Как работает защита от перенапряжения Ethernet, коаксиального кабеля и телефона?

There are advantages and disadvantages to both options (protected and not protected).

In many cases, the surge protection can cause intermittent issues with different services such as...

• the cable's signal quality
• DSL (high speed Internet through a phone line) signal quality
• Ethernet connection instability

That said, it will prevent minor power surges that could indeed damage the equipment connected. I specify minor surges because major surges can not be entirely prevented without major impacts to your services, as mentioned in the previous section.

If there are small spikes in the electrical current going into your home, the power surge protector could save some of your equipments, but most important surges like those coming from lightning (thunder storm) will go through the surge protector and may damage everything anyways. Minor surges actually rarely cause any serious harm to your equipment.

In conclusion, to maintain a good quality of service, it is rarely a good idea to connect anything other than power cables to surge protectors. I've worked for both DSL and cable companies and it is one thing to check with the customer when they have issues as those protectors often cause issues.

Here is a fair explication (in 8 to 9 web pages) about surge protection

Basically surge protection either tries to absorb the incoming energy or block the incoming energy. A good "absorbing" protection will let the current pass at the lowest possible voltage with the fastest reaction and is capable of absorbing the required energy without breaking. A good "blocking" protection will block very fast and possibly break in order to efficiently block the energy (like a fuse).

DSL and TV coaxial cables are transmission lines for the signal and changes in impédance on the line result in signal reflexion and hence signal degradation. A good protection will typically come at a higher cost which is in line with the observation that protectors often cause issues.

So:

1. A surge should not fry every phone, as the energy is absorbed/short circuited (progressively) by initially broken devices.
2. When a cable is split, I expect the energy to be split too and travel down every line (at 15 to 30cm/ns). Hence the risk is higher to actually damage more devices.
3. The router is more likely to provide a global protection than the PC. The router might break and stop the current going to the other devices. The PC would only get part of the arrived energy (as there is a split of the surge) and might absorb "its part" - if the other devices "resist", the PC may continue to absorb incoming energy (and voltage build up).

An adjacent protector can only 'absorb' or 'block' a surge. Manufacturers provide a specification number for 'absorbing': hundreds or a thousand joules. How many joules can be in a destructive surge? Hundreds of thousands. Why do near zero (plug-in) protectors fail catastrophically? Most are recommend by ignoring numbers.

How does its 2 cm protector part 'block' what 3 miles of sky cannot? It doesn't. It does not claim to protect from typically destructive surges. It claims to protect from a surge that is already made irrelevant by protection inside every appliance. For example, your computer will convert a hundreds joule surge into rock stable low voltages to safely power its semiconductors.

A completely different device is also called a surge protector. This does something completely different with numbers that claim protection from all types of surges including direct lightning strikes. It costs tens or 100 times less money. It works for the same reason that Ben Franklin's lightning rods are effective.

Protection from destructive surges has always been about how electricity finds and gets into earth - be it via a lightning rod or surge protection. For example, TV cable already has best protection installed for free by a cable company. Cable must be hardwired to earth ground where it enters a building. Protection means a low impedance (ie less than 10 feet) connection to single point earth ground. No protector required. That hardwire means hundreds of thousands of joules are 'absorbed' harmlessly outside in earth. Then no surge is inside hunting for earth ground destructively via appliances and plug-in protectors.

AC electric and telephone cannot connect directly to earth ground. So a 'whole house' protector does what a hardwire does better. Again, a connection must be low impedance (ie less than 10 feet) to one particular earth ground: single point earth ground. (Telco installs that 'whole house' protector for free.)

Near zero protectors can be $3 power strips with ten cent protector parts while selling for $25 or $90. They are not selling protection. They are selling a massive profit margin to consumers who routinely ignore numbers. Those protectors do not connect to earth ground (wall receptacle safety ground is not earth ground). Manufacturers will not even discuss THE most critical component in every protection 'system' since they are marketing profits; not effective protection.

A UPS provides even less protection. Again, read spec numbers. A UPS typically has so few joules that its number cannot be smaller. If it has more than zero joules, then it can be hyped as 'full surge protection' in sales brochures to naive consumers (who ignore numbers). A UPS is not effective protection. A UPS has one purpose - to provide temporary and 'dirty' power during a blackout. To protect unsaved data. UPS does nothing and does not claim to protect hardware. Otherwise someone cited a spec number that justifies that wild speculation. Nobody did. A UPS also must be protected by recommended 'whole house' protection.

Protection is always about where hundreds of thousands of joules (ie a direct lightning strike) are 'absorbed'. An effective protector must connect a surge low impedance (ie hardwire has no sharp bends) to earth - and remain functional. Lightning is typically 20,000 amps. So a minimal 'whole house' protector is 50,000 amps. Only 'whole house' protectors have numbers that describe protection from direct lightning strikes (and other just as robust surges).

Protectors must remain functional after every direct lightning strike. 50,000 amps defines a protector's life expectancy over many surges. Quality of protection during each surge is defined by quality of and low impedance connection to what 'absorbs' energy - single point earth ground. Numbers describe what happens during each surge and what happens over many decades. An effective protector is only a connecting device to what actually does protection.

How many amps can an 18 AWG (lamp cord) conduct? Something less than 60,000 amps. All wires can conduct massive surge currents without failing. A surge incoming to appliances will more often damage semiconductors that routinely make tiny surges (hundreds or thousand joules) irrelevant - not damage wire. Protection is always about earthing before that electricity gets inside. So that a surge does not go hunting for earth destructively inside via appliances or near zero protectors.

A protector is only as effective as its earth ground. As was true and well understood over 100 years ago. But little known to so many today only educated by advertising, hearsay, and speculation. And who ignore numbers.

That "HowStuffWorks" citation is chock full of lies, myths, and half truths. It contradicts how electricity works. Is obviously best ignored. It never discusses what defines all protection - earth ground. And it ignores what is always found in honest recommendations: numbers. Where are hundreds of thousands of joules harmlessly 'absorbed'? A damning question.