Print

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

[Basil]

Having received a new router from Zen and found it cured some of my ills (a Technicolour TG582n) I decided to wire it up properly. I had been testing with the router connected to the master socket in the loft via a LONG Ebay special cable. I fitted a new filtered faceplate with the extra two AB connectors so the wiring to the router socket was hidden. I reconnected the BT installed twisted pair cabling to the AB connectors, and another pair to 2 and 5 for the phone socket in my office. All seemed well for a few hours, but in the night routerstats showed serious noise and the connection speed dropped substantially. Connecting via the extension cable brought a quiet connection again. I decided to follow the BT laid cable and in so doing was forced to recall I had 2 switched mode power supplies on the opposite side of a studded wall, just near the sockets in my office. The cable ran *RIGHT* by them and alongside a mains cable feeding their 13 amp dual mains socket. One is for a Thunderbolt 10 MHz frequency standard, the other for a frequency divider board. I use them in my amateur radio hobby. The Thunderbolt box itself is slso very near the socket and wiring. These supplies runs 24 hours a day, seven days a week and have done for at least 2 years. I decided to turn them off and reconnect the BT wire. All has been quiet since and I am slightly optimistic one may have been the problem. I am perhaps lucky and not had issues with switch mode supplies causing noise that I have been aware of, so could some of the more knowledgeable folk tell me if this screen shot of the noise margin when playing up could be a SMPS gone awry? Thanks

[roseway]

I think I would expect SMPS noise to be more continuous. Your graph looks more like a temporary line fault or a dubious connection in your wiring.

[Basil]

You must be correct, I have been out and just returned, routerstats shows this noise, and the two PSU's I thought may have been the culprits were unplugged. I am thinking this is totally random, and they were just flukes when the line was quiet plugged direct into the test socket. The noise below is with the filtered cover plate on and just the one BT wire from the filtered cover to the router and phone in my office. I don't think a hard wired phone could cause this, could it?
Thanks for the reply. I will just bite the bullet and get the ISP to get BT in, although satellite internet does look very tempting right now. Cheers roseway

Latest noise graph:

http://www.gatesgarth.com/noise2.jpg

[JGO]

I agree with Roseway.   If it was winter I'd suspect a sparking central heating thermostat. It could be a pig to locate so it might be an idea to try a power line filter on your modem rather than hoping BT could help; certainly cheaper !   

[Basil]

Is there any future in running the router off a small car accumulator for a day or so? It's 12 volts @ 1 amp. I have some race car batteries that would run it for at least 48 hours, and are light enough to cart about inside the house. If my laptop hadn't got a tired old lump of a battery in it I would turn all power off in the house and run the two whilst I was out just off battery power and see what happened.

I do have another question, but I can't be sure the seeming randomness of all this isn't confusing the issue. I'll put it forward for critique by those who know.

It *SEEMS* that I get a slightly lower noise margin (17 down to 15 dB) when I run the router off my cheap extension cable, about 30 feet long, plugged into the test port of a brand new filtered main ADSL 1.0 "box". It's just draped along the boarded loft and then the carpet in my office. But it seems it's stable connected in this manner. The exact set up is a brand new ADSL 1.0 box from the Italian company that do the one with 2 extra A/B terminals. I can't plug the router extension cable directly into the test socket as it has 2 RG whatever connectors, so I am using a new dongle filter that came with the new router as an adaptor. So ADSL 1.0, dongle filter in test socket, extension cable in small dongle socket, then cable direct to router. Seems good like that. Not seen noise, even over extended periods.

I then try to return to a "normal" hidden wired circuit. I put the original BT twisted pairings cable that runs through the loft space onto the A/B terminals inside the cover of the ADSL 1.0 and the other end of the cable goes straight to a double new socket in my office. I only connect the two blue wires (blue and blue / white) to A/B and at the other end they terminate in my new double socket (phone / router).  I plug the router into the router socket with its new short cable. In that format I see noise.  It was the same with the old sockets, and I am 110% sure the wires are inserted properly, with the correct tool.  As far as I can tell the cabling is now not near any SMPS or anything likely to radiate RF. 

So how come the el cheapo extension from router port of master socket, via a filter dongle just used as an adaptor to mate differing plug / socket types, straight to the router is quiet, and the original BT wire is not? I have tested continuity and resistance with the cable pair shorted at one end and it seems perfect. I have trimmed the ends to be sure there are no near breaks inside the insulation near the ends. And how come the el cheapo extension cable shows a slightly lower RX noise margin than the BT wiring?

I am amazed I am still pursuing this, I usually have great impatience with this sort of stuff I guess I am now very curious as to WHY it isn't working properly.


Thanks for reading!