Print

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

[npr]

Can I please ask for advice on my new Plusnet (40/10) fibre BB connection.

Stats:
Max:   Upstream rate = 1420 Kbps, Downstream rate = 27980 Kbps
Path:   0, Upstream rate = 1368 Kbps, Downstream rate = 23360 Kbps

Link Power State:   L0
Mode:         VDSL2 Annex B
VDSL2 Profile:      Profile 17a
TPS-TC:         PTM Mode
Trellis:      U:ON /D:ON
Line Status:      No Defect
Training Status:   Showtime
      Down      Up
SNR (dB):    6.6       6.1
Attn(dB):    0.0       0.0
Pwr(dBm):    11.2       6.6
         VDSL2 framing
         Path 0
B:      63      42
M:      1      1
T:      63      22
R:      16      0
S:      0.0872      0.9885
L:      7343      348
D:      371      1
I:      80      43
N:      80      43

I'm about 700 metre from the cabinet by road, don't know the true cable length.

I suspect the download speed to be reasonable but I'm rather disappointed with the upload speed.  Any thought as to whether it's worth pursuing this with PN.

[burakkucat]

I wonder if there has been a mix-up, somewhere and you have been provisioned on the old 40/2 profile? 

I would certainly raise it with PN, if it was my service.

[Bald_Eagle1]

I wonder if there has been a mix-up, somewhere and you have been provisioned on the old 40/2 profile? 

I would certainly raise it with PN, if it was my service.

That was my first thought too as Plusnet seem to have a bit of an issue where some new users have recently been provisioned on the wrong (lower speed) service. i.e. 80/20 customers ending up on 40/10, so thinking maybe this 40/10 user has ended up on Plusnet's 'obsolete' 40/2 service.

However, these seem to suggest that the connection itself is physically currently unable to allow higher US speeds:-
Max:   Upstream rate = 1420 Kbps, Downstream rate = 27980 Kbps
Path:   0, Upstream rate = 1368 Kbps, Downstream rate = 23360 Kbps

SNR (dB):    6.6       6.1



This is what my connection looked like when I was on Plusnet's 40/2 service
US Sync speed 1997 Kbps from an attainable 6071 Kbps & US SNRM at 14dB:-

Code: [Select]


# xdslcmd info --pbParams
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason: 2
Max:    Upstream rate = 6071 Kbps, Downstream rate = 24376 Kbps
Path:   0, Upstream rate = 1997 Kbps, Downstream rate = 20316 Kbps

Discovery Phase (Initial) Band Plan
US: (0,95) (696,1183)
DS: (32,687) (1192,1627)
Medley Phase (Final) Band Plan
US: (0,95) (696,1183)
DS: (32,687) (1192,1627)
       VDSL Port Details       Upstream        Downstream
Attainable Net Data Rate:       6071 kbps         24376 kbps
Actual Aggregate Tx Power:        1.1 dBm           9.6 dBm
============================================================================
  VDSL Band Status        U0      U1      U2      U3      D1      D2      D3
  Line Attenuation(dB):  8.5     52.6     N/A     N/A    21.5    65.5     N/A
Signal Attenuation(dB):  9.3     51.1     N/A     N/A    21.5    65.5     N/A
        SNR Margin(dB): 14.1     14.0     N/A     N/A     7.0     7.0     N/A
         TX Power(dBm): -9.3      0.7     N/A     N/A     9.2    -0.7     N/A




@ npr,

Could you post your xdslcmd info --pbParams data for comparison against mine?
It may just be that your connection's attenuation is too high for its distance (maybe a fault or aluminium cabling?)
Even your DS speeds seem a shade low for 700m or so.

I'm 1000m or so from the cabinet & these are my currently attainable rates:-

Max:   Upstream rate = 5620 Kbps, Downstream rate = 30220 Kbps

[npr]

Thanks, here's my pbParams stats:

Code: [Select]

# xdslcmd info --pbParams
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason: 0
Max: Upstream rate = 1206 Kbps, Downstream rate = 25852 Kbps
Path: 0, Upstream rate = 1220 Kbps, Downstream rate = 22405 Kbps

Discovery Phase (Initial) Band Plan
US: (0,95) (880,1195) (1984,2771)
DS: (32,859) (1216,1959) (2792,4083)
Medley Phase (Final) Band Plan
US: (0,95) (880,1195)
DS: (32,859)
       VDSL Port Details       Upstream        Downstream
Attainable Net Data Rate:       1206 kbps         25852 kbps
Actual Aggregate Tx Power:        6.3 dBm          11.2 dBm
============================================================================
  VDSL Band Status        U0      U1      U2      U3      D1      D2      D3
  Line Attenuation(dB): 7.1 63.5 63.4   N/A 26.2 0.1 0.1
Signal Attenuation(dB): 7.1 60.6   N/A   N/A 26.2   N/A   N/A
        SNR Margin(dB): 5.5 6.0   N/A   N/A 5.6   N/A   N/A
         TX Power(dBm): 2.0 4.5   N/A   N/A 11.2   N/A   N/A


I raised a ticket with PN a few days ago, rather disappointed with their response -- it was mostly meaningless techno babble.

"Unfortunately due to the length of the line there has been a banded profile assigned to the line to ensure stability and persistent downstream throughput speeds. Unfortunately the profile does place a large proportion on the downstream therefore the upload can be placed lower than your necessary estimate."

Looks to me as though they are blaming a IP profile when in fact it's the sync speed which determines the max upload speed.

