What is an arc flash protection device (AFDP) and what is it used with?

We continue the topic of modern protection devices for home electrical panels. Next in line are Arc Fault Protection Devices (APD - wording from GOST), aka “Arc-fault detection devices” (USIS), aka arc-fault detection device (AFDD), aka arc-fault circuit interrupter (AFCI)... There are many names , but the essence is the same: this device is designed to disconnect the line if an arc breakdown is detected somewhere on the line. This is the theoretical and methodological part. Tests and dissection of devices will continue.

This article has a video version for those who prefer to listen:

▍ Hot little thing

Let's imagine something wrong happened in your electrical wiring - mice chewed the insulation, a terminal came loose, or wires broke where the cable was bent.
These, as well as a number of other faults, can lead to arc breakdown. An arc breakdown occurs when two conductors are at a very short distance from each other, which causes a spark to jump, an electric arc to ignite, and the electric current to flow through the air. An electric arc is very hot, and in moments it can ignite flammable materials around it, char the insulation and cause disaster. Moreover, the charred insulation becomes a conductor, which greatly simplifies the re-ignition of the arc.

A distinction is made between parallel and sequential arc breakdown. Parallel arc breakdown - when an arc ignites between conductors L and N or L and PE, for example, due to a self-tapping screw screwed into the cable. Or, for example, it begins to break through damaged insulation. In this case, most likely, the parallel arc breakdown will develop into a short circuit and the overcurrent protection will operate. Sequential arc breakdown, when the arc burns in an open circuit in series with the load, is the most dangerous. Neither the RCD nor the circuit breaker will work! There are no conditions for these types of protection to operate - the current is not exceeded (its value is limited by the load), and there is no differential current either. The arc will burn until contact is accidentally restored or broken. However, you’ve probably already encountered it - this is the same “crinkling” of a bad contact in a switch or socket.

If your wiring is done in strict accordance with all standards, then an arc breakdown will not cause a fire, but will generate streams of abuse from the electrician who will be repairing the outlet, where two charred stumps of wires are sticking out of the socket.

The key word here is “if”. Unfortunately, the harsh reality may be:

• Old aluminum wiring, which was repaired, I don’t understand how and I don’t understand where
• Wiring laid inside combustible walls
• Rodents that have eaten wire insulation down to bare copper
• Unfortunate builders who damaged the insulation of wires with a self-tapping screw
• A huge number of carriers, tees and other electrical products of dubious quality, lying in hard-to-reach places surrounded by flammable objects

In an unfortunate combination of circumstances, an arc breakdown can cause a fire, resulting in casualties.
It turns out: with a careless attitude towards electrical maintenance, we can get a phenomenon that can lead to a fire, and which none of the protective equipment used can detect. Sounds unacceptable.

Five-pin relay

We connect the alarm output to the contact

Imported relays under the Saturn and San Hold brands have proven themselves to be the most reliable and commercially available; relays from other manufacturers are also used. That is why a voltage stabilizer in the running lights connection circuit is extremely necessary. Contacts 30 and 86 are swapped.

As you already know, the use of DRLs in conjunction with other lighting devices is prohibited. Operation voltage: not less than 8.0V. Options for circuit solutions for connecting relays.

Otherwise, traffic police officers will issue you a fine, or even take your car to the impound lot. So what's going on? The magnitude of the control voltage and the voltage switched through the contacts can be different and do not depend on each other. Obviously, your DRLs will always work as long as the key is turned in the ignition, no matter what lighting you use.

Read additionally: To repair a broken wire in an electrical appliance, you need

The simplest scheme

It is important to note that if the relay has been operated for a long time when switching power circuits in extreme modes, then the spark that jumps when closing or opening the contacts creates carbon deposits between the contacts and because of this, the actuator may not work or will not work correctly. Case temperature The relay coil consumes about .5 watts of power, which is why its case can get quite hot during operation - this is not criminal.

Everything that is not prohibited is permitted.

Now, if you try to start the car with the security switched on, contact 30 will open with contact 87A and will not allow the engine to start. I would like to emphasize that DRLs are intended to indicate your vehicle in front of other road users, and not to provide additional illumination of the roadway. Self-locking motor blocking circuit with self-locking. Accordingly, the other two terminals should show infinite resistance - these are our normally open working contacts. And so when at rest - the circuit is broken - that is, nothing comes to the plus of the fog lights, as soon as the ignition is turned on, the plus from the battery is switched on.

Simple homemade products for a car, tips for car enthusiasts and do-it-yourself diagrams

IMG: link And so. For every change in the design of a vehicle, a certificate must be obtained, which in itself is neither quick nor cheap.

Installation method: terminal blocks, terminal pins, soldering into a board or installation on a DIN rail. In the figure: Contacts 85 and 86 are control contacts. How to properly connect DRLs (daytime running lights) to a VAZ via a 5-pin relay

▍ Catching the ghost by the tail

Engineers are still searching for a reliable way to detect arc faults; if you look through publications in scientific journals, you can see attempts by researchers to use different techniques, including fashionable neural networks. The better the technique, the higher the probability of detecting an arc fault and the lower the number of false alarms:

In this case, the device in the electrical panel has only one way to detect an arc breakdown - to analyze the magnitude and shape of the current supplied to the load. All manufacturers of modular arc flash protection devices collect a signal from a current sensor, but process the data differently and do not disclose details, citing know-how. Therefore, I can only tell you the general approaches that are disclosed in scientific publications, but in the hunt for details you will have to catch and solder the developers in a pub.

It is still possible to detect an arc breakdown due to one feature - the arc does not ignite immediately. The voltage must rise to the breakdown voltage, after which a spark jumps in the gap, which ionizes the air and allows the electric arc to ignite steadily. And since we have alternating current in our network, and the current changes direction 50 times per second, passing through zero, the arc lights up and goes out 100 times per second, leading to specific distortions!

I’ll show you with examples why I made a small stand. I measure the current in the circuit with a current transformer (blue line), voltage - through a divider (yellow line), the scale in this case is not important. Almost ideal load - fan heater:

It’s simple: as the voltage in the line increases, the current increases proportionally. The voltage drops - the current in the circuit drops. Pay attention at the point where the voltage crosses zero - the current increases immediately. And this is what the current looks like in the same circuit if I separate the contacts and cause an arc breakdown in series in the circuit. A step appears - the current appears only after the voltage reaches the breakdown voltage of the gap between the conductors:

You might think that it is enough to simply monitor whether there is a step in current consumption when the voltage passes through zero. But alas, this method does not work, since such a step appears in many types of load. For example, if the device has power regulation with a thyristor regulator, which creates such a step and, by changing its width, regulates the power in the load. Just look at what the current graph looks like for a vacuum cleaner with a power regulator:

In addition, the ideal case of only one load on the line is rare. More often there are several consumers on the line, and their currents are summed up. As a result, the graph begins to look completely unclear. The graph below shows four consumers (1 kW heater, 2 kW electric kettle, vacuum cleaner with a regulator at half power (approximately 800 W) and a powerful switching power supply loaded with ballast (approximately 180 W)). On the left there is no arc breakdown, and on the right there is a sequential arc breakdown of a 1 kW heater, i.e. The arc current is only a quarter of the total current consumption:

What to do? Let's look carefully at the graph with sparking - the rate of current increase in the circuit after a breakdown is enormous, the step is almost vertical! This means we need to look not at the appearance of the step, but at its verticality. The easiest way to do this is by analyzing the spectrum of the signal; the steeper the step, the wider its spectrum. I visually depicted it in this picture: As a result, the principle of operation of the protection is simple - we constantly analyze the spectrum of the signal from the current sensor. If suddenly it changes dramatically, we determine how it changed. If we observe a rise in the high-frequency part of the spectrum, this means an arc breakdown and we turn off the load. True, in reality there are nuances...

