Weird Light Switch - Any Ideas?

[ban-all-sheds]

I've seen non contact testers used a lot and come highly recommended.

Neither by anybody who knows what they are doing.

 

[PBC_1966]

I find it sad that these devices seem to be so widely promoted in certain quarters these days. Bernard has given the thorough technical explanation, but if you want it summarized, they are simply not reliable. They can indicate something is live when it is not, and - more worringly - can indicate that something is dead when it is live. Get a proper multimeter and stay away from non-contact testers (and neon screwdrivers).

 

[Cinobite]

Thanks for the great explanation Bernard.

I'm finding it amazing how many tools are being sold that are simply dangerous.
I recently bought a cable/stud detector, it told me there were no cables in the wall beneath the sockets and live cables above. I didn't believe it so opened the wall up to check and the detector was completely wrong. Had I have blindly put my faith in it like most people would, I could have drilled into a live cable. How are these things being sold?

 

[Taylortwocities]

How are these things being sold?

Very easily to people who are too dumb to understand what they are being offered, and/or to understand what they are doing.

You may remember to case of the guy who sold virtually empty boxes for £10,000 each to governments all round the World as Bomb Detectors.
http://www.bbc.co.uk/news/uk-england-23768203

 

[PBC_1966]

Then there are those "miraculous" devices sold to gullible homeowners which, supposedly, just plug into any convenient socket and will cut your power bill by up to 20% or some such nonsense. And more expensive versions to install at the distribution panel for which I've seen even wilder claims. At best they're just a few capacitors which might correct for power factor if you have a lot of inductive loads, but as residential supply tariffs never include a penalty for p.f. anyway, they wouldn't save anything even if they did that.

Then how about all the ridiculously overpriced interconnection leads for audio systems? Some of them sell for thousands of dollars (or pounds) a piece. You can even find power cords which cost hundreds, supposedly "specially treated" sockets to improve the sound at similar sorts of prices - The list goes on and on.

P.T. Barnum would be proud.

 

[Cinobite]

Very easily to people who are too dumb to understand what they are being offered,

Selling crap to gullible people is fine. But what I mean is, if these things are actually dangerous to the point people can be killed, why are they legally being sold?

FWIW, in addition to spec the quality of material and shielding in AV cables does make a difference. I'm not saying spend £1,000 on a HDMI, but a £100 cable is not the same as a £2 from Asda

 

[ban-all-sheds]

Then there are those "miraculous" devices sold to gullible homeowners which, supposedly, just plug into any convenient socket and will cut your power bill by up to 20% or some such nonsense. And more expensive versions to install at the distribution panel for which I've seen even wilder claims.

I was at an exhibition once where a company selling that sort of thing had a stand.

After a short time discussing some of the basic laws of physics with them I was invited to choose between stopping questioning their product and allowing security to escort me from the premises.

 

[ban-all-sheds]

FWIW, in addition to spec the quality of material and shielding in AV cables does make a difference.

Even for digital signals?

I'm not saying spend £1,000 on a HDMI, but a £100 cable is not the same as a £2 from Asda

You should invest in cables of the same sort of quality as those used in the studios where they make the recordings and programmes you listen to and watch.

 

[eveares]

If he can't figure out how to do something that simple, he has no business to be calling himself an electrician!

Agree to!, use the 3 cables coming down as the line and switched lines for each light at the switch and do the loop in a JB up in the attic.

Testing for dead ? ( electrically )

Use a two probe tester and you will be almost dead certain it is dead ( or not dead )

Use a single probe or non contact tester and you may end up just dead.

True that Non Contact Voltage testers must not be used to prove dead, but they still have a use. Also how is one supposed to ensure that a circuit or cable is truly dead before you get access to the potentially live terminations/conductors to allow you to prove dead. By definition you can't be 99% sure that a circuit is dead (And thus safe) until you make direct contact with a two probe tester or turn off the entire DB.

Neither by anybody who knows what they are doing.

They are fine a first call for testing live/possibly dead PROVIDING that you follow with a proper 2 probe tester before commencing electrical work. (i.e Tracing the correct circuit to isolate in large commercial DB before testing again with a proper GS38 voltage tester.)

Even for digital signals?

Of course, if the shielding and quality of it is not good enough to protect the sensitive digital data from outside EMF interference and inductance, then you are going to have a noisier a signal what may affect the communication. True that shielding is more important in the analogue domain, but digital signals can be affected to by EMF and stray inductive currents.

After a short time discussing some of the basic laws of physics with them I was invited to choose between stopping questioning their product and allowing security to escort me from the premises.

What did you chose to do?, Should have persisted and told everyone that they were scammers.

 

[PBC_1966]

FWIW, in addition to spec the quality of material and shielding in AV cables does make a difference. I'm not saying spend £1,000 on a HDMI, but a £100 cable is not the same as a £2 from Asda

Of course cable quality can make a difference and it's not something to be skimped on. Try running TV distribution over the cheapest Chinese-made junk you can pick up on eBay, for example, and assuming that you realize the error of your ways you'll wish you'd just gone and bought a decent quality coax like Belden 8281 from the outset. And even with good quality cable, you need to choose the right cable for the job, so try to run high-power audio to low-impedance speakers over 100 ft. of 26 AWG wiring and the quality of reproduction will suffer regardless of it being good cable.

