Print

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

[marineio]

Hi all, wasn't sure whether to just update my current post or start a new one, so here goes.

The connection has been relatively rock solid since my last post - occasionally would drop but comes straight back (normally dropping in the early hours of the morning) - with no latency issues or other noticeable problems... until recently. Whilst it has been rock solid, we have also been using powerline adapters to get internet into a dark corner of the house (from a WiFi and ethernet perspective), with no issues noted.

We are back to getting a reduced connection speed, with sporadic *ridiculous* latency that makes FPS games unplayable (looking at you COD). Other usage (such as video streaming, normal browsing) isn't affected as much except for when the connection completely drops. The recent high winds seem to have tipped it over the edge (i'm still betting something to do with the line between pole and house).

Have a load of data from DSLstats, not sure what is useful and what isn't to post? Cheers!

Edit: just did the BT broadband checker, and it immediately went to arrange an engineer visit - so I presume this is something they can see too!

[burakkucat]

Have you recently performed a quiet line test from a (wired, not cordless) telephone?

From the SNRM plot that you've shown, above, I strongly suspect there will be audible noise present.

[marineio]

Have you recently performed a quiet line test from a (wired, not cordless) telephone?

From the SNRM plot that you've shown, above, I strongly suspect there will be audible noise present.

Hi,

Unfortunately we only have a cordless phone - 17070 and option 2 just leaves a very faint hiss, but no popping etc. I have asked my folks to let me know if they hear anything of the sort during phone calls as I never use the landline myself.

Saying that though, the connection has now been relatively solid again for the past hour or so and SNR improved again too!

[burakkucat]

From sight of just that SNRM plot, I suspect that a joint is beginning to show high resistance (HR) or semi-conductive tendencies.

If the BT Wholesale checker has auto-selected an Openreach engineering visit then the automation must have sensed something abnormal. Hopefully it can be quickly resolved.

[meritez]

I'm curious if the cordless phone is affecting the REIN.
Is the cordless in the same socket as the router?

[marineio]

I'm curious if the cordless phone is affecting the REIN.
Is the cordless in the same socket as the router?

It is. And it's crossed my mind as well (we have moved the base station away from the modem - about 4ft away). Unfortunately this is the only socket in the house now as the upstairs extensions were disconnected when they ran the new drop wire to the new master socket (last year) and we weren't using them. The thing is though, up until I stopped recording data (as the line was solid), after the last engineer visit, it was solid aside from disconnections in the early hours. That cordless phone has been there since the start!

From sight of just that SNRM plot, I suspect that a joint is beginning to show high resistance (HR) or semi-conductive tendencies.

If the BT Wholesale checker has auto-selected an Openreach engineering visit then the automation must have sensed something abnormal. Hopefully it can be quickly resolved.

Entirely possible - apparently our line was testing the best in the neighbourhood according to their systems though on about the 3rd engineer visit last year - i'd love FTTP as I know it'd fix it, but it's not available here, but is available at my own house, and i'm extremely tempted to get it... 500mbps here I come!

Talk of the devil - it's back again. Quiet line test is a bit more hissy but still not anything i'd instantly get concerned about if I was on the phone.


Something i've noticed too on the UI of the modem, it's giving me 17.6db for SNRM down, and 0 for up. I presume the data from DSL stats is far more accurate than the UI?

[Alex Atkin UK]

Depending on the cordless phones, you could just be hearing a side effect of how they work.  If you unplug the base from the BT socket while listening to the dial tone they may still remain live, so you'd be able to hear if it the noise was the line or the phones own circuits.

The thing with the cordless phone introducing noise in the socket, that could happen at any time if say the PSU filtering is getting on a bit and failing.  Again unlikely, but you never know.

Hissing can also be a failing filter, easily diagnosed by turning off the modem to see if it goes away.

I think its a long shot to think any of this is the actual cause of the issue, but its always worth looking into just be sure.

[marineio]

Depending on the cordless phones, you could just be hearing a side effect of how they work.  If you unplug the base from the BT socket while listening to the dial tone they may still remain live, so you'd be able to hear if it the noise was the line or the phones own circuits.

The thing with the cordless phone introducing noise in the socket, that could happen at any time if say the PSU filtering is getting on a bit and failing.  Again unlikely, but you never know.

Hissing can also be a failing filter, easily diagnosed by turning off the modem to see if it goes away.

I think its a long shot to think any of this is the actual cause of the issue, but its always worth looking into just be sure.

I will try the unplugging thing tomorrow - or see if I can borrow a corded phone from my relative round the corner (i'm currently isolating as apparently been in contact with someone with covid ).

Again, appears to have gone back to normal now... i hate intermittent problems! Thanks for all the help again though, loads of good ideas!

[marineio]

As an addendum to this mornings post:

The faint hiss persists after unplugging the phone from the wall, so must just be how it is.

After unplugging the phone (at approx 4am) and plugging it back in, the SNR immediately tanked again and took til 7am to recover. Have fiddled with the wires with the master socket and trying this again, and seems like it was just a coincidence as it won't do it again (speed now 55000/10000 with good stable SNR), however through the day it's dropped multiple times.

Have ensured that the modem wire is not crossing any power supplies or cables, and it's a new one (received with the new BT Home Hub we had delivered last year).

[marineio]

Engineer has visited, and they've replaced the overhead line. Confirmed they'd replaced everything else, swapped fibre port, pair from pole to cabinet and drop wire.

Didn't seem overly interested in the graphs/snrm (including when he was doing his tests and it showed a small drop as he was doing them). Went through the history for the line of disconnections and speeds which didn't totally tally with the data I had (including where it'd synced lower at 50/6 or so, but then resynced higher later on in the day, it only showed the higher rate on their system).

Here's hoping it's fixed!

[burakkucat]

Engineer has visited, and they've replaced the overhead line. Confirmed they'd replaced everything else, swapped fibre port, pair from pole to cabinet and drop wire.

<snip>

Here's hoping it's fixed!

Hopefully so . . . as there is nothing else to swap out.

[marineio]

As a further addendum to the mystery. Following replacement of the overhead line, it seemed much more stable. However, we still had the occasional random interference/drop occur but it would go back to full speed.

Something the engineer mentioned was that the power strip (which had a surge protection element) neon was intermittent, and if you touched it, it would turn the neon on/off (power to the devices was maintained though). This strip is 20 years old.

We have replaced that with a new one, and the connection has been solid since going on for 5 days (except a resync at around 2am), Downstream SNRM 6.3 and Upstream SNRM 18.

Not something i'd have ever thought of, but for anyone trying to troubleshoot is something to consider!

[burakkucat]

Something the engineer mentioned was that the power strip (which had a surge protection element) neon was intermittent, and if you touched it, it would turn the neon on/off (power to the devices was maintained though). This strip is 20 years old.

We have replaced that with a new one, and the connection has been solid since going on for 5 days (except a resync at around 2am), Downstream SNRM 6.3 and Upstream SNRM 18.

Not something i'd have ever thought of, but for anyone trying to troubleshoot is something to consider!

Ah, now that is interesting.