In reply to a post by WWWombat:

I don't know enough about the Hlog graph to be able to read or interpret it.

Loosely, Hlog depicts attenuation over frequency.

The ideal graph is a gently sloping (almost horizontal) & smooth graph with no peaks or troughs.

A reasonably wide & deep 'v'-shaped valley tends to suggest a bridged tap/High Resistance issue & a peak tends to suggest some sort of high capacitance.

Edited by deleted (Tue 05-Feb-13 19:59:09)

In reply to a post by Bald_Eagle1:

The ideal graph is a gently sloping (almost horizontal) & smooth graph with no peaks or troughs.

I'm not sure about horizontal - the ones I've seen seem to be a smooth descending line, that gently curves to be less steep as it descends.

The OP's Hlog graph, to me, looks fine. With your interpretation, it suggests there are no obvious line problems to note, right?

This is what I meant by ideal/almost horizontal:-

Almost flat Hlog Graph


& here is the pbParams data from that connection - ONLY 13.8dB ATTENUATION in the D3 band!!!
Apparently around only 60m from the cabinet.

Text

1 23 45 67 89 1011 1213 1415 1617 1819

Retrain Reason: 0 Max:    Upstream rate = 41318 Kbps, Downstream rate = 129700 KbpsPath:   0, Upstream rate = 10000 Kbps, Downstream rate = 39998 Kbps  Discovery Phase (Initial) Band Plan US: (0,95) (868,1207) (1972,2783) DS: (32,859) (1216,1963) (2792,3939) Medley Phase (Final) Band PlanUS: (0,95) (868,1207) (1972,2783) DS: (32,859) (1216,1963) (2792,3939)        VDSL Port Details       Upstream        Downstream Attainable Net Data Rate:      41318 kbps        129700 kbpsActual Aggregate Tx Power:    -   8.6 dBm          13.3 dBm ============================================================================  VDSL Band Status        U0      U1      U2      U3      D1      D2      D3   Line Attenuation(dB):  2.2     7.8     10.0     N/A    4.4     9.2     13.8   Signal Attenuation(dB):  0.0      N/A    7.2      N/A    4.4     9.2     13.8           SNR Margin(dB):  30.7     N/A    30.8     N/A    31.4    31.2    31.4            TX Power(dBm): -25.1   -128.0  -10.2     N/A    10.5    7.7     6.2

I agree. The OP's Hlog graph looks just fine.


Here's one before the bridged tap was removed:-

Bridged Tap

& the same connection after it was removed :-

Bridged Tap Removed

Edited by deleted (Tue 05-Feb-13 22:00:35)

your analysis is similiar to what I see on my line.

my D1 bitloading is similiar to lines with higher attenuation than mine in other words its underperforming. Indeed my D1 has very noticeable lower bitloading than when I had a 90meg attainable sync.
my D2 also took a noticeable hit.
my D3 didnt appear to make much of a hit and is almost the same as when I had the higher sync.

The upstream again similiar however with something more interesting.

U1 does not have increased QLN unlike D1 D2 and D3 and U2, however when I had a 90/36 attainable U1 was been fully utilised with high bitloading, now U2 is been used more and U1 has lower bitloading.

I concluded the same as what you have for this guy.

I think a longer line is crosstalking with me, openreach cutback my upstream power (they now have admitted this to me via my install engineer). However this cutback seemed to have had consequences on my downstream as well although I do have very high QLN on my D1 and D2. My QLN is the highest I have seen on any HG graph. The QLN is also high on D3 but seems to have affected D3 much less, my conclusion is when attenuation is higher crosstalk has less affect since signal is lost anyway due to the higher loop loss.

Seems clear to me vectoring is needed on BTs network the 40% or so loss is not a myth from my own conclusions.

Although also notice the slight change in attenuation (pair swapped?)

Before crosstalk
After Crosstalk

--edit--

I see Madpom has severe QLN like myself, so now I seen someone else with as bad. That QLN is pretty high in my view compared to the average.

Edited by Chrysalis (Wed 06-Feb-13 00:16:55)

I guess I have a little issue with my line looking at shape of the Hlog graph? Here.

Not that i'm really bothered because my line syncs at 73 Mbps.

In reply to a post by simon194:

I guess I have a little issue with my line looking at shape of the Hlog graph?

It looks that way.

At the speeds your connection is achieving though, it's probably not worth the effort to pursue resolving it.

@all

Thank you for the in-depth analysis and wonderfully-detailed explanations.

A few extra points:
Whilst I don't know exact cable routes in my area, using "worst case scenario" the longest possible line from my cab that could run parallel to mine (closer to the cab) would be 1.2km long. However only an idiot would have wired the estate that way...! More likely it would peak at 800m in length.
My phone line itself is ok, a little crackly.
Something I am aware of is the cable from the pavement to my property runs under my lawn but wasn't buried very deep. When I first moved in (12 years ago) I put a spade through it by accident - and noticed several joins in it already - before the BT guy came to repair it. I know joins add to attenuation, but would it be this significant? I've mentioned the cable to Openreach engineers in the past after every fault and because it's buried they always offer the same response that they wouldn't deal with it unless the phone failed completely.

I've reported it to BT via their Twitter/email contact form. I've given them the Wholesale speedtest screenshot demonstrating IP profile, not heard anything back since. To be honest I'm not getting my hopes up, the response I expect is "it's working, be grateful". Although if the reason is crosstalk as you all suggest then is there much they can do?

I'll update this thread with any response I get in-case anyone else is in a similar position and is interested.
Thanks all!
Rob

I think you have your line length about right based on your attenuation.

On crosstalk in the long term they can doa lot with vectoring.

In the short term they can try a pair swap or replace cables, the latter is pretty unlikely tho. Its also possible as stated there could be an issue making the crosstalk worse than it should be, but of course to find that issue openreach would have to believe it to be a fault first to look for it.

Edited by Chrysalis (Wed 06-Feb-13 08:14:50)

If its not deep dig it up, it won't be buried then

On the subject of buried cables, does anyone know if the metal amour on the phone cable was exposed, could it cause interference?

My cable was damaged by the builders about 8 years ago exposing just the amoured metal. It was then just buried again, without being insulated. But I read that if this shielding is earthed at two points it could cause curent to flow due to the potential difference between the two earth points?

I think if both ends are "earthed" by being in the actual earth the potential difference isn't anything to worry about. A pseudo-earth in a piece of kit may be a different matter.