Operating principle of spark protection devices

How does the spark protection device, which is located in the electrical panel at the entrance to the house, see the sparking of the wire in the farthest socket in the bedroom or living room? What magic is used here?

Of course, there is no magic here, everything is based on the laws of physics. The device mainly monitors the spectrum of the current passing through it.

When sparking begins anywhere in the electrical wiring circuit, firstly the sinusoid is distorted and it becomes jagged. The current and voltage begin to change abruptly. Interference occurs.

However, if the protection were configured to monitor only these parameters, there would be a lot of false positives. This is exactly what the very first copies sinned with.

Therefore, the latest high-quality ultrasonic sensors or ultrasonic sensors analyze a lot of parameters:

• size

• form

• polarity

• duration

• and the pace of the jumps

Manufacturers of spark and arc protection devices are required by the GOST standard to have the following three main tasks:

• analyze the current, and at the same time make sure that its source is the arc, and not the payload

The device has the right to ignore and let through anything that sparks with an arc current of less than 2.5A.

• find out how dangerous this arc is by its power

After all, simply plugging a plug into a socket also causes sparking. But nothing should be disabled.

• if the first two tasks are successfully solved and the current is detected, then it must be broken in a given time

▍ False positives and invisibility

False alarms are a headache for AFDD developers.
There is complete anarchy in the electrical network - every load consumes current as it wants, low-quality devices are still actively creating interference. For example, look at what the current looks like when I just turned on a sander with a dying motor:

And this is what the current of the welding machine looks like (I took a regular transformer and twisted it with a carbon electrode):

In this case, the device should not formally work - there is no arc breakdown. Now imagine that you have a dozen such devices on one line - their currents will add up, the noise will be summed up, and the developer will despair.

It turns out to be a rather non-trivial task - on the one hand, you need to increase sensitivity, and on the other hand, prevent false positives. Therefore, developers are in no hurry to reveal their cunning algorithms. I found the only description of the operating algorithm here.

And here it is important to note: Not a single AFDD is immune from false alarms!

Moreover, of all the protection devices, the AFDD is probably the only one that can give a false alarm in
good condition
. This is important to remember when designing! (but more on that towards the end). For example, there is a bastard like me who will dig up an old Soviet UFO-B lamp (a high-pressure mercury arc lamp with a resistor ballast) and plug it into the network. When ignited, the current consumption graph looks like this:

It causes an arc breakdown on the igniting electrode, and the lamp caused a false alarm every time it was turned on! It was difficult to find such problematic devices, but I succeeded. In the process of testing the AFDP, I tried different types of loads and found my kryptonite for each model of the AFDP. However, the vast majority of household devices do not cause problems.

Any state does not tolerate anarchy, and therefore fights it. Many countries have electromagnetic compatibility requirements for devices - they must not interfere with the operation of other devices on the electrical network. Therefore, the power and range of interference that can leak from the device back into the network is limited. The consequence of this was the installation of filters in devices. The filter reduces high-frequency noise generated by the device. For example, any switching power supply has such a filter in its circuit, so I took the first Meanwell power supply circuit I came across (I like them) and circled the filter:

The surge protector is an invisible cap: everything that happens behind it becomes invisible to the AFDD. Technically, in addition to chokes or capacitors, you can use an isolation transformer. For this reason, my ersatz welding machine for welding strands did not cause false alarms - the arc breakdown was in the secondary winding, so the transformer worked as a filter. Adding a simple filter (taken out of a microwave) in the form of a common-mode choke completely eliminated the problem of false positives with the UFO-B lamp, which I described above.

It follows that the probability of false positives increases sharply if a device is connected to the network with:

1. There are no such filters, simply because it is old. For example, the 1960s, when the requirements were simpler
2. There are filters, but they are not effective due to the design curve or cost savings. This is often the case with noname devices, where, to save money, everything that is responsible for quality or safety is thrown away. A good filter is heavy, as it requires a lot of copper and iron.

It turns out that high-quality electrical devices for AFDDs that meet modern requirements should not cause problems.
And if you have one such problematic device (for example, your grandfather’s favorite electric razor), then you can “hide” it from the AFDD with an invisible cap in the form of an additional surge protector. Specialized filters in the form of radio components can be viewed here: (https://www.promelec.ru/catalog/409/455/494/) although, I hope, SPD manufacturers will have such a product as an option. And I think many people are concerned about the question - does the AFPD work for welding? - no, I tried several inverter welding machines - everything is fine.

CS-CS.Net: Laboratory of the Electroshaman

AFDD example from ABB - S-ARC1

This post is called “You are tired of everyone sending me news that is not news! Before sending it, have you even thought about where it would be advisable to install arc protection?!” Exactly. Here are all these links and “news” of yours, sent for the tenth and next time “Do you know? Here they released something like this!” (and if some bastard “reminds” me about the UZM-51Ts, I’ll fucking ban her) made me think about who is trying to produce what and where to use it all.

There are a lot of photographs in the post that I borrowed from different authors: from my own blog, from my own community and from other blogs and sources. I cropped these photos a little and tried to preserve the original copyrights, and signed them here and there.

If you found this post just by searching and immediately wanted to write to me about new samples, then first take a look at the Cooperation section, and then read that 99% of the time I will use AFDD from ABB, because they are more structurally thought out and made for the application exactly where they should be. I don’t want to test different knee crafts.

Attention! In 2022, Meander released a defective batch of UZM-51m and UZM-50Ts, I got fed up and switched to voltage relays from NovaTek in 2022! Read this post for detailed information!

Since 2022, ABB began selling their AFDD S-ARC1 and DS-ARC1 in Russia. Here is a link to the catalog: ABB_AFDD-Brochure-2020.pdf, and here is a link to detailed technical information on how AFDDs from ABB are designed and work: SP-256.1325800.2016-Changes-2021.pdf.

So, as I recently wrote here, now no one reads or comprehends anything so much that everyone is now trying to quickly post a “link to a friend,” even if this already happened a couple of messages ago. Therefore, information is now spreading according to the viral scheme: one heard it, passed it on to others, those passed it on to others, and so on. Even MetroElf laughed about this when he talked about one simple “accident” in the subway. Here is a link to his post with video.

