Fibers are loaded into the welder
Hello Habra readers! Everyone has heard about optical fibers and cables. There is no need to tell where and for what optical is used. Many of you come across it at work, some develops backbone networks , someone works with optical multiplexers However, I did not meet the story about the optical cables, couplings, crosses, about the very technology of splicing optical fibers and cables. I am an optical fiber splicer and in this (my first) post I would like to tell and show you how it all happens, and I will often in my story be distracted by other related things. I’ll be relying mostly on my own experience, so I’m quite willing to let someone say "this isn’t quite right" or "this isn’t really canonical".
There is a lot of material, so it was necessary to break the topic into parts.
In this first part you will read about cable construction and cutting, about optical tools, about preparing fibers for welding. In other parts, if the topic is interesting for you, I will tell about methods and show on video the process of splicing optical fibers, about basics and some nuances of measurements on optics, I will touch upon welding machines and OTDRs and other measuring instruments, show splicer workplaces (roofs, basements, attics, hatches and other fields with offices), tell a little about fixing cables, about soldering schemes, about placing equipment in telecommunications racks and crates. This will probably come in handy for those who are considering becoming a splicer. I added lots of pictures (sorry in advance for the paint-quality) and pictures.
Careful, lots of pictures and text.
Part 2 here
First, a few words about me and my work.
I work as an optics splicer. I started out as a telephonist and installer, then I worked in an emergency crew for mainline fiber optics. Now I work in the organization, which takes general contracting for the construction of objects and lines of communication from various companies. A typical construction project is a cable line connecting several GSM base station containers. Or several FTTB rings, for example. Or something smaller, such as running a cable between two server rooms on different floors of a building and welding cross-connections at the ends of the cable.
If the tender is won, we look for suitable subcontractors to perform the work (design and survey and construction and installation). In some regions these are our subsidiaries, in some regions we have our own equipment and resources, in some regions independent companies are hired. We are mostly responsible for the control, elimination of subcontractors’ faults and various force majeure factors, coordination of all kinds with land owners and administrations, sometimes preparation of executive documentation for the constructed facility (documentation – mainly RD 45.156-2000, see here). here there is a list, plus there is also added a section with different licenses) and other things. Often you need to work with optics: to weld or reweld somewhere optical coupling or cross, to eliminate consequences knocked down by streetcarriers supports or fallen on the cable tree, to conduct the incoming inspection cable drum, take traces and so on. These are the tasks I perform. And incidentally, when there is no task on the optics – other tasks: from the loading and installation through the courier-delivery to the copying and paperwork. 🙂
Optical cable, its types and insides
So, what is an optical cable? Cables come in a variety of forms.
By design – from the simplest (sheath, under it plastic tube-modules, in them the fibers themselves) to the most sophisticated (a lot of layers, two-level armor – for example, in the underwater transoceanic cables).
By location of use – for outdoor and indoor installation (the latter is rare and usually found in high-end data centers where everything has to be perfectly right and beautiful). According to laying conditions – for suspension (with Kevlar or cable), for ground (with iron wire armor), for laying in the cable duct (with armor of corrugated metal), underwater (complex, super protective multi-layer construction), for suspension on transmission towers (besides information transmission, they perform a role of lightning protection cable). In my practice I mostly find cables for suspension on poles (with Kevlar) and for laying in the ground (with armor). Less common are cables with cable and corrugated armor. Still often there is a cable, which is basically a thin paired optical patch-cord (yellow sheath for single-mode and orange – for multimode, a bit of kevlar and one fiber; two sheaths are paired). Other optical cables (without protection, underwater, for laying indoors) are exotic. Almost all of the cables I work with are of the design shown in the picture below.
