alterations to internal wiring complete (I hope)

[PaulSky]

Are all the telephone sockets on the line ADSL filter plates?

[eyemdee]

There are still quite a lot of errors and the noise margin is hovering on the low side, although it doesn't appear to have disconnected in 24 hours.

It might be worth getting hold of an extension lead and trying the router in your new test socket behind the master faceplate, just to rule out the possibility of internal wiring issues causing the noise.

[Lymmranger]

PaulSky - no only the extension I use for BB - the one downstairs has a filter in it - would it improve matters if I did?

Eyemdee - connection seems better (to me) and I think errors are improving (see previous thread stats) - I am still struggling to interpret the stats - are they acceptable or if not what else can I do.....

I cannot justify (to her indoors) ripping plaster off the walls to re-wire....
I have "carefully" changed all faceplates... I cannot do anything with the wire between them (or can I?)

...........It would be nice to finally get back to "unattended downloads"

[PaulSky]

Why did you have an adsl plate (is it pre-filtered) fitted in the office instead of on the main socket in the loft? Which socket is the first socket on the telephone line?

[eyemdee]

Latest 1 day time = 1 hours 15 min 42 sec
SF = 267240
CRC = 616
LOS = 0
LOF = 0
ES = 132
Latest 15 minutes time = 42 sec
SF = 2529
CRC = 0
LOS = 0
LOF = 0
ES = 0
Previous 15 minutes time = 15 min 0 sec
SF = 52930
CRC = 16
LOS = 0
LOF = 0
ES = 15
15 minutes interval [-30 min to -15 min] time = 15 min 0 sec
SF = 52931
CRC = 9
LOS = 0
LOF = 0
ES = 8
15 minutes interval [-45 min to -30 min] time = 15 min 0 sec
SF = 52930
CRC = 9
LOS = 0
LOF = 0
ES = 9
15 minutes interval [-60 min to -45 min] time = 15 min 0 sec
SF = 52989
CRC = 20
LOS = 0
LOF = 0
ES = 14

I just read right through your original post again and it looks like you still have the original problem, ie that of error "spikes". The stats above cover the last 1 hour 15 minutes and 42 seconds of which 1 hour and 42 seconds are detailed. By calculation, in that missing 15 minutes, there were 562 CRC errors and 86 ES "errored seconds", both of which are very high rates.

Try in the loft removing the faceplate from the master socket. This should disconnect all your extensions. Try them all. If any are not disconnected, then you still have extension wiring connected "upstream" of the master socket which ideally should be removed as it could be acting as an aerial.

Your speed is much better than it was and the noise margin reads better - but I think it still can be improved if you can get the the root of the problem.

[Lymmranger]

Why did you have an adsl plate (is it pre-filtered) fitted in the office instead of on the main socket in the loft? Which socket is the first socket on the telephone line?

BT line comes into the loft. Master socket is first connection. From master socket a cable descends (fixed in wall plaster to "office" - about 1.5m below). This (extension) face plate has been changed to a "PHILEX" ADSL Faceplate" which "seperates voice and data signals, prevents background noise, and is a permanent alternative to plug in ADSL filters" - as written on package!
Also a second cable leaves master socket and (somehow) gets to understair cuboard where there is a second extension faceplate. This has a plug in filter which then feeds skyplus box (contractually obliged to leave this in for 10 more months). I have also plugged a filter in to master socket (even though nothing else plugged in)(extensions are hard wired)

[Lymmranger]

I just read right through your original post again and it looks like you still have the original problem, ie that of error "spikes". The stats above cover the last 1 hour 15 minutes and 42 seconds of which 1 hour and 42 seconds are detailed. By calculation, in that missing 15 minutes, there were 562 CRC errors and 86 ES "errored seconds", both of which are very high rates.

Try in the loft removing the faceplate from the master socket. This should disconnect all your extensions. Try them all. If any are not disconnected, then you still have extension wiring connected "upstream" of the master socket which ideally should be removed as it could be acting as an aerial.

Your speed is much better than it was and the noise margin reads better - but I think it still can be improved if you can get the the root of the problem.

