When the phone rings

[#1]

The problem with the sort of faulty connection that this is likely to be, is that the ringing current & voltage  has a 'cleaning' effect on the connection, so that for a while after an incoming call, the problem can seem to disappear  However an experienced OR engineer should still be able to track down such a connection with his Hawk tester.

[#2]

I called Demon and they ran a test by ringing the number whilst pinging my IP address here.  This worked (I.e. dropped the broadband connection) on the second attempt.  But Demon have now asked me to report this fault to BT myself.  Is this really the correct protocol?  If our ISP here were good old Plusnet I would have expected them to take charge at this stage. 

[#3]

They probably are doing that on the basis of the line rental being with BT. But BT won't accept a voice fault report that it's affecting broadband and will tell you to contact Demon. I'd just do that and then ring Demon back telling them what BT said. If they still refuse conclude the call by saying "your support is useless, can I have my MAC" - that might make them take you seriously.

[#4]

This is very frustrating.  BT ran their test which gave the (voice) line a clean bill of health so back to Demon.  They asked me if I had changed the filters etc.  This line has a long history of unexplained disconnections and I had tried several filters without successfully eliminating the problem.  But that was a while ago before I discovered the correlation between disconnections and incoming calls so I agreed to try again.
With filter 1, I made four calls (with the fax machine disconnected (so the call was not answered).  Each time there was a 4 to 5 dB downward spike in the SNR (target 9dB) but no loss of sync.
Then I tried filter 2.  This time on the first attempt there was a loss of sync, followed by several others over the next minute or two.  But after that and even after I reconnected the fax machine nothing at all, incoming calls had no traceable effect.  There seems to be some sort of memory effect so that after the line misbehaves badly it is less likely to do it again for some time thereafter.  Is these type of phenomenology explicable?       

[#5]

Yep - classic symptoms of a poor joint. The extra voltage applied when ringing an incoming call bridges the the poor joint and makes it better - for a while till it degrades again.
I'm guessing that with the fax in place the phone doesn't ever ring for long enough for the joint to be made good.

[#6]

Quote from: jelv

Yep - classic symptoms of a poor joint. The extra voltage applied when ringing an incoming call bridges the the poor joint and makes it better - for a while till it degrades again.

Exactly as I said earlier in the thread.

[#7]

Oh yes, Anotherone.  Sorry, I completely missed that earlier comment (still haven't made it to Specsavers).