1 – central power element (simply put, a rod of fiberglass, although it can also be a cable in a polyethylene sheath). Serves to center the tube-modules, giving stiffness to the whole cable. It is also often used to fasten the cable in the socket/crossover by clamping it under the screw. If the cable is bent too much, it has a sneaky tendency to break, breaking the modules with some of the fibers in the process. More advanced cable designs have this rod encased in polyethylene sheathing: then it is harder to break and it causes less damage in the cable when it breaks. The rod can be as shown in the picture or even thinner. The tip of the rod is an excellent abrasive tool for delicate jobs like cleaning relay contacts or soldering copper parts. If you burn it a couple of centimeters, you get a nice soft brush. 🙂
2 – themselves optical fibers (pictured in lacquer insulation). The very thin filaments and light conductors, for which it is all about. This article will only talk about glass fibers, although somewhere in nature there are plastic ones, but they are very exotic, they are not welded by optical welding machines (only mechanical connection) and are suitable only for very short distances and I personally have not encountered them. Optical fibers come in single-mode and multi-mode, I have only encountered single-mode, as multi-mode is a less common technology, can only be used for short distances and in many cases is perfectly replaced by single-mode. The fiber consists of a glass "shell" of glass with certain impurities (I will not dwell on the chemistry and crystallography, as I do not know the subject). Without lacquer, the fiber has a thickness of 125 microns (slightly thicker than a hair), and in the center of it there is a core with a diameter of 9 microns made of ultrapure glass with a different composition and with a slightly different refractive index than the shell. It is in the core that radiation spreads (due to the effect of total reflection at the "core-shell" boundary). Finally, the top of the 125-micrometer cylinder of the "shell" is covered with another shell – made of a special varnish (transparent or colored – for color marking of the fibers), which is also two-layer, EMNIP. It protects the fiber from moderate damage (without lacquer, though the fiber is bendable, but bad and easy to break, the fiber will crumble elementary from accidentally put a cell phone on it; but in lacquer it can be easily wrapped around a pencil and yank quite hard – it will hold). It happens, that the cable span is sagging on only fibers: all the sheaths are torn (burnt, cut), kevlar, the central rod is broken, and some 16 or 32 125-micrometer glass fibers can hold the weight of the cable span and wind loads for weeks! Nevertheless, even in lacquer, fibers can easily be damaged, so the most important thing in a splicer’s job is meticulousness and accuracy. One awkward move can ruin the results of a whole day of work or, if you are particularly unlucky and there is no redundancy, you can drop the trunk for a long time (if, while digging in the "combat" trunk, you break a fiber with DWDM under the stub at the cable outlet).
Fibers come in many varieties : regular (SMF or just SM), biased dispersion (DSF or just DS), non-zero biased dispersion (NZDSF, NZDS or NZ). Externally it is impossible to distinguish them, the difference is in chemical/crystalline composition and, probably, in the geometry of the central core and in the smoothness of the boundary between it and the shell (unfortunately, I have not completely clarified this question for myself). Dispersion in optical fibers is a severe and complicated thing, worthy of a separate article, so I will explain simply – you can transmit a signal without distortion on fibers with biased dispersion, than on simple ones. In practice, splicers know two types: simple and "biased". In a cable, the first module is often allocated for "offset" and the rest for simple fibers. It is possible, but not desirable, to splice "offset" and simple fibers together. This causes one interesting effect, which I will tell you about in another part, about measurements.
3 – plastic modulus tubes , in which are floating in hydrophobic fibers.
Cable stripped down to modules
They easily break (or rather, suddenly bend) when bending like telescopic antennas of household receivers, breaking the fibers inside them. Sometimes there is only one module (in the form of a thick tube), and in it a bundle of fibers, but in this case you need too many different colors to mark the fibers, so usually make several modules, each of which has from 4 to 12 fibers. There is no single standard for color and number of modules/fibers, each manufacturer does it their own way, displaying everything in the cable data sheet. The passport is attached to the cable drum and is usually stapled to the wood right inside the drum.
Cable data sheet
Typical cable data sheet. Apologies for the quality.
However, there is hope that, say, the cable "DPS" from manufacturers "Transvok" and "Beltelekabel" will still be the same configuration. But you still need to look at the cable data sheet, which always shows a detailed coloring and what type of fibers are in which modules. The minimum capacity of an "adult" cable I’ve seen is 8 fibers, and the maximum is 96. Usually 32, 48, 64. Sometimes 1 or 2 modules of the cable are occupied, and then instead of the other modules they put black blank plugs (to keep the dimensions of the cable unchanged).
4 – film , which braids the modules. It plays secondary roles – damping, reducing friction inside the cable, additional protection against moisture, retaining the hydrophobe in the space between the modules and probably something else. Often it is additionally tied with threads crosswise and moistened on both sides with hydrophobic gel.
5 – thin inner sheath of polyethylene. Extra protection against moisture, protective layer between kevlar/armor and modules. May be missing.