Not sure what else to do... I have removed every "unwanted connection" that I can - see above.
I do have a DECT phone base station in "phone" part of "PHILEX" plate....
....could it be that?
......I have noticed that BB is still connected over 2 days later.... That certainly is a first!!!!!!!!
I am about to redo my stats - see below

[Lymmranger]

So see below for 24hrs later...

What should I be looking for??????????

I feel guilty posting all these stats....

/usr/sbin/adslctl: ADSL driver and PHY status
Status: Showtime Channel: FAST, Upstream rate = 448 Kbps, Downstream rate = 7328 Kbps
Link Power State: L0
Mode: G.DMT
Channel: Fast
Trellis: ON
Line Status: No Defect
Training Status: Showtime
Down Up
SNR (dB): 4.9 18.0
Attn(dB): 36.0 20.5
Pwr(dBm): 19.8 11.9
Max(Kbps): 7360 932
Rate (Kbps): 7328 448
G.dmt framing
K: 230(0) 15
R: 0 0
S: 1 1
D: 1 1
ADSL2 framing
MSGc: 1 1
B: 230 15
M: 1 1
T: 1 1
R: 0 0
S: 1.0000 1.0000
L: 1848 128
D: 1 1
Counters
SF: 10660836 10660834
SFErr: 137596 12
RS: 0 0
RSCorr: 0 0
RSUnCorr: 0 0

HEC: 75319 4
OCD: 114 0
LCD: 0 0
Total Cells: 3132072575 0
Data Cells: 364869 0
Drop Cells: 0
Bit Errors: 0 0

ES: 10545 0
SES: 2371 0
UAS: 13 0
Total time = 1 days 2 hours 20 min 54 sec
SF = 10660836
CRC = 137596
LOS = 95
LOF = 0
ES = 10545
Latest 1 day time = 2 hours 20 min 54 sec
SF = 497311
CRC = 1431
LOS = 78
LOF = 0
ES = 322
Latest 15 minutes time = 5 min 54 sec
SF = 20819
CRC = 374
LOS = 56
LOF = 0
ES = 29
Previous 15 minutes time = 15 min 0 sec
SF = 52931
CRC = 611
LOS = 19
LOF = 0
ES = 84
Previous 1 day time = 24 hours 0 sec
SF = 5082263
CRC = 45691
LOS = 9
LOF = 0
ES = 4270
15 minutes interval [-30 min to -15 min] time = 15 min 0 sec
SF = 52930
CRC = 160
LOS = 1
LOF = 0
ES = 54
15 minutes interval [-45 min to -30 min] time = 15 min 0 sec
SF = 52990
CRC = 72
LOS = 0
LOF = 0
ES = 36
15 minutes interval [-60 min to -45 min] time = 15 min 0 sec
SF = 52930
CRC = 20
LOS = 0
LOF = 0
ES = 11

[eyemdee]

So see below for 24hrs later...

What should I be looking for??????????

SNR (dB): 4.9 18.0

Total time = 1 days 2 hours 20 min 54 sec Router bug - should read 2 days!

SF = 10660836
CRC = 137596

If you divide the CRC errors by the SF superframes you arrive at your error percentage, which in this case in 2 days 2 and a bit hours is 1.29%.

Some 15 minutes segments are much better, signalling a noise "spike" pattern of errors.

The noise margin at the point when you took these stats is also quite low which can cause both errors and disconnections.

When you say that the remaining extensions are hard wired to the master, have you checked that the ring wires on the master socket terminal 3 are disconnected? They are not needed and are a common source of noise like this.

[Lymmranger]

If you divide the CRC errors by the SF superframes you arrive at your error percentage, which in this case in 2 days 2 and a bit hours is 1.29%.

Some 15 minutes segments are much better, signalling a noise "spike" pattern of errors.

The noise margin at the point when you took these stats is also quite low which can cause both errors and disconnections.

When you say that the remaining extensions are hard wired to the master, have you checked that the ring wires on the master socket terminal 3 are disconnected? They are not needed and are a common source of noise like this.

Yes only T3 and T5 remain connected. Additionally, I have changed the faceplate of the extension for a new one. The only thing I cannot change is the wire.

Should i be worried about the router bug.... can it be fixed?/tweaked?

thanks