Because of this, all of our arc protection devices have now become a new magical artifact - no less! If you laugh like that, then the archetypes of society do not change. Only the things through which they manifest themselves change, heh. In general, everything around arc protection devices is now covered in mystery, rumors, superstitions, and they are turning almost into a new panacea, just like a cure for all diseases. At the same time, the game of “higher and lower” started to play and the first throwing of poop began: “Oh, don’t you have arc protection? Wow, you're a loser! But you know, here Meander/ABB/horseradish-who-else-released...”

But does anyone ask the main questions: HOW does it work and WHERE to use it? AND WHAT HAPPENS IF THIS DEVICE GLUCKS ? So I wondered. And I didn’t like the result, so - I’ll spoil it right away - I decided that I would wait a while until it was all debugged. And, most likely, I will only use AFDD from ABB. I’ll tell you why in this post.

Let's understand the terminology. What exactly to call such devices is not yet entirely clear, because in our country the attempt to produce them immediately begins to turn into some kind of game of patents and “We’re cooler than Europe, now we’ll come up with our own and cheaper.” Moreover, each manufacturer (and I know only two so far) calls their device differently. Maybe so that they are well searched by search engines, or maybe to grab some patent for the name. Therefore, at the moment there are a bunch of names:

• AFDD (Arc Fault Detection Device) - international, generally accepted, which I will use
• Arc protection is my version without acronyms. Simple and clear.
• AFDP (Arc Flash Protection Device) - variation of Meander
• UZIS (Spark Protection Device) is a variation of EcoLight.

What is the purpose of arc protection? Yes, damn it, it’s simple - detect sparking in the wiring and disconnect the problem line. Wiring sparking is determined by analyzing the spectrum of the current that flows through the line. This is done using a powerful microcontroller, which calculates the signal on the fly and decides whether to consider it an arc or not. Of course, the signal analysis algorithm must be such that it is guaranteed to distinguish an arc from, for example, welding or sparking of a switch. And this algorithm is the MOST valuable. But making iron is not that difficult.

Why is all this necessary? This first appeared in Europe and America, because there this arc protection device is needed like air: there are a lot of frame houses in which the wiring lies open among the wood, and all connections are made using PPE. There, if some connection starts to spark, there will be a fire, because there is something to burn around. For example:

Example of wiring in a USA/Europe frame house

And here is one of the shields that are installed in such frame structures. Please note that there the SIP goes directly inside the shield (and here this is prohibited).

Example of a shield used in America

In our country, the rules have always been more stringent and already designed for the fact that the connection will burn. Well, or on the Harsh Russian LED:

Harsh Russian LED

In our country, all wiring that goes into combustible structures must be done in metal pipes. That is, if you are building a frame house in our country, then use steel pipes. Moreover, I remind you that the wall thickness of steel pipes is also indicated in the rules in such a way that the melting copper of the cable cores does not burn through it.

But in other countries, initially no one thought about it or cared. And now, probably, they are biting their elbows, because according to their standards, they now have to install an arc protection device. Of course, these are my guesses, based on facts and logic, but for me such conclusions are better than a stupid repetition in the form of “But in Europe it’s been ten years now... but here... we’re backward.” Even the statistics are different. If “with them” most of the fires are caused by crappy connections of PPE and wiring in the tree, then “with us” it is due to low voltage in old houses, because of which refrigerator compressors burn. In any case, such statistics were in 2011-2014 for the Podolsky and Klimovsky districts of the Moscow Region. Almost first hand.

Why is an arc protection device not a panacea? Let's come up with different cases of wiring problems and think:

• If the wiring is located in a non-flammable place (steel pipes among wood, concrete screeds, plaster) then there will be nothing to burn there. It will boom, smoke, and most likely nothing will happen.
• If the connection just gets very hot, for example as in the photo above, then there will be no arc. And AFDD won't work! Take this into account! And the temperature from this connection can melt or set something on fire.
• Alternative option: “Master! We saved you some money! All sockets were made with a 1.5" cable (I now also consider such a shield) with circuit breakers for 16..25A. In general, when the cable heats up, it will be long, tedious and intense. Arc protection won’t help here either, and the cable can heat something around it and set it on fire.

That is, AFDD (arc fire protection) does exactly what it does - saves you from prolonged SPARKING. But NOT from overheating of the crappy connections!

Where will an arc or spark be harmful? Let's think together again (written based on the page about AFDD from ABB):

• EXPLOSIVE premises (but those where a dangerous concentration of something for an explosion is not constantly created). All kinds of gas stations, gas stations. Or maybe places where they work with a thin flammable suspension - flour, coal dust, toner, sawdust. There, the slightest spark can lead to a massive explosion. And here paranoia will not hurt: it is better to cut off the lines than to get an explosion.
• Sparking due to poor connections (I remind you: but NOT heating of these connections!) - twists, poorly tightened contacts of sockets, circuit breakers, terminals. It will help when the connection has become so oxidized due to heating that it begins to spark. While it's heating up, it won't help.
• Wooden houses - as additional protection if the wiring is done according to the rules.
• Places where a person can fall asleep with devices turned on: living rooms, all sorts of hospitals, and the like.

In general, we draw the main conclusion : arc fire protection (AFDD) is NOT A PANACEA! AND IT WILL NOT HELP WITH HEATING CONNECTIONS OR CABLES! IT WILL ONLY WORK IF A SPARK OR ARC IS PRESENT! In other cases, you can easily get burned if the cable overheats and the insulation around it in a wooden house catches fire. Look at this post about a burnt-out shield in Irkutsk: AFDD will NOT help there either - due to the crappy contact in the meter, the connection did not spark. It just started heating up and gradually set the shield on fire.

Conclusion two : AFDD is extremely necessary in Europe and Western countries, because there the wiring is laid “as is” inside the wooden structures of houses, and they are FORCED to install arc protection so that it does not flare up. And that’s why they have arc protection “Strongly Recommended” - “strongly/strictly recommended” for use. But so far we have stupidly copied this GOST, without even understanding what’s what. And, of course, the majority immediately began to shout: “But with them... it’s mandatory, strict, dangerous! We need this too." But to think about the reasons for this - no, we don’t want to do that. Why repeat stupidly without understanding the reasons?

And now that I understand the reasons why and why arc protection was created in Western countries, I can understand the tasks and requirements that it must solve. Let's say these are:

• Let all equipment work normally in normal mode and not react to sparking forks, operation of commutator motors or welding machines in harsh American garages.
• Disconnect the line only if there is a stable arc. In simple terms, it’s not when something goes wrong, but when the wiring starts to go completely haywire and a decision needs to be made immediately. That is, have rough sensitivity and good noise immunity.
• Disconnect the line permanently. Without any restart. We turned it off and won’t turn it back on!
• To avoid interfering with other loads, turn off only the line on which the problem occurred. This is VERY important for houses, because no one wants to get burst pipes in the winter, having arrived from a vacation at the seas and seeing that everything has been cut off due to a forgotten charger in the attic bedroom. We will talk about this point very strictly and a little lower in the text.