6 – Kevlar threads or armor In the figure, the armor is made of rectangular rods, but much more often it is made of round wires (in imported cables, the wires are steel and hard to cut even with wire cutters, in domestic ones, they are usually made of nail iron). Armor can also be in the form of fiberglass bars, the same as the central element, but in practice I have not come across such. Kevlar is needed so that the cable can withstand a lot of tensile force without being heavy. It is also often used instead of cable where the cable should not contain metal to avoid surges (for example, if the cable hangs along the railroad, where the contact wire with 27.5 kV is nearby). Typical values of allowable tensile force for a cable with Kevlar are 6…9 kilonewtons, it allows to withstand a large span under wind load. Kevlar is terribly blunt with cutting tools when cutting. 🙂 That’s why it’s better to cut it either with special scissors with ceramic blades, or to bite it off with a wire cutter, which is what I do.
As for the armor – it’s meant to protect the underground cable lying directly in the ground, without protection in the form of plastic pipe, cable ducting, etc. However, the armor can only protect from the shovel, the excavator will tear any cable in no time anyway. That’s why underground cables are put into the ground for 1m 20 cm, and above it at a depth of 60 cm yellow or orange signal tape with "Beware! Do not dig! Below the cable", as well as putting bollards, warning signs and slogans along the route. But still they dig and tear.
7 is external thick polyethylene sheath It is the first to bear all the burdens of installing and operating the cable. Polyethylene is soft, so it is not difficult to cut when the cable is tightened carelessly. It happens that when laying underground cable contractor tears up to the armor this sheathing for a few meters and did not notice, in the ground in the cable enters the moisture despite the hydrophobic, and then at the delivery, when testing the outer sheathing megohmmeter megohmmeter shows low resistance (high leakage current).
If the dangling cable touches a concrete pole or tree, the polyethylene can also quickly rub through to the fibers.
A polyethylene film and some hydrophobic gel may be present between the outer sheath and the armor.
In Russia, unfortunately, optical fibers are no longer produced (here, alas, a joke about polymers would be appropriate). There is a Russian laboratory that makes experimental fibers for special purposes, as prompted by esvaf
They buy them from companies like Corning, OFS, Sumitomo, Fujikura, etc. But they make cables in Russia and Belarus! Moreover, in my practice, 95% of the cables I have worked with are cables from Russia or Belorussia. At the same time in the cable is laid with imported fiber. From my experience, I can think of such cable manufacturers as Beltelekabel, MosKabel Fujikura (MKF), Eurokabel, Transvok, Integra-kabel, OFS Svyazstroy-1, Saransk-kabel, Inkab. There are others. Of the imported cables in memory remains only Siemens. Subjective all cables are similar in design and materials and quality are not particularly different.
So, I told you about the construction of the optical cables. Let’s move on.
Cable cutting : necessary tools and technique
Cable cutting, like welding, requires a number of specific tools. A typical splicer’s kit is the NIM-25 tool case, which contains all the necessary strippers, wire cutters, screwdrivers, side cutters, pliers, layout knife and other tools, as well as a pump or vial of alcohol, a supply of D-Gel hydrophobe solvent, non-woven lint-free wipes, duct tape, self-adhesive cable and module markers and other consumables.
Once you complete it with consumables (couplers, worm clamps, etc) and some auxiliary tools it is quite sufficient for working with optics. There are also other kits, richer and poorer ("NIM-E" and "NIM-K"). The weak point of most kits is the poor quality "aluminum type" case, which only looks nice, but actually consists of thin fiberboard covered with textured/corrugated foil and aluminum thin corners with rivets. It doesn’t hold up well in the field or in the city, and has to be repaired and reinforced. In my case the case survived for 3 years and, being all torn, stiffened with corners and bolts, with a "kolhozny" organizer instead of the native one, it was replaced by a regular plastic toolbox. Some of the tools and materials in the standard set may be of low quality. Some of the tools I personally don’t need. Some were already replaced after 3 years of work. As the "branded" consumables are used up, some are replaced with "handy" ones without affecting the quality of work. For example, the factory non-woven lint-free wipes for wiping fibers are easily replaced with "yawns plus" type toilet paper. 🙂 The main thing is that it should be unscented. Instead of expensive (about 800 p/liter) D-Gel, if you work outdoors, you can use gasoline AI-92.
When cutting cables it is important to keep the lengths of the cable elements according to the instructions of the coupling: in one case it may be necessary to leave a long power element to fix it in the coupling/crossover, in another case it is not necessary; in one case the Kevlar cable is braided and clamped under the screw, in another case the Kevlar is cut off. It all depends on the particular coupling and the particular cable.