But there's a big difference between choosing the right, good quality cable for the job and being taken in by all the snake-oil products available, like $200 per foot speaker wire which has supposedly been specially treated every which way from Sunday, is supposed to be "directional," needs to be "burned in" for several days before you'll notice the improvement, etc. Or "super-conductive oxygen-free copper" power cords, a mere snip at a few hundred or so each, and you'll wonder how you ever listened to music without them (never mind the miles of plain old copper or - shock - aluminum wiring the power traveled over to reach your home and all the non-treated switchgear it went through on the way).

Of course, if the shielding and quality of it is not good enough to protect the sensitive digital data from outside EMF interference and inductance, then you are going to have a noisier a signal what may affect the communication. True that shielding is more important in the analogue domain, but digital signals can be affected to by EMF and stray inductive currents.

Indeed, digital signals are generally more tolerant of noise but only to the extent that the original signal can still be recovered at the distant end. Once you reach that threshold of detection, you lose the whole lot, or if hovering around the threshold are subject to frequent data loss. And even without any external interference, there is still the issue of proper impedance matching, capacitance and inductance in the connecting cables, crosstalk between different channels carried in the same cable, etc. Some of the cables and patch cords sold for Cat5e Ethernet, for example, are of abysmal quality and come nowhere near meeting the required standard. Try running near the limits, especially, and you'll be subject to all sorts of data problems. And again, that doesn't mean you need the overpriced snake-oil stuff at $200 per foot or whatever, just a good quality cable which is properly made and within the required specifications.

 

[Cinobite]

Even for digital signals?

Of course, if the shielding and quality of it is not good enough to protect the sensitive digital data from outside EMF interference and inductance, then you are going to have a noisier a signal what may affect the communication. True that shielding is more important in the analogue domain, but digital signals can be affected to by EMF and stray inductive currents.

Indeed, one of the common arguments for digital signal is "it works or it doesn't", this isn't exactly true. All digital data is sent, like a jigsaw, in chunks called packets. When it gets to the other end, the packets are put back together to form the information. Digital signals, therefore, can suffer from "Packet Loss", where not all of the pieces get to the other end. There can be lots of reasons for it, one of which being loss or interference due to a poor connection. Whether it's a lack of shielding, EMF from power cables, impurities in the wire material. You can test it yourself on a computer with; Start - Run. Type in "ping www.google.com" and look at the results, you'll typically see 4 packets sent with 0% packet loss.
For a layman's example - we've all seen freeview and digital radio break into blocks right?

But there's a big difference between choosing the right, good quality cable for the job and being taken in by all the snake-oil products available,

Absolutely agree. It's all about use. I have £120 HDMI on my £2,500 living room TV. But a £4 one on my £80 tesco 22" in the kitchen.
It's the same concept as these idiots who buy things like Beats headphones (lets assume they're as good as they claim), they spend hundreds on headphones or Bose speakers and then run MP3 music through them. Absolute waste of time. As they say, things are only as good as their weakest component!

 

[bernardgreen]

Add in the complication that some digital systems will detect a missing or corrupted packet ( of data ) and request a retransmit of that packet. Obviously not possible on a broadcast system but can be used on a point to point link.

 

[PBC_1966]

All digital data is sent, like a jigsaw, in chunks called packets.

Not all - There are plenty of examples of digital transmission which are not packet based.

There can be lots of reasons for it, one of which being loss or interference due to a poor connection. Whether it's a lack of shielding, EMF from power cables, impurities in the wire material. You can test it yourself on a computer with; Start - Run. Type in "ping www.google.com" and look at the results, you'll typically see 4 packets sent with 0% packet loss.

But as you already noted, there are different reasons for packet loss in the wide IP network, including in some cases packets which actually reach the destination machine intact but are lost there, so anything you get back from a ping test is not necessarily representative of a problem in actual cabling.

It's the same concept as these idiots who buy things like Beats headphones (lets assume they're as good as they claim), they spend hundreds on headphones or Bose speakers and then run MP3 music through them. Absolute waste of time. As they say, things are only as good as their weakest component!

Never heard of those headphones, but I cringe when I hear people (mostly the younger generation and slick salesmen, it seems) go on about how fantastic everything digital is, simply because it is digital. As you say, MP3 music is hardly high quality to begin with. Some of the digital audio broadcasts in the U.K. don't come anywhere near the quality of good stereo multiplex FM, some of the digital TV broadcasts are nowhere near the quality of a good PAL transmission of 20 years ago, etc. Of course, in many of these cases it's not that the system is not capable, but that technical quality has been compromised to squeeze those extra channels in (never mind the quality, just look at the number of channels you have!).

 

[Cinobite]

Indeed, iirc, freeview originally had enough bandwidth for around 30 channels at a proper quality. Then they crushed, squeezed and compressed the **** out of it. I don't even try to watch channels like Viva, Movie Mix etc - and to top that off some of them really cover the screen in logos and badges, some like viva you can expect to see 3+ frequently.
in fact, I watch little to no TV.

 

[PBC_1966]

I know what you mean. I've found the presentation annoying for years with the on-screen logos, banners scrolling across one show telling you about other shows later, shrinking the closing credits down to a corner to preview other stuff etc. And that's not to mention that with movies especially, many broadcasters now hack them to pieces to make them fit their schedules and to squeeze in more commercials.

My wife and I don't have any broadcast TV here at all, and don't miss it. DVD's are cheap enough now that we're building up our library slowly (mostly older movies and shows) and find that quite enough, as we seldom watch more than an hour and half or so each evening anyway, sometimes less, sometimes more on the weekends, and we're definitely not the sort of people who can stand having whatever garbage is being broadcast droning on incessantly all day long. I may put up an antenna sometime later just so we can get the local Redding stations for news, but even that's nowhere near even my "top 100" list of jobs to do!