That is, the most important thing that needs to be highlighted here is that arc protection should not have paranoid sensitivity, as our manufacturers try to do (and get a lot of glitches) and turn off with every sneeze. Its task is to cut when the arc is already stable, when we are definitely on fire. And chop it so that it doesn’t turn on later.

And now we come to the most important question. Where should arc arc protection be installed on the shield ? This is where I, in fact, have the most important wave of swearing and the desire to hammer nails into everyone’s heads. Because for some reason almost all of our manufacturers decided that not only should arc protection respond to every sneeze, but it should also be installed immediately at the input of the switchboard - instead of or in the same place where the voltage relay is installed !

And that’s why our manufacturers are currently turning out complete shoals. If arc protection is installed at the input of the switchboard, then it must be able to detect an arc in a wide range of currents. Let's say from 1 to 63A. And also define it the same way. If we measure the current through a current transformer, then problems will begin: with small currents it will lose sensitivity, and with large currents it will be lost again, because the transformer will go into saturation! But they want to literally understand the Western standard (which has already become our GOST - it was stupidly translated and accepted) and make sure that arc protection is very sensitive...

Second. My beloved Meander took a radical path. He decided to force arc protection on EVERYONE. Because the new GOST, and because they were one of the first to receive patents for it. Moreover, there are no alternatives, and you remember the post where I cursed at Meander for this. They were offended by me, said that the post was an order from competitors and “we wanted the best” and have not communicated since then.

What is the problem with input arc protection ? The fact is that it becomes uncontrollable by different currents and glitches with sensitivity will MANDATORY. Why - I described above: due to the fact that it is impossible to obtain the same sensitivity of current sensors in different ranges of this current. That is, for now, in my opinion, arc protection at the input will either be almost insensitive, or buggy as daylight.

And now - the most luxurious part. Let's compare the European approach and Our approach. In Europe, there are more private residential buildings (and even with wiring lying open in the tree) than apartments. With us it’s the opposite. Therefore, when they created AFDD, they thought about protecting houses and that houses have a BOILER ROOM. Well, or at least a gas water heater or boiler, which is responsible for all the HEATING of the house. That’s why they initially designed AFDD as a SEPARATE RCD-type device: if you want, install it, if you don’t want, don’t install it. If you want, put several on different groups of machines; if you don’t want, put one or two for the whole house.

But if you look at our crafts, you get the feeling that their creators (and the same Meander) thought only about apartments and did not know that there are residential buildings and especially cottages in the world. And these houses have boiler rooms. And in our climate there are also severe frosts. In Europe, a frost of -15..20 degrees is a state of emergency, but here we have the norm, and for some regions it’s completely garbage.

In such frosts, the life support system at home is very important. And that’s why we install generators and think about all sorts of alternative energy sources, inverters and SMS notifications. If you are not at sea, then you receive an SMS saying that the power has been lost, you need to run and add fuel to the generator.

And we are being asked to install arc protection at the entrance to a house or apartment. What happens if you went to the seaside for the New Year (let’s take the situation: it’s frosty, and you’re gone for a long time), and the arc protection glitches or really honestly turns off along the arc? The worst thing in an apartment is the loss of communication (security, servers, video surveillance) and the leaking refrigerator being thrown away. And then you can try to wash it.

But in the country house there is a PI##A BOILER ROOM. Yes, in the worst case scenario! And complete. This is real, especially if the house is heated with gas, because for gas boilers it is necessary to make a window (grid, slits) in the door of the boiler room so that air can pass normally to the boiler for gas combustion. The boiler stops working - the cold comes through the window - the boiler room freezes - the pipes burst from the frost and you end up with a lot of money. And in the worst case scenario, you can’t quickly start up the entire system to warm the house.

Here it is - the price of UZM-51MD with input protection. And that’s why I’m so persistently pushing Meander so that they either leave the UZM-51M in production, or make arc protection switchable in the UZM-51MD. I don't need arc protection directly on the input. At the input, it is necessary to protect against emergency voltage in order to protect equipment from it. And my boiler room - let it burn from the arc. Most often, the boiler room is made into a separate room, tiled or with bare walls, where there is nothing to burn. And I would install arc protection on sockets and household appliances. Separately, in groups, like UZOshki.

I wrote a letter to Meander, but never received an answer. And I still don't understand their policy: why do they want all their protection devices to ONLY have arc protection and why can't this arc protection be disabled? Are they trying the Steve Jobs method? “You must want to buy this”©? But the cost of a mistake is not the cost of the UZM-51MD (even if it costs 5 thousand) - but this is the boiler room of a house costing tens and hundreds of thousands and wasted time replacing defrosted pipes and equipment!

If the arc protection in the UZM-51MD could be turned off, then I would be the first to switch to MDshki and start testing them. And my customers would be calmer and decide on their own responsibility. For example, if you live in the house permanently, then you can turn on arc protection and become a tester of Meander’s products (and Meander will only have to collect reviews like “I turned on such and such a tool, glitches started”). And if you leave this house for a long time, then:

• The shield has a shutdown of everything unnecessary. Only the boiler room, life support for the house (well pump, security) and other little things like the gate drive remain operational. That is, 3/4 of the house simply turns off. And there is nothing for the arc to arise from at all.
• At your own discretion, you could decide between “everything will shut down” and “everything will burn to hell.” And if I wanted to play it safe, I would turn off the arc protection when going to the seas in winter frosts (and in the summer you can leave it on - there are no long subzero temperatures in the summer).

The next thing that no one thought about in our country was HOW to then determine the line (or group of lines) on which the arc arose? EVERYTHING has gone out for you. You turn on the protection - it turns off again. What should I do? How to understand? If the arc protection device were separate, then it would be possible to decide where to install it. For example, don’t put them on non-switchable lines at all, but put them on the rest in groups in the same way as we put UZOs: “Light”, “Room Sockets”, “Kitchen Appliances”, “Kitchen Sockets”, “Bathroom”, “Air Conditioners and Warm Floors” "

That is, the correct method for using AFDD is the same as for ultrasonic protection devices - divide the lines into logical groups and protect these groups from the arc separately so that the operation of one AFDD does not cut off the entire house or apartment. European manufacturers do everything this way: AFDDs are produced as separate devices in such a way that they are attached before or after the outgoing line machine. There is a lot of logic and precision here:

• We can then issue the AFDD at different currents and adjust the arc current sensor to operate within certain rated current ranges. Let's say from 20 to 100%. And then inside for ratings “16A” and 32A” it will be possible to install current sensors that will give normal sensitivity.
• It is logical to place AFDD on a small group of lines or strictly one per line. This will reduce false alarms, because if you install AFDD on a large number of lines (or on the entire apartment/house), then the total current of interference and various crap can be detected by the device as an arc. This is what we observe in all our crafts as incomprehensible glitches. Manufacturers either coarse the sensitivity of the sensor so much that it doesn’t even detect sparks, or they make it so paranoid that it goes off when you start a vacuum cleaner or microwave. This is why foreign AFDDs are produced with low rated currents. For example, ABB S-ARC generally only has a rating of up to 20A. Do you think this is for a reason? Oh yeah. This is no accident! Just so that no IDIOTS would think of putting AFDD directly on the darkness of the lines and get glitches. In exactly the same way as with 10 mA (0.01 A) RCDs - they are produced at 16 or a maximum of 25 A, so as not to receive false alarms from natural leakage currents.
• A separate device will immediately show which line the problem is on and will in no way turn off the boiler room: its small range of ratings simply will not allow you to install it on the whole house at once. Eureka, damn it!