Consider cutting the most typical cable :
a) Before cutting a cable that has been in damp or without a waterproof end for a long time, you should cut off about a meter of cable with a hacksaw (if the stock allows), because prolonged exposure to moisture negatively affects the optical fiber (can cloud) and other elements of the cable. Kevlar strands in the cable are an excellent capillary that can "pump" water into itself for dozens of meters, which is fraught with consequences if for example there are high voltage wires in parallel with the cable: currents can start to walk on wet Kevlar, water evaporates, crushes the outer sheath from inside, the cable gets blistered and new moisture gets through the bubbles from rains.
b) If there is a separate cable for suspension (when the cable in the cross-section has the shape of "8", where the bottom part is a cable, the top part a cable) it is bitten out with a wire cutter and cut with a knife. When cutting the cable it is important not to damage the cable.
c) To remove the outer sheath of the cable, an appropriate shear knife is used. The NIM-25 is usually supplied with a "Kabifix" knife as shown in the photo below, but you can also use an electric cable stripper with a long handle.
The cable cutter has an all-round rotating blade which can be adjusted in length according to the thickness of the cable sheath and a clamping element to hold it on the cable. Important: if you cut different brands of cables you have to try the blade on the tip before cutting the new cable and if it cuts too deep and damages the modules the blade has to be adjusted shorter. The worst thing is when the socket is already welded and suddenly when laying the fibers one fiber suddenly "pops" out of the cable because the knife caught the module during cutting and broke that fiber: all the work is in vain.
With the cable stripper a circular cut is made on the cable to remove the outer sheath, and then from there two parallel cuts from opposite sides of the cable toward the end of the cable so that the outer sheath splits into two halves.
It is important to set the length of the blade correctly, because if the blade is too short, the outer sheath will not split easily into two halves and it will take a long time to peel it off with pliers, and if the blade is too long, you can damage the modules deep in the cable or blunt the rotating blade on the armor.
d) If the cable is self-supporting with Kevlar, the Kevlar is cut with cable cutters or scissors with special ceramic blades.
Kevlar should not be cut with a knife or simple scissors without ceramic pads on the blades, as Kevlar quickly blunts metal cutting tools. Depending on the design of the coupling it may be necessary to leave a certain length of Kevlar for fixing, this will be mentioned in the installation instructions of the coupling.
If the cable is designed for laying in the telephone sewer and from the armor contains only metal crimp (so that rats can not chew), it can be cut lengthwise with a special tool (plow knife) or you can carefully make a small pipe cutter or even a regular knife on the crimp circular risk and wiggle to get the metal fatigue in the place of risk and cracking, then you can remove some of the crimp, bite on the modules and tighten the crimp. Such cutting should be done with special care, because it is easy to damage the modules and fibers: the corrugation is not very strong, it can crush in the place where you poke it with tools, and while pulling the fibers the sharp edges in the place of bending can puncture the modules and damage the fibers. Corrugated cable is not the most convenient to cut.
If the cable is armored with round wires, they should be bitten off with wire cutters in small batches, 2-4 wires at a time. It takes longer and is harder with wire cutters, especially if the wire is steel. Some couplings require a certain length of armor to secure, and armor (including corrugated armor) often needs to be grounded.
e) For the inner, thinner sheath present in some cables (e.g. self-supporting cables with Kevlar), you should use a separate, pre-configured blade stripper (can be the same as for stripping the outer cable sheath) in order not to mess up the blade length setting each time you cut the cable. In this case it is especially important to set the blade length in the stripper correctly, it will be less than in the stripper for stripping the cable outer sheath, because the inner sheath is significantly thinner, and right under it there are modules with fibers. With some skill, you can use an ordinary mock knife to remove the inner sheath, making a longitudinal cut with it, but there is a significant risk of damaging the modules. You can also use a clothespin stripper to cut the coax.
f) Threads, plastic film and other accessories are removed from the modules with wipes and D-Gel/gasoline. Threads can be twisted one at a time, or can be peeled off with a special sharp "plow" hook (may be included in the design of some sheath strippers). D-Gel solvent (colorless oily liquid, smells like orange, toxic) or gasoline can be used to remove the hydrophobe. However, be careful with gasoline: the employees of the office, where the gasoline is pouring, will not be happy with the fragrance. It is also a fire hazard.
You have to wear disposable gloves (surgical, polyethylene or construction gloves) because hydrophobe is a very unpleasant substance (the most unpleasant thing about a splicer!), it is hard to wash, after petrol or hydrophobe your hands are greasy for some time, and after cutting cable you have to weld fibers which requires clean hands and a clean workplace. In winter, hands soaked in hydrophobe get very cold. However, if you get the hang of it, you can cut cables without getting your hands dirty.
