Eratic U/S noise margin.

[#1]

Just when I thought I had solved my connection problems, something is playing silly beggars again.

My D/S SNRM is now fairly stable around 6dB (thanks @Gandalf), but this morning I had to give Routerstats a helping hand, and noticed a horrible jagged trace - see first pic. That was during a phone conversation, but since then, even without using the phone, I am seeing bouts of instability.

Anyone got an explanation?

[#2]

Narrow band interference (REIN) coincidentally in the upstream frequency band?

[#3]

I think I understand what you are saying @Baldrick1, and if it is I'll have the devil of a job tracking it down. Looks like a 'wait and see' situation.

[#4]

@Baldrick1 Thanks for your suggestion. As always, I should check before I complain

Following your comment, I went and checked plugs, sockets and connections, and found my printers power plug had somehow been partially pulled out of the wall socket! Pushed it back in, and had no problems since.

I'll monitor overnight, and mark your answer as a 'fix' if it continues to behave.

[#5]

@Baldrick1 - whatever is causing the problem isn't that plug, although it was strange that it seemed to fix it last night - I still have the same problem this morning.

[#6]

Thats a pity. I see that it kicked in in the morning. Was there any correlation today? Are there any clues if you keep monitoring and look for any consistency? 

[#7]

@Baldrick1 MAY have found the culprit! Have just changed my phone, and I'm suspecting the base unit on the old one, which had an aggrevating habit of appearing to not charge the handset.

The initial disruption you see WAS during a call, but the later ones were when the phone was not in use, so I'm still going to monitor the situation, but hope I've cracked it.

Thanks for your help.

[#8]

My fingers are crossed for you! This sort of problem is a nightmare as you find yourself chasing coincidences. There is also the distinct possibility that the source is external to your home. The other question is whether the noise (assuming this theory is correct) is getting in to the line from direct coupling into the telephone cable or mains borne interference getting into the modem via the power supply. Probably a longer shot but if you have a filtered extension lead around it might be worth putting the modem power through that.

The reason I asked if there was any correlation over days was because I noticed that it started around 08:30 and stopped just before 12:15, which seemed to be around normal work hours. It's all about looking for clues...