed was the extensions being connected to the existing and they were not filtered.

You can buy a filtered faceplate on line.

OK - that's not a filtered faceplate. Either buy an Interstitial plate (see my other post) and insert it between the faceplate and the master or buy a filtered faceplate to fit the NTE5. The advantage of the former is there's now wiring work to do, but it's a bit ugly as it pushes the faceplate out a couple of cm. The advantage of the filtered faceplate it's a not neater and gives you both a VDSL and phone outlet but you have to connect your extensions to the new plate. Easy enough, but if you do it, get one of the good spring-loaded Krone tools, not one of those nasty all plastic jobbies. They cost little more and are a heap better.

Thanks you very much guys! I'm thinking about just buying the Interstitial Faceplate. the Master Socket is actually hidden behind a desk so aesthetics isn't necessarily an issue.

So all I have to do is fit the plate in between the current plate and the socket? Do I then just use my standard filter for both phone and VDSL, plugged into the outer plate, as before?

I'm assuming I have to replace the current middle plate with the BT Openreach logo on the top?Or is that just left and the two openreach plates will sit in parallel?

Edited by deleted (Wed 29-Jun-16 12:49:39)

Ok, I just had a little look online and it seems as though it simply slots on top. I've ordered the one linked.

Great stuff, thanks for all your help guys. Really appreciate it!

Yes, it's easy enough as there's a slot cut in the bottom and you slide the extension cable through it. Instructions should be with the plate. No wiring required.

Given that VDSL is so sensitive to these wiring issues really all the ISPs ought to include a "best practice" booklet or website. Heaven knows how many people are running way below what is possible. It looks like you were losing about 64% of potential download throughput.That's a huge hit.

Glad to be of help.

nb. I forgot to say, the plate has it's own RJ11 output socket. Plug the modem lead into that - not need for the microfilter (and if you do use it, it won't work as the voice socket will be filtered).

Great stuff! I agree, no where on the Zen website / correspondence I received was there any explanation surrounding necessary modifications. This seems as though it's a pretty vital step in achieving the speeds customers are expecting.

Thanks again!