After removing the strands and separating the harness into individual modules, each module is wiped clean with wipes or a rag with D-Gel/gasoline solvent and then with alcohol until clean. Although, to save time and to get less dirty, you can do the following way – firstly cut the cable to the modules not all the way through, but in the place where it starts, about 30 centimeters, without wiping anything, bite the modules (see point "e") and pull the whole harness of modules with winding and threads, holding the clean cable end with your hand as a handle. Your hands stay almost clean and time is saved. However, with this method of cutting, there is a risk of tearing some of the fibers or applying excessive tensile force to the fibers, which will have a negative effect on future fiber attenuation, as well as a greater chance of damaging the modules, so this method is not recommended, especially in winter, when the hydrophobic filler gets thicker. First you have to learn how to do it right, and then you can try different optimizations.
g) At the required length, each module (except for pacifier modules, they are bitten out to the root, but first make sure that they really have no fibers) is bitten with a module stripper (suitable for copper coax as well), after which the module can be effortlessly pulled from the fibers.
The notching of the modules with the stripper is a very responsible moment. You have to choose the exact diameter of the notch, because if it is too big, the module will not get a good enough bite to get out, if it is too small, you risk cutting the fibers out of the module. Besides, you have to watch carefully for the stripper catch: if at the moment of biting the module it blocks the stripper backwards and fixes it in "closed" state, then in order to unbite the stripper and release the catch, you will have to close the tool on the biting module again, and there is a great risk to bite the module, which will make you have to cut the cable all over again. Keep in mind that when you bite one of the modules, the other modules are in the way, and the cable itself has to be held in the balance somehow. Therefore, it will be very uncomfortable at first, and the cable should be cut by two people.
There are cable constructions where the module is the only one and it looks like a rigid plastic tube in the center of the cable. To remove such a module properly, it should be cut in a circle with a small pipe cutter (not included in the NIM-25) and then gently break at the circumferential groove.
When tightening the modules, make sure that all fibers are intact and no fibers are left sticking out of the tightened module.
If the temperature is low, the modules are thin, there is little hydrophobic (=grease) in the modules by design, or the length of the modules to be removed is significant – the module may not pull off the fibers without effort. In this case, do not pull hard, because stretching can affect the attenuation of the fibers in this place, even if the fibers are not torn. You should bite down and remove the module in two or three batches, piece by piece and slowly.
When cutting the cable, pay attention to the length of the fibers. It should not be less than specified in the manual, usually it is 1.5-2 meters. Basically it is possible to cut the cable to 15 cm and then even weld it somehow, but then you will have a big problem when putting the fibers into the cassette: a big fiber reserve is needed so that you can "play" with the length and put all the fibers into the cassette in a beautiful way.
Sometimes it is necessary to weld into the transit cable without cutting it. In this case it is cut in the same way as the usual one, but the requirements to the care of cutting are stricter: because the cable can already be used for communication. It is cut to modules and modules are carefully introduced into the "oval" input sleeve (in the usual round will not go – will break), for this input, using a special set of heat shrink and metal clips with a block of hot melt. This glue melts when it shrinks from the high temperature and fills the space between the two cables, providing a seal. Then the module that needs to be welded in is cut, those fibers from it that do not need to be unsoldered are welded back in transit, and the ones we need are welded to the "unsoldered" (branching) cable. Very rarely we may have a situation where we need to take a fiber from a module, but we cannot cut the module (it carries an important link). In that case the following is used module slitting kit : the module is "chamfered" lengthwise, the fibers are extracted from it, wiped of the hydrophobic and sorted. The ones we need are cut and welded to another cable according to the scheme, and the rest are simply placed in the cassette. In this case, if you start a non-cut cable, the length of the fibers should be twice as long (2-3 m), this is understandable.
The fibers must be clean (thoroughly wiped of the hydrophobe), and special care must be taken to ensure that all fibers are intact. Fibers must be handled with care, because if the cables are cut and terminated, welding is almost complete, and a fiber breaks at the cable exit, it will be necessary to cut and weld the cable again, which is time consuming and highly undesirable and unprofitable for rapid restoration of communication on an existing trunk line.