So let's take a look at ABB's S-ARC1 (I haven't seen it in person yet - I'll try to ask ABB for a sample to test):

ABB S-ARC1 arc protection device (©ROMUZ)

Doesn't remind you of anything? Of course it does! First of all, the difavtomat of the DS201/DS202C series, and secondly, the usual module of the S200 series. What can we glean from this photo and the data on the ABB page:

• All its clamps will coincide with others in the System Pro M Compact series: automatic devices, RCDs, and automatic devices. Therefore, you can use the same combs and other installation delights here as with the rest of the ABB module.
• Voltage can be applied both from above and from below. The designation of contacts “1/2” or “2/1”, as on a difavtomat, tells us about this.
• AFDD is immediately combined with a line circuit breaker of a certain rating (B, C and from 6 to 20A). This will greatly simplify the assembly of the shield and the logic of its operation: such AFDDs can be immediately connected with a PS2/xx comb, and since both phase and zero are connected to them, the zero busbars will completely disappear in the shield. We also have a normal 6 kA breaking capacity, and God knows what.
• This device is electronic. In general, all arc protection will be electronic, because a powerful microcontroller is needed to calculate the arc spectrum. This is fine. But in this case, all the stuffing: electronics, AFDD and overcurrent protection circuit breaker for the contacts themselves are perfectly combined in one device. Installed - and there is no need to mess up anything else in the shield.
• It is most likely made at a factory in Italy, in the same place where the RCDs with Difavtomats are made (this should be clarified): the body parts are very similar.
• The design itself (the fact that AFDD is combined with a machine gun) and a number of denominations forces you to place it CORRECTLY - on a specific outgoing line. ONE line. And the manufacturer, in turn, can guarantee that all sensors and algorithms will work normally on the same line. And not in the whole house with such a closet (Povarovo).

Here are the answers to all your questions, right? And I pulled this logic out of my head and a brief description of ABB’s AFDD. But here the meaning put into AFDD is clear: there is foolproof protection (“I put one on the whole shield, and it’s buggy”), and it’s immediately combined with a machine gun (it’s more convenient to remake old shields), and it will most likely work reliable - because the task of AFDD (I remind you again) is to turn off the real long arc, and not the sparking from the engine manifold.

And now - dessert and delicious. What's going on with us. Logics? What's the logic? Let's cram in as much of everything as we can to make it cool! And fewer alternatives, gentlemen - the buyer should not think, he should consume the product! And the more often he changes it, the more profitable it is for us.

The first is my beloved and long-suffering Meander UZM. Why start cutting the goose that laid the golden eggs? For the sake of the new GOST? For the sake of government procurement? For the sake of “it will be better than the old one,” although we found out the cost of the risk of shutting down the entire house due to an arc glitch and saw that it is not customary for Western manufacturers to set AFDD once at the input...

At the moment, Meander produces both the UZM-51M - good protection, licked by the entire forum (here is a copied post from Ksiman about it), and the UZM-51MD is produced - the same version with arc protection. There are rumors that this year, 2022, Meander will again cut off oxygen and force everyone to switch only to “MD”, removing “M” from sales. What's the point of doing this? I still don’t understand, and my letters to Meander remain unanswered - they were deeply offended by me for the harsh truth of life.

Example of UZDP - Meander UZM-51MD (©Designman)

I know only a few people who wanted to buy "MD" and I know many, many people who buy "M" in packs for their shields. That is, at the moment Meander’s scarlet flower © is not coming out, no matter how hard they try to promote it.

Let me explain once again briefly and directly, without verbal passages or games. I have no complaints personally about Meander and no conservatism towards the new. My complaint is that Meander leaves us without alternatives to choose when he is going to release only “MD”. I believe that if such a combined device is being made, then it should have a “turn off” or “give me current, even if everything is on fire” button. There is even such a button on some leakage protection systems and it’s called “Give me water.” This is exactly what I am emphasizing: make the arc protection switchable - and there will be no questions. You will receive a sea of ​​free testers who will help you and say a kind word. And people will have an alternative, slowly getting used to the new.

Our user Designman bought the first version of the UZM-51MD and began to torture it. Here are links to his community posts:

• The long-awaited test of the UZM-51MD. Part 1
• The long-awaited test of the UZM-51MD. Part 2 - Disassembly
• The long-awaited test of the UZM-51MD. Part 3 - Operation
• The long-awaited test of the UZM-51MD. Part 4 - VER.3 firmware

We observe what I wrote about: in trying to make one single device, we got glitches due to different sensitivity of the current sensor and attempts to catch a small arc from a spark both at a current of 1A and at a current of 63A.

And here are photos of the UZM-51MD from the inside (from his posts). Here you can see a funny, damn it, Kulibin inductive current sensor.

Arc sensor for Meander UZM-51MD (©Designman)

The large piece of iron is the phase contact of the voltage relay, which disconnects the load. The little green things next to it are inductors into which current is induced due to magnetic induction. And this current is analyzed by the microcontroller and a decision is made: whether it looks like an arc or not.

Arc sensor for Meander UZM-51MD (©Designman)

This sensor immediately had one cool problem. He fucking reacts to tips. And not only for interference from the internal phase conductor, but for all sorts of things. For example, from the switching power supply of the Meander VAR-M01-08 voltmeter-ammeter (thanks to Ksiman for the review). Here is this link - “The meander burns. Arc. UZM-51MD and VAR-M01-08 do not work on the same rail” - user gregory writes about this joke. What if there is an introductory machine next to the UZM-51MD? After all, the machine has an electromagnetic release, which is a coil of wire. Which also emits something into space, heh!

At the moment, Meander writes that he has debugged all the glitches with the MD, but it still reacts badly to welding. It’s so bad that somewhere on the forums Meander suggests shunting the MD head automatically during welding (!!). What kind of fuck-up is this?!

And - most importantly - this protection CANNOT be turned off for Meander. For example, in the same example with welding, I would cut off the arc protection for the duration of construction, while workers are wandering around with damaged tools. And then I would turn it back on. And it also has Automatic Recloser (ARC) for arc protection. How's that? Like it caught fire - turned it off - turned it on - burn out, house? That is, it is already directly readable that I am against all-in-one devices and do not want to use them.