Note:
When I ordered fibre I was told the predicted speed was 18 mbps down and 3.8 mbps up.
I can live with this, it's a lot better than the 3.5mbps down and 0.85 mbps up I was getting on adsl2.
My main problem is that I don't know if this speed, especially the upload speed, is reasonable. That and the fact I don't find PN's techno babble replies reasonable,  so if it's worth pestering PN a bit more then I'm up for that

[Bald_Eagle1]

Firstly, it doesn't look like a banded profile to me.

Your attenuation values look really high for a 700m line length from the cabinet, suggesting your line is actually longer than my 1000m line.

Users are not always connected to the closest cabinet, so that could be the reason or the cable may be aluminium or has a 'fault', possibly high resistance (HR) somewhere between you and the cabinet.

Also, your own internal wiring COULD be causing some issues.
If you have an ADSL microfilter, you could try removing the VDSL faceplate & plugging the filter directly into the test socket. That should disable any internal wiring & give you a 'clean' connection for testing purposes.

If you are a Windows user, you could try using the new logging/graphing programs that were released today & post some graphs/stats from your connection for us to have a look at.
These are freely available here:-

Freewarefiles.com

[npr]

Thanks, I'm already using your previous graphing scripts. Will this do?



Yes I am on the closest cabinet, I know this because I had to direct the fibre installer to the cabinet The cable could go all round the houses though.
Always had good results from ADSL2 considering the distance from the exchange, so doubt there's anything wrong with my home wiring but the same should apply to the line from the cabinet.   
I'll try from the test socket tomorrow and report back.

[Bald_Eagle1]

Thanks, I'm already using your previous graphing scripts. Will this do?

I'll try from the test socket tomorrow and report back.

The graph that might tell us more about the physical quality of the wiring (internal and external) is the Hlog graph.

If it has any pronounced valleys or peaks, it would suggest a physical wiring problem such as a bridged tap etc.

Maybe graphs plotted before using the test socket & during using the test socket would help to rule in/out your internal wiring.

As you are connected to an ECI DSLAM :-
Discovery Phase (Initial) Band Plan
US: (0,95) (880,1195) (1984,2771)
DS: (32,859) (1216,1959) (2792,4083)

you may just have some graphs knocking about that show Upstream data for QLN, Hlog & SNR.

Huawei DSLAMS NEVER show that data & ECI DSLAMS sometimes show it.


 

[npr]

I'll get the graphs from the test socket tomorrow.

[npr]

Graphs from the test socket.
There's a noticeable difference in the QLN graph, don't know the significance.

[asbokid]

Graphs from the test socket.
There's a noticeable difference in the QLN graph, don't know the significance.

The ECI DSLAM has now given you the QLN tonemap (the background noise) for the upstream bands.

How far are you from the cabinet?  EDIT: Sorry, just seen.. 700 metres.  It's a poor service for that distance.

cheers, a

[npr]

I've had a bit of a strop with PN, mainly regarding the meaningless rubbish they put on the ticket.
Result -- they are now sending out a BT engineer
I'll report back on how things go.

[Bald_Eagle1]

I've had a bit of a strop with PN, mainly regarding the meaningless rubbish they put on the ticket.
Result -- they are now sending out a BT engineer
I'll report back on how things go.

When the engineer arrives, I would ask him to determine from his JDSU or Exfo tester, the electrical line length.

If it's a lot more than the 700m, it COULD suggest a physical line problem.
Both devices can also be used to run a TDR test that should identify the distance of the problem from wherever it is tested i.e. from your master socket, or from the DSLAM in the cabinet.

If you are lucky, the engineer may still have a 'Hawk' or 'Mole' or other similar specific TDR tester that would report things in more detail.

The engineer should also be able to show you the actual line route & cable guage/type & be able to approximate its length from BT's Network Records that he should be able to access from his laptop.

Bacon butties, chocolate hobnobs & copious amounts of tea/coffee often seem to pave the way to an engineer being particularly helpful.

As mentioned by asbokid, the ECI DSLAM has decided it would send the US QLN data to your modem.
It probably also sent the US Hlog & SNR data too (as per the attached example from a connection with a potential bridged tap problem).

The new programs (& later versions of the scripts) can capture US data & plot it in green to distinguish it from DS data.

If you don't wish to use the new programs, you could zip the Plink log from your data obtained whilst connected via the test socket & post it here (as long as the zip file is less than 200K in size).

I'll then plot it & post the montage here as an attachment for you.

Have you been logging ongoing stats 24/7?
That data would give us (& the engineer) a general feel for what is going on with your connection over a period of time.

[npr]

Well, the engineer has been and he put in a tremendous amount of effort, even changed the pair of wires between cabinet and pole, but unfortunately the connection is pretty much the same.

He says the cable distance is 900 metre.
He also thinks it's aluminium wire although he seemed a little unsure about this.
The estate was built in the early 1980s. I thought I'd heard they stopped using ally before then -- can anyone confirm this?
I can only assume it is aluminium wire and that's the cause of the problem

Thanks for all the help
npr


Latest stats and graphs

[asbokid]

The problem looks like widespread Radio Frequency Ingress (RFI) across the VDSL2 spectrum.  The interference looks terrible. Is your line pole-strung for long distances of that 900 metres, where it would serve as an unwanted radio antenna?  Do you live on elevated ground?

The price of copper soared in the early 1970s (iirc, which i don't!) There were copper mine crises, price ramping in the City, etc and the GPO was forced to look for alternatives, i.e. aluminium.  Whether the price had stabilised and stocks replenished by the 1980s, don't know.

cheers, a

[npr]

The cables are under ground to the pole, the pole is only outside the house next door.

The modem is next to the TV, just for completeness I'll try unplugging the TV and rebooting the modem -- see if that gives a better sync.
Can't thing of any other RF source. Had no such problems with ADSL so hadn't considered this.