Optical fibers damaged as a result of careless cable stripping (the length of the stripper blade was set incorrectly to remove the cable inner sheath, resulting in cutting through modules and damaging some of the fibers)
g) The fibers should be wiped down well with lint-free alcohol wipes to remove the hydrophobic filler completely. The fibers are wiped first with a dry cloth, then with cloths moistened in isopropyl or ethyl alcohol. This is the order because the first wipe leaves a huge drop of hydrophobe (alcohol is not needed here), but on the 4th-5th wipe, you can call for alcohol to dissolve the hydrophobe residue. Alcohol evaporates quickly from the fibers.
Always clean up the used wipes (as well as cable sheath scraps, chipped fibers, and other debris) – spare the environment!
The cleanliness of the fibers, especially near the ends, is important for quality welding. Where microns are being worked on, dirt and dust are unacceptable. The fibers should be inspected for the integrity of the lacquer coating, absence of dirt, broken parts of fibers. If the varnish on some fibers is damaged, but not yet broken – it is better not to risk and redo the cable. Spend 10-15 minutes, otherwise you risk wasting the whole day.
h) Special adhesive heat shrink sleeves are put on the cut cables, which are often included in the sleeve kit (if the sleeve has a cable entry spigot). If the socket provides for clamping the cable in raw rubber with sealant, then a heat shrink is not necessary. A very common and very unpleasant rookie mistake is to forget to put on the heat shrink! When the socket is welded, the heat shrink is pushed over the socket joint and shrunk with a gas torch, blowtorch or industrial hairdryer to provide a tight seal for the cable in the socket and extra fixation of the cable. The most practical is to use a small torch put on a cylinder of travel gas with a price clamp: one cylinder is enough for dozens of welded sockets, it simply ignites unlike the blowtorch, it weighs little, there is no dependence on electricity unlike the industrial hair dryer.
Before shrinking you have to sand the socket and cable with a coarse sandpaper for better adhesion of the glue. If you don’t do that, you might end up with this kind of misunderstanding:
If you have forgotten to use a heat shrink sleeve, a heat shrink sleeve with a lock (known as XAGA) will help. Do not use duct tape to seal it!
Some heat shrinks (e.g. by Raychem) are covered with dots of green paint that turn black when heated, indicating that this place doesn’t need to be heated anymore and that this should be heated some more. This is done because the heat shrink can break if it is overheated in some place.
It is better to shrink after the socket is welded. If trouble happens during welding (e.g. a fiber breaks and you have to re-partition the cable) you won’t have to pick at the cured thick adhesive shrink, and the shrink itself won’t be wasted.
i) The isolated cables are entered in a socket or cross-country are fixed, and the socket or cross-country is fixed on a desktop. At cable fixing in the socket or in cross-country it is necessary to be guided by the instruction on installation – for different sockets there all is different. In some cases (armored cable and, for example, a socket MTOK A1 with the corresponding complete set for input) cable fixing in a socket – separate not simple operation with trimming of an armor, winding of the sealant, etc.
So we got the cut cable into the socket/cross, now we have to measure and strip the fibers, put on the FAC and weld according to the diagram. I will tell you about it in the next part, because it is too much for one article.
Let me tell you a little bit about optical couplings and cross-arms. I’ll start with the couplings.
An optical coupler is a plastic container into which cables are wound and connected there. Previously, in the late 90’s and early 2000’s, when all the specialized materials for optics were in short supply with exorbitant prices, some smart guys molded sewer fittings or plastic bottles as couplers. Sometimes it even worked for several years. 🙂 Today it is certainly the wildness, normal couplings can be bought in any medium and large cities, and prices start from 1500-2000 rubles. There are a lot of coupling designs. The most popular and familiar design for me personally – this is how a series of couplings Svyazstroydetalev "MTOK. There is a headband, from which the outside stick out sockets for cable entry. From the inside of the headband attached a metal frame to which the optical cassettes are attached. On top of the cap is put (which for strength can be made with ribs), sealed with a rubber band. The cap is fixed with a split plastic clamp: the sleeve can always be opened and closed without wasting a thermal shrink kit.
In general, "Svyazstroydetal" makes a generally good coupling for various applications. From the MTOK series I personally like the L6 coupling best: universal, inexpensive, easy to install.
There are also other couplings in the MTOK series – small-sized, for sewers, for entering armored cables, for burying underground. Each coupling has the possibility to buy additional accessories and kits for cable entry: for example, cast-iron armor protection underground coupling "MCHZ", an extra set of optical cassette with consumables or an additional set for the entry of another cable.