...but Meander has the best voltage protection. Therefore, if they continue to forcefully impose only the “MD” version, then I will take them and bite (or short) this damn arc sensor and use them like regular “M”. Just like that.

Let's move on to the next craft - an anti-sparking device (note - not from an arc!) UZIS-S1, reviewed by Alexey Nadezhin (Ammo1) : Review: Anti-sparking device UZIS-S1-4. Next, I used the photographs that were posted on his LiveJournal via the link (I saved them to my blog):

Example of an AFDP - USIS (spark protection device) ©Ammo1

It is made by someone unknown to me, and the first thing that catches my eye is that this device has an OUTPUT FROM TOP and an INPUT FROM THE BOTTOM. And this is a kicker for all shields. You know, I am very tough in places, so from that moment on I immediately had a preconceived negative opinion about this device.

Included with it, Alexey Nadezhin was also given a test plug, which generates a signal into the network, which the device detects as an arc. This device also has protection against overvoltage. Well, this is understandable - the filling in all AFDDs is electronic, so it must be protected.

Example of an AFDP - USIS (spark protection device) ©Ammo1

And we are interested in photos of the insides of this AFDD, which the manufacturer himself posted in that LiveJournal. Here's the whole set. Let's try to look at it:

Ultrasonic arc protection device inside (manufacturer's photo)

Here is the electronics board. A varistor is visible to protect against voltage surges. True, it would be nice to put it in heat shrink so that it does not explode too much if damaged. Well, a large microcontroller is visible (its brand is not visible). The zero is not broken, but passes through the device in transit.

Arc protection device inside - electronics (manufacturer's photo)

And here's the power part. And this is where the first fuck-up comes in. WHERE THE FUCK IS THE ARC CHAMBER? What is the breaking capacity of this device? Has anyone tested it? But the device is designed for 40A nominal current! Even at UZMka Meander the breaking capacity is indicated at 4.5 kA. What's going on here?

Ultrasonic arc protection device inside - power part (manufacturer's photo)

We also see a current transformer, with which the current in the line is measured, and a large electromagnet, which should turn off the circuit breaker. A power contact is also visible in the lower left part of the photo. There is a solder on it - at least that's good.

Regarding this device, I had the following thoughts and questions:

• Still, what is its breaking capacity in kA at such and such operating current ratings? Where is the arc suppressor? It is not in any photo, and not a single hole is visible in the housing for the release of hot gases from the arc.
• Where is the “Test” button on the device itself? The test plug can be lost, but the button will allow you to test the device on the device itself. She must be.
• The device works as a release. Very good: once it works, it will turn off the line forever until a person comes to sort it out.
• Why input from below? After all, in our country it is unspoken that input is always done from above, but here this is violated.
• The power supply seems to have a quenching capacitor, if I'm not confusing anything. Has anyone tested these capacitors so as not to repeat Meander’s mistakes with them (when UZMs malfunctioned in 2013 due to the fact that capacitors from a bad batch lost capacity)?
• Why are “IN” and “OUT” not written in our language on the case? Did they steal the entire device from somewhere or steal the case (the die for casting the case costs hundreds of thousands and therefore it is always expensive to develop your own)?
• How well does such a current transformer work in the range from 0 to 40A? Is there saturation of the ferrite core? What tests were carried out?
• How resistant is this device with such a huge ferrite to external interference from a neighboring module? Maybe it will perceive the pickup from the coil of the linear contactor of the frequency switch as an arc?

The funny thing is that four days ago this manufacturer wrote to me and offered to send me samples of this ultrasonic device so that I could promote it or test it. For review, in general. I sent it through the woods, because again no one thinks about the methodology of use. Where should I put it? For input? No, that's not good! On separate lines? How can it be combined with RCDs and automatic machines? No thanks. I'd rather wait and choose ABB =)

Damn, it feels like now all the companies in a row will do AFDD, without understanding a damn thing about it. Wait, damn, a wave of AFDDs, similar to IEK, TDM and others. Because there are a lot of such cases for modules =)

Just don’t send them to me for tests and write about them - I’m NOT INTERESTED.

So, let's summarize. Here are some points that can be extracted from this post:

1. Arc protection devices were designed specifically to protect against ARC, not sparks. Normal AFDDs most likely should not be triggered by sparking forks or brushed motors. And, most likely, for welding.
2. AFDD, if used thoughtlessly, gives a FALSE sense of security! This is wrong! AFDD will NOT help if some connection takes a long time to heat up and sets fire to something around with its heat. AFDD will not detect this overheating as an arc. Everything will burn, and AFDD may not work.
3. In Western countries, AFDD is strongly recommended for installation due to the fact that they previously had more lenient wiring rules. And in most houses, the wiring lies inside wooden structures. Connections in America, for example, were most often made using PPE. This is where prolonged sparking of wiring can lead to fires. And this is exactly what Western AFDDs were developed for.
4. It is technically difficult to make an arc sensor that will have good protection against interference and the same sensitivity over a wide range of currents (for example, from 0 to 63). Therefore, Western devices are produced with small current ratings.
5. The small current rating of Western AFDDs makes it possible to install them on only one or two switchboard lines. And this automatically eliminates false alarms, which can be due to the summation of various interferences and current forms, if many long lines go after AFDD.
6. Putting AFDD at the entrance to the apartment is fraught with the loss of only the refrigerator. But installing AFDD at the input in a country house is fraught with loss (“defrosting”) of the boiler room or pipes if the AFDD malfunctions. Therefore, the “correct” AFDDs are released as separate devices so that developers can install them only on certain lines or groups of board lines.
7. Most often, Western AFDDs are specially combined with a line circuit breaker: this allows you to limit the current rating for the internal arc sensor, and immediately protect the line from overcurrents and “force” the developer to install an AFDD instead of a circuit breaker to protect only one line of the shield (to eliminate interference and interference ).

Why all our manufacturers decided that they would be able to deceive the system and make a cooler device on their knees - I don’t know. But now it turns out like in the story about the Germans and production automation (here, DI HALT had it) - about people's ingenuity. Well, there’s no need for resourceful ingenuity here, damn it! Need not! Here you need an understanding of where and why this device will be used. Before doing this, we would ask any shield workers. And the shield people would answer them that putting AFDD at the shield input is a risky lottery. And our AFDD market would have taken a completely different path.

In general, I decided the following for myself:

1. With competently made (and maintained) new and fresh wiring in the apartment , AFDD is still completely useless. Let me remind you: it should not paranoidly turn off when sparking, so it will not (and should not) go off with every sneeze.
2. a country house/cottage is properly done (and always maintained) (if it is a wooden house, then in steel pipes), AFDD is also not entirely necessary.
3. Setting AFDD at the input is fraught with a complete loss of life support at home. Therefore, all AFDDs that will be combined with voltage relays/RCDs and other crap like “just one thing will protect you entirely” can be immediately excluded from consideration. For me they simply will not exist.
4. If we decide to install AFDD in the shield, then it must be placed on one or more outgoing lines. Exactly the same way we treat RCDs.
5. I will focus on AFDD from ABB, because they are immediately combined with the line circuit breaker and have the same clamps as the rest of the module. They are expensive, so they will only need to be installed where it is really needed. And this is good - let it make people think and not consider AFDD a new panacea for “everything.”
6. If Meander again removes the UZM-51M from production forever, then I will take the UZM-51MD, open them and turn off the arc sensor there (bite it off or short it out - I’ll conduct experiments).

