Bookkeeping - to the masses! Typing up the optical budget

The expression “optical budget” causes deep sadness among technical translators - they say that the English “optical budget” is more correctly translated as “line energy potential” or, at least, as “optical balance”. Most likely, they are right: tracing paper has never been the best way to translate. However, in this case, they, apparently, will have to put up with it: the “optical budget” has entered into the everyday life of telecom people so deeply that you can’t replace it with any balance ...
But what is this optical budget and what does it eat?

Making ends meet

Any oscillation propagating in a real physical medium with resistance gradually fades away. Upon reaching a certain boundary, our signal dissolves into a sea of ​​noise, and no sensitive receiver can decode it anymore - this boundary determines the range of this type of communication. Some theoretical, ideal range indicators are given in the description of the standards, but it should be understood that in most cases it is nothing more than a “spherical horse in a vacuum,” which can be considered with caution as a hardly achievable maximum.
In order for the calculation of signal propagation to be at least approximately consistent with reality, it must take into account the real characteristics of the transmitter, receiver and everything that is between them. For fiber optic communication lines (FOCL), this equation is called the optical budget. Let's see how it is composed.


Optical Budget Calculation Scheme

The signal source is characterized by its power, which varies within certain limits; take its maximum value. On the propagation path of the optical wave, it will weaken at the junction points of the cables (connectors, optocrosses, welding) and gradually along the entire length of the cable. At the end point, the luminosity of the stream should be no less than the limit threshold of the receiver sensitivity - otherwise the network will not work. The budget should converge in both directions if the traffic is duplex. All parameters related to the line can be considered constants for a certain type of optical radiation - here they are.

Parameter	Multimode, 850 nm	Single mode, 1310 nm
Line attenuation	3 dB / km	0.4 dB / km
Connector attenuation	0.5 dB	0.2 dB
Weld Attenuation	0.1 dB	0.1 dB

Calculation Examples

Consider the budget of a particular optical line. We believe that the single-mode communication line between the server with the Intel network card and the Cisco Catalyst switch has a length of 10 km, there are 2 optical crosses on it. Let's agree that hereinafter we will talk about 10-gigabit Ethernet - from the point of view of optics, the data speed does not matter, but the parameters of the network interfaces are different:

	Luminance, dBm		Sensitivity, dBm
	Min	Max	Min	Max
Intel Ethernet Server Adapter	-8.2	0.5	-12.6	0.5
Cisco X2 LR	-8.2	0.5	-14.4	0.5

Of the pair of boundary values, we need only one, the second is given for completeness. Note that the parameters of the optical modules are very similar, but this is not always the case. One more remark: in order not to embrace the immensity, we will miss the optical subtleties associated with the characteristics of the interfaces and their measurements and limit ourselves to the final figures - this is quite enough for us. These characteristics for a specific interface can be found in the product description on the manufacturer's website.
So, let’s do the calculation (it will be the same in both directions):



As you can see, the budget converges with a large margin - this is generally typical for single-mode lines, as, indeed, for multi-mode lines, if you use them normally. Consider the same problem with SR optics and a line length of 300 meters:

	Luminance, dBm		Sensitivity, dBm
	Min	Max	Min	Max
Intel Ethernet Server Adapter	-5.0	-1.0	-11.1	-1.0
Cisco X2 SR	-7.3	-1.2	-9.9	-1.0

Optical budget from Intel to Cisco and vice versa:

-1.0 - 0.5 - 0.5 - 0.5 - 0.5 - 0.5 - 0.5 - 0.3 * 3 = -4.9
-1 , 2 - 0.5 - 0.5 - 0.5 - 0.5 - 0.5 - 0.5 - 0.3 * 3 = -5.1

The margin is slightly smaller (taking into account the lower sensitivity), but still sufficient. The question is, why then consider the budget, if everything is always the same? The answer is this: far from all optical lines operate in a mode comfortable for themselves. The main purpose of the budget is to come to the rescue when something goes wrong. Let's say there is a line that does not behave as it should. What is the problem - in equipment malfunctions or in principle, it can’t work stably in this configuration? It is here that our tsifirki can help out.
So, we conclude: in the case when the line is made "wisely", in compliance with standards and according to common sense, it is not necessary to consider the optical budget, this can only be done if problems arise. However, there are cases when the calculation is simply necessary - we will move on to them.

Special cases

Imagine the opposite situation: you need to connect the equipment through an optical line of very short length, literally “back to back”. Can this mode be considered comfortable for network interfaces? Not at all. Overloading (flashing) can lead to errors on the port, or even to its physical failure. In order to bring the port to life, you need to reduce the luminous flux - put the attenuator. Attenuators are fixed or variable attenuation; how to substitute it into the equation above - I think everyone has already guessed.

Optical attenuator

It is well known that when selecting attenuators they often use the “poke method”, but, really, it’s not worth guessing, when in a few minutes you can get the optimal indicators for a given case by calculation.

This case is even more complicated. Sometimes passive couplers - splitter are placed on optical lines. A splitter is a translucent glass prism that passes part of the light through itself, and part reflects in the opposite direction. Thus, literally “free”, without any active devices, you get two copies of the same signal. Most of the light remains in the line, and the second copy is fed to the traffic analyzer, anomaly detector, corporate security device, etc.


Optical splitter The

peculiarity of the situation is that in the case of splitters, the budget must be considered for two lines - the main and the branch. In addition, the splitter makes significant attenuation in the line, which also makes life difficult. Let's add a splitter to our optocrosses and see what happens.



Let's calculate the budget for the 70/30 splitter. For a single-mode line, the attenuation introduced by it is as follows: 2 dB in the main line and 6.1 in the branch line. It is only necessary to add this attenuation to the previous equations (or rather, subtract it).

Basic: 0.5 - 0.2 - 0.2 - 0.2 - 0.2 - 0.2 - 0.2 - 10 * 0.4 - 2 = -6.7
Branch: 0.5 - 0, 2 - 0.2 - 0.2 - 0.2 - 0.2 - 0.2 - 10 * 0.4 - 6.1 = -10.8 The branch

line operates at the limit of its sensitivity, it is better to change the type of splitter to 60 / 40 - this will give us an extra decibel in a dangerous area.
Now the multimode (2.4 / 6.3 dB attenuation), for simplicity we calculate one direction:

Basic: -1.0 - 0.5 - 0.5 - 0.5 - 0.5 - 0.5 - 0.5 - 0.3 * 3 - 2.4 = -7.3 Branched
: -1.0 - 0.5 - 0.5 - 0.5 - 0.5 - 0.5 - 0.5 - 0.3 * 3 - 6.3 = -11.2

As you can see, the branch line is out of the budget, and even replacing the splitter with 60/40 will not solve the problem. In general, we can say that installing splitters on multimode lines is a serious problem and it is better not to do this without a strong need.

I would like to complete the topic with the following words. Probably, the ability to calculate the optical budget of the lines is not one of the primary skills of a network administrator. However, if you are in charge of optical networks or if you intend to try yourself in this field, we recommend that you at least have an idea about this subject, in order to know where to look and what to do in the future.

If the topic is habrainteres, next time we’ll talk about practical ways to control optical lines, typical optical problems and the influence of curvature of the hands on the trajectory of the light beam.