If you need a cheaper – they have a series of couplings "MOG", of which the most popular – coupling "MOG-U" (Muff Optical City, shortened): the price of less than 2000 rubles, we get a simple and high quality coupling, which, incidentally, some think inconvenient for mounting.
On the pole such a coupling will not look good, and to coil the stock of cable with such a coupling, standing on the stairs, it is inconvenient, so they are usually put in the hatches. This clutch and created to be put in the manhole on special standard consoles. Minus "mogushka" – the fact that it does not have a locking connector clamp and to open it will have to cut the heat shrink, and when closing it to spend repair kit from a wide heat shrink (if the cables are brought in from one end) or heat shrink sleeve (if the cables from both sides). MTOKs of the A-series suffer from the same problem. In addition, if you put the cables in from both sides, it is important not to forget to put the plastic tube on one of the "sides" of the cables beforehand, otherwise it cannot be put on later without cutting it: this is also the problem of beginners.
There are also sometimes spigotless couplings where the cables are sealed by clamping them in raw rubber or sealant. Here, for example, is the "SNR-A" coupling that my partner and I were welding out as part of the FTTB ring construction.
This method of sealing cables requires great care, as otherwise water can get into the coupling, which is undesirable. Firstly, water in the coupling can cause clouding of the fiber glass and deterioration of the varnish over time. Secondly, all sorts of metallic structural elements will corrode and the armor ground wire will rot, if there is one. Third, Kevlar will draw water into itself. And most importantly, a coupling full of water will simply get crushed together with the fibers in a frost.
You usually wind at least two cables in an optical coupler. Of course, you can come up with a wild unwelding scheme where one cable will be inserted and unwelded on itself, but usually 2-3 cables are inserted. If you enter 4-5 cables, and even all the cables are different colors and different number of fibers in the modules, the coupling gets complicated for installation and the subsequent analysis of what is soldered to what. My first such coupling, my partner and I welded 3 days! 🙂 So it is better to design the network so that the coupling does not include more than 3 cables.
Optical cross is designed to terminate the cable in the place where it was brought: at the base station, in the data center, in the server room. A typical crossover is a 19quot size metal box for standard rack mounts, with the cable being terminated entered from the back and the port strips in the front.
24-port FC/APC type welded cross-connect, single-unit
Welded cross-over 64-port LC type, 2-unit
96-port FC-type working cross-connect
There is also a cheaper variant – when they cut everything out of the cross, then it turns out something like this :
Open cross-connect for 8 SC/APC type ports, 1 unit. The bad thing is that the optical pigtails are not protected in any way and they can be broken by those who will dig in the box/rack, pulling in, say, a new cable.
All of these crossovers are rack-mounted, but there are also wall-mounted variants and other rare ones.
Wall-mounted cross-connector for 16 FC ports. By the way, poorly welded: the yellow pigtail shells do not go into the CDC and the fibers can break, and the fibers in the cassette are stacked with small bending radii
The cable entering the cross is welded to the so-called pigtails: in the pictures these are the thin yellow cords inside the crosses. Each fiber goes to its own pigtail. The other side of the pigtail contains an optical connector "plug" which is inserted into the optical adapter "socket" from inside of the cross. Outside of the cross the switching is made by optical patch-cords (thick yellow cords). From the pigtail patch cord differs more durable connector and the presence of Kevlar inside, so that if someone gets a hold of the patch cord and yank, it was difficult to tear. Well, the connectors of patch cords have connectors on both sides and pigtails have connectors only on one side. If necessary, you can weld a temporary patch cord from two pigtails.
In principle it is possible to have several cables in the cross, some of the fibers can be welded together and some of them can be connected to the ports. Then you get something that can be called "crossover muff", and we save on materials and welding. This is sometimes done in FTTB installations, but it’s not a good idea to do it this way because it increases the complexity of the circuit.
Adapters and connectors
Optical cross-connections are characterized by the adapters used in them (simply – optical sockets). There are also a large number of standards and substandards.
This picture is just some of the "genera" and "kinds" of optical sockets
The standard is a set of adapter (socket) and connector (plug). Of course, there are adapters between different standards, but these are crutches that are only useful for measurements and should be avoided in a permanently working communication line. The less welds and especially mechanical connections in the line, the better. Of course, if the distance is small, the line will work, even if a couple of decibels will be lost on some of the cross-connections. In the case of short lines sometimes specially put optical attenuators. But for very long lines, where the equipment is working at the limit, adding one more cross or coupling (that is some 0.05-0.1 dB loss) can be fatal: the line will not go up.