That's all I have!

▍ From one extreme to another

The opposite problem is loss of sensitivity on long lines. Any piece of conductor has its own inductance and distributed capacitance. If we have a long line, then this is how it will be different: The long line itself begins to work as a surge filter, and the high-frequency part of the spectrum is attenuated more strongly the longer the line is. Therefore, there is a certain maximum range at which the AFPD is capable of detecting an arc fault. Only one manufacturer of an AFDP includes a simulator that allows you not only to check the serviceability of the AFDD, but also to determine whether it has lost sensitivity due to a long line. Therefore, the AFDD may not work due to sparking in the security booth, from which there are a couple of hundred meters of cable to the shield with protection devices. As a rule, there are no problems on lines shorter than 100 m.

The most popular ways to protect the route

Corrugated pipe (corrugated pipe, corrugated pipe)

It is a flexible plastic thin-walled tube with a cross-section from 16 to 63 mm. The most universal and common method of cable protection. It is supplied in coils of 5–500 m and has a broaching wire embedded inside. To place a wire or group of wires in a corrugation, just secure the wire at one end of the wire and pull it out of the corrugation - the wire will stretch behind it.

The cable is corrugated before installation. It is suitable for any type of wiring, but practice has shown that it is most convenient to use it with flexible wires - low-current, twisted up to 2.5 mm2, telephone, monolithic up to 1.5 mm2.

Advantages:

• Easy to install.
• Easy to cut.
• Flexibility.
• The most affordable price.

Flaws*:

Least protection. Fragility (tightness may be broken). Flammability. It is difficult to push the wire through it - the ribs get in the way.

* Applies to low and medium price categories. Expensive corrugation is quite durable, non-flammable, and there are also frost-resistant models.

In general, corrugation satisfies all the needs for protecting household wiring - it can be concreted, laid in frame walls, or run outside instead of a cable duct.

Table. Approximate prices for Russian-made thin-walled corrugation

Plastic pipe

This is an ordinary smooth-walled technical pipe made of polyethylene. Wall thickness from 1 to 3.5 mm, diameter from 16 to 50 mm. Wire range: twisted - any section and shape, monolithic - up to 20 mm2.

Advantages:

Creates a reliable channel that is difficult to damage unnoticed even with a drill. Durable walls prevent leakage or accidental damage. It is convenient to push the wire along smooth walls.

Flaws:

Inflexible, because of this - a narrow scope, often technical objects. When folded, a crease may occur, which will complicate further work. Flammable.

Any technical pipe can be used as a channel - water supply, sewer or drainage. However, laying it in the groove can be too labor-intensive. If the diameter is too large, break the group of wires into two thinner pieces.

Scope of application - any objects that are suitable for the technical properties of the pipe and do not have a fire, explosion or thermal hazard.

Table. Prices for plastic pipe

▍ Why only now?

If fuses have been known for more than a hundred years, circuit breakers for about the same, RCDs for fifty years, then AFDDs appeared quite recently - already at the end of the 20th century.
And all because without electronics it is impossible to detect an arc breakdown. And relatively cheap microcontrollers on which digital signal processing can be implemented have appeared quite recently. So it turns out that only now has it become possible not only to technically implement this type of protection, but also to do it at a price affordable to individuals. Legislation is also actively changing - the new device is being introduced into various rules and regulations, making it mandatory for use in certain tasks. I don’t want to refer to various regulations (then I’ll be tired of running around and making changes with the next change), but in our country, AFDDs have begun to be legalized with GOST IEC 62606-2016, which is a translation of the IEC standard. Actually, the standard not only defines the required characteristics of the AFDD and the testing methodology, but finally determined the very name of this type of device - the AFDD.

Types and types of AFDP

With all this, GOST does not determine exactly how to do this. Each manufacturer solves the problem in its own way and issues appropriate patents.

• Meander UZM 51MD

• AFDD Eaton

• UZIS-S1-40 Ecolight

• Siemens 5SM AFD

• ABB S-ARC1

• Hager

Only when all factors are combined does the protective device determine that an arc has appeared in the circuit and turns it off.

If the pulses in the network are less than the specified amplitude, then this is considered not dangerous and the device does not respond.

Unlike the voltage relays we are used to, there are no manual settings on such arc-protective circuit breakers.

In voltage switches, you can adjust the triggering both at the upper and lower limits. Here, all parameters are set at the manufacturer’s factory.

Of course, the very first such specimens still have errors and false positives. The technology cannot be called fully developed.

However, most of the blunders have already been eliminated. For example, an ordinary vacuum cleaner, blender or drill, when turned on, can generate a certain wave characteristic similar to an arc. An arc also occurs when the stove is electrically ignited.

Any brushed power tool sparks, especially if its brushes are already worn out enough. Not to mention the initial inrush current.

Manufacturers take all these operating points into account and there are fewer and fewer false alarms in high-quality models.

How quickly should these arc flash detection devices respond? Everything here depends on the voltage and rating of the arc current.

According to the requirement of the IEC 62606 standard, at a current of 10A, the response time should not exceed 0.25 seconds.

Here is a table of all values:

▍ Where to include?

The AFDD is not an independent device - it usually requires a separate circuit breaker.
Manufacturers, in pursuit of our wallets and compactness, can combine an AFDD with a circuit breaker - such a hybrid can already be used independently. When using several types of protection devices, the connection sequence does not matter. You can install an AFDP both before and after the RCD. Please note that in some AFDP models the input is made from the BOTTOM, and this is not a foolishness of the developers, and is also found in famous Western manufacturers. I'm sure the designers tried to the last to do as everyone was used to, but something got in the way.

Typical circuit diagram for switching on an AFPD:

Considering the non-zero probability of false alarms, it makes sense to use several AFDDs and divide the lines by type of load - conditionally stationary and variable. Include stationary consumers whose current consumption profile has not changed for years - circulation pumps, refrigerators, ventilation, etc. A sudden activation of the AFDD on such a group will most likely signal a real problem. Variables should include all sockets into which anything is constantly plugged - blenders, kettles, vacuum cleaners, lighting, etc. The activation of an AFPD on this line should be alarming, but it is much easier to connect it with a new device on the network.

In an ideal world, of course, each line would have its own automatic machine and AFDD, but given the prices and average salaries, this is a dream. But one AFDD for an entire private house can create many problems, how can you look for the location of the problem if it is triggered? Therefore, at least some division into groups should be provided.