The tip of the "fork" is, roughly speaking, a cylinder with a thin through hole for the fiber in the center. The end of this cylinder is not flat, but slightly convex. The tip consists of an amazingly hard and scratch-resistant ceramic-metal, although metal ones are very rare. Rumor has it that people have broken side cutters trying to unscratch this tip. 🙂 I myself have easily scratched steel and glass with these tips. Nevertheless, they must be handled carefully, don’t let dust get in, don’t touch the end of the tips with your finger, and if you have touched them, wipe with a cloth soaked in alcohol. Ideally, a special microscope (optical or camera) is used to control patch cords. Dirty – clean, scratched, if the scratch crosses the center with glued fiber – to be written off or polished. Dirty and scratched sockets and patch cords are a common cause of line attenuation.
Optical fiber is fixed in the tip by pasting epoxy (or some other) glue and subsequent grinding on a special machine, although this is done only if you need to make long non-standard patch cords: it’s easier and cheaper to buy ready-made. The price of an ordinary optical patch cord length of 2 meters – about 200-400 rubles.
Manufacture of patch cords. Emilink
In practice, standards such as FC, SC, LC are most commonly used. FC/APC, SC/APC, ST are less common. LC comes in both duplex and single-ended.
Pros – excellent connection quality, so suitable for critical mains. Old proven standard. Metal (hard to break). If you move the hand well screwed connector – it will not affect the connection.
Minuses – it takes a long time to unscrew/screw when switching. If the crossover is tight – it can be very inconvenient to get under to unscrew one of the connectors in a crowd of others.
The connector is held in place by a notch in the connector and a notch in the adapter, and only the notched nut can be twisted with the fingers.
The contact side of the tip is not flat, but slightly convex (this also applies to other standards) so that the two fibers from the two tips on opposite sides of the socket (pigtail and patch cord) are guaranteed to align without air and dust between them.
The receptacle contains a hollow, thin-walled ceramic cylinder with a longitudinal cut. When a plug is inserted into the socket, the cut gives out some microns, springing up and centering the plug. In this way, precision alignment of the two connectors in the socket is achieved (remember that the signal is transmitted over a 9 µm diameter fiber core and a shift of even 1 µm causes a loss of signal power at the socket and a parasitic back reflection). Therefore, dust and dirt is destructive for optical cross-connections, patch cords and pigtails should be regularly wiped with lint-free alcohol wipes, and sockets – blow with compressed air or clean with special cleaning sticks. A common cause of loss of communication is a burst ceramic insert in the socket.
In order for the connectors to press tightly together in the socket, in every FC and FC/APC connector (whether it is a patch cord connector or a pigtail connector) the ceramic-metal tip is spring loaded and can "press" into the plug by about a millimeter to a half. In SC, LC, ST standards the whole plug is spring loaded, and in the case of ST the fixing element is very similar to the one used in local networks on thin coaxial.
Everything is the same as FC, only the adapter and connector are square, plastic, and the connector is snapped in, not screwed in. Advantages – cheaper than FC, more convenient and faster switching, disadvantages – plastic is easier to break, less connection-disconnection life. Sometimes it happens that the amount of reflection and attenuation on the connection changes markedly after touching the connected connector, which is undesirable for critical lines. The color of the connectors is usually blue.
LC and LC Duplex
LC cross offers the same properties as SC but in much smaller dimensions: a two-unit LC cross holds 64 ports, while SC holds only 32. Due to their small size, they are often mounted directly on optical multiplexer boards.
FC/APC, SC/APC, LC/APC
Same as FC, SC and LC, but with an oblique (A – angle) polished tip.
The difference between ceramic tips with regular and oblique polishes. The image is a bit inaccurate: in fact, in both polishes, the ends are not flat, but slightly convex, so only the fiber centers of the tips will be in contact when joining them.
These adapters and connectors are made in green color and when comparing with the usual UPC (or just PC) polish the difference is visible to the eye. This is to reduce the back reflection at the junction of the two connectors. As far as I know this type of polish was developed for analog TV over fiber optics to avoid double imaging on the screen, but I could be wrong.
It is possible to dock "normal" and "oblique" polish together, but only if it is necessary to shoot a trace on the principle of "just to see the length of the trace": a big air gap will give strong losses and strong back reflection.
That’s the end of my story for today. Ask me questions, I will try to answer them. If you find this topic interesting I will write a sequel.