Special caution is required when using AFDDs on lines with important loads, the disconnection of which can cause problems no less than an arc breakdown. Circulation pumps, network switches, etc. Moreover, the standards explicitly prohibit the use of ultrasonic devices for some consumers - for example, with ventilators.

Connection diagram

To begin with, here is a diagram of the AFDP itself:

Schematic diagram of the AFDD

Pay attention again - both in the diagram and in the real device, INPUT is from below, OUTPUT is from above!

The connection diagram for our device is simple, the main thing is to protect it automatically!

Connection diagram of the IEC AFZ through a circuit breaker

RCDs and difavtomats are connected according to the usual rules. As a result, the connection diagram will be like this:

Inclusion of an AFDD in a circuit together with a circuit breaker and an RCD

Instead of a combination of AB + RCD, you can use Difavtomat. Both options have their pros and cons, but that’s not about that now.

It is better to install an AFPD closer to consumers, in the apartment panel. This will ensure maximum sensitivity. At the same time, you need to understand that the AFDP will not detect any problems with the electrical wiring BEFORE the installation site.

▍ It sparks at my neighbor, but turns off at mine

Unfortunately, this is possible with low-quality AFDDs. Although the AFDD analyzes the load current, it would seem that it should be blind to everything that happens before it. But power lines are not an ideal source of current and have internal resistance. Therefore, on long lines, the sparking of a powerful load will cause noticeable fluctuations in the supply voltage, which in turn will cause fluctuations in current consumption (very significant if the load is nonlinear). This is called crosstalk. Developers are taking measures and using various techniques to reduce sensitivity to crosstalk with varying degrees of success.

How AFDD+ works

Eaton's AFDD+ solution uses digital technology with built-in signal processing and intelligent evaluation to provide fault current detection. Thus, it is capable of digitally monitoring electrical wiring for specific frequencies that may indicate arcing. The device indicators display status and detailed information when the AFDD+ is turned off due to a fault (Figure 2).

Rice. 2. Status indicator

Communications over power lines can produce intense signals that "mask" the noise generated by the arcs. The AFDD+ solution has been designed with this feature in mind, meaning it will not impact the device's arc detection capabilities.

LLC "Eton" 107076, Moscow, st. Elektrozavodskaya, 33 building 4 +7 +7 [email protected] www.eaton.ru

▍ It worked - now what?

Probably the most interesting question.
I am sure that when the protection is triggered, most will simply go and turn everything back on without trying to figure out the reasons. But we’re not like that?) If the AFDD worked, then there was a reason, and it is advisable to try to find it. The task is simplified if, when turned on, the AFDD turns off again, then the problem is stable - using circuit breakers (now you understand that the more developed the division into groups, the easier it is to look for the problem?) We turn on the groups sequentially. If, when connecting the next group, for example, “garage”, the AFDD is triggered again, we begin to look for the problem there. Troubleshooting can be tedious, but in general it is no different from finding the reasons for the operation of any other protection device, for example, an RCD.

If, when you turn on the AFDD, it does not turn off again, it is enough to carry out a preventive inspection to see if all the sockets are intact, if there are any melting or darkening on the plastic. You can turn the voltage back on and listen carefully - poor contact can sometimes be heard through a characteristic “squeaking” sound. Inspect flexible cords and carriers for damage. If the power cord is bent at the fastening points, nothing should change.

Now it is obvious: the more developed the division into consumer groups, the less work there is to localize the problem. It’s one thing to inspect ALL the electrics of the house, since the AFDD is one thing, and another thing is to inspect the children’s room if the AFDD has been triggered in the children’s room.

▍ More functions, and for free

If the AFDD incorporates fairly advanced electronic “brains” to perform the main function, then why not add more functions with minimal hardware changes? Almost all AFDDs in my test have an overvoltage protection function - if the voltage in the network increases above the standard value, for example, due to a fallen “zero”, not 230V but all 400V has arrived, then the AFDD will also turn off normally. Alas, when the voltage returns to normal, it will not turn back on due to the free release mechanism. Thus, the use of some AFDD models allows you to get additional protection against zero loss practically for nothing. (Reservations: there is no automatic re-switching - when the voltage returns to normal, nothing will turn on automatically. Many SPD models also do not have undervoltage protection.)

▍ It also tests itself?!

Yes, if you take a closer look at the readings of the indicators on the facade of the AFDD, you will see the option “AFD is faulty.” The device contains additional circuits that allow it to send a reference signal to its input and make sure that the signal is perceived as expected. In this case, the serviceability of the analog part of the device is checked, but it is not checked, for example, the serviceability of the release mechanism (this would lead to self-shutdown, which is unforgivable). Those. The AFZ is capable of independently identifying certain types of faults and notifying about its non-functionality. When the user regularly checks the RCD (remember the “test” button on the RCD?) he will notice a problem and replace the RCD.

Surge suppressor parameters

Before you go to the store and buy this device, you need to know the following:

The number of modules (terminals) depends on the type of your network. 1 module can be purchased when there is a single-phase TN-C system. 3 modules when the installation is in a three-phase TN-C network and 4 modules when the network is three-phase in TN-S or TT. Class (type) - you can choose between classes B, C or B + C. If you are not sure that a type B limiter is used in front of your apartment, you should choose a B + C solution. Otherwise, a type C limiter will be sufficient. The rated voltage at which the limiter operates. Uc is the operating voltage of the protector, that is, the maximum voltage level that will lead to operation. In is the rated current of the limiter, that is, what current can flow through the arrester in the event of a short circuit. Imax is the current that the arrester is capable of accepting during an atmospheric discharge

Please note that both values ​​(In = 30,000A and Imax = 60,000A) will be relatively large in relation to the current during normal operation of appliances in the house. Up - the voltage to which it decreases in the event of a rupture. For example, if the potential reaches a voltage of 10,000 V in the event of a surge, the final value drops to 150.

▍Criticism

For the sake of objectivity, it is worth saying that the widespread use of AFDCs also has its critics.
The most significant argument is that the role of arc breakdown as the root cause of a fire is ambiguous; when conductors are heated from overcurrent, arc breakdown is formed in the late stages of melting of the conductor, when the insulation from heating is already smoking and draining. And in this case, the activation of the AFDD may not prevent the fire. And the open question is what is the cause of the fire - a fire from an overload (which should be prevented by circuit breakers and fuses), or an arc breakdown. Here I will leave a link to the noteworthy channel of test engineer Vladimir Semenovich Melnikov, as a critic of the UZDP (https://www.youtube.com/channel/UCCem6jemMX_3ce6dDKk3gdw), in particular, this video (https://www.youtube.com /watch?v=fsy20dMmp-w). My opinion is illustrated by the phrase “If you try to automate a mess, you get an automated mess” - if the electrical equipment has been brought to a state where the wires fall out of the terminals, then the AFDD will not become a panacea (although it will probably constantly trip and irritate electricians, and may force them to find problem areas ). Although many safety measures that are already familiar to us, such as seat belts in cars, were also implemented with difficulty and found their critics, who very convincingly spoke out about the uselessness and redundancy of such measures