COMPUTER NETWORKS

LAB3: Optical Fibers

Date of prosecution: 08/03/2020
Team: Daskalakis Stylianos

Lab is focused on measurement of parameters of optical fibers,

observing their features and evaluation of influence of different
disturbances, e.g. an incorrect coupling in connectors or
mechanical stress.
Two test modules are used for measurement tasks. Tx module
contains signal generators and light sources (LED, IRED – infrared
LED, laser diode), Rx module allows reception and contains
photodiodes, amplifiers and other measurement circuits.

1. Attenuation introduced by axial displacement

During this measurement, influence of increasing axial gap size on
the power loss is evaluated. This is depicted on Fig. 1.

Following configuration is used for this measurement.

We configured the transmitter (Tx) and the receiver (Rx) in the
appropriate manners. (Fig. 2)
We measured the attenuation for axial displacement in range x =
0–35 mm (step 1 mm for gap up to 10 mm, then with 5 mm steps).
Specifically, we wrote down the optical power in both dBm (decibel
above mW) and in μW, as we can see in Table 1 .

Table 1: Optical power for a given distance x in both dBm and in μW

Distance x (mm) dBm μW

0 -11.88 16.8
1 -8.02 41.0
2 -9.50 29.2
3 -12.54 14.4
4 -14.91 8.4
5 -16.74 5.5
6 -18.45 3.7
7 -19.62 2.84
8 -20.77 2.18
9 -21.90 1.67
10 -22.87 1.34
15 -26.57 0.57
20 -29.22 0.31
25 -31.02 206nW
30 -32.66 141nW
35 -34.00 103nW

The power conversion of dBm to mW is given by the formula:

P(mW) = 1mW ⋅ 10(P(dBm)/ 10) (*formula 1)
Unfortunately, the conversion between the power in dBm and in
μW does not seem valid! In Table 2 we see some valid results
using the formula.
Table 2: Valid conversion between the power in dBm and in μW

Distance x (mm) dBm μW

0 -11.88 0.065
1 -8.02 0.16
2 -9.50 0.11
 3 -12.54 0.06
4 -14.91 0.03
5 -16.74 0.02

We calculated the dependence of attenuation in dB on the axial

gap size (considering that reference – 0 dB – is for x = 1).
For example, attenuation for x = 5mm:
a = Px=5 – Px=1 (dB) (*formula 2)& P-Power in dBm.

Table 3: Attenuation in dB on the axial gap size

Distance x(mm) Attenuation a (dB)

1 0
2 -1.48
3 -4.52
4 -6.89
5 -8.72
6 -10.43
7 -11.6
8 -12.75
9 -13.88
10 -14.85
15 -18.55
20 -21.2
25 -23
30 -24.64
35 -25.98

As we can see in Table 3, attenuation decreases as the distance

increases.
Graph 1: Dependence of attenuation on the axial gap size.

Distance(mm)
0
0 5 10 15 20 25 30 35 40
Attenuation a (dB)

-5

-10

-15

-20

-25

-30

2. Attenuation introduced by axes rotation

Fig. 3 presents the measurement of the power loss caused by
rotation of fiber axis. Optical power decreases with increasing
angle, from limit angle no power is transferred at all.
We used the following setup for this measurement:

Keeping the settings of both Tx and Rx modules we set the axial

distance between fibers to 5 mm. We measured for angle in range
x = −25°  +25° with 5° step. We see the results in Table 4.

Table 4: Optical power for a given angle θ in μW

θ(°) μW
-25 0.57
-20 2.35
-15 4.1
-10 5.0
-5 5.4
0 5.2
5 4.7
10 3.6
15 1.85
20 0.48
25 110nW

Table 5: Dependence of received power ratio (100% for 0°)

θ(°) Received power

ratio
-25 10.9%
-20 45.2%
-15 78.8%
-10 96.2%
-5 100.3%
0 100%
 5 90.4%
10 69.2%
15 35.6%
20 9.2%
25 2.1%

Graph 2: Dependence of received power ratio by axes rotation.

120.00%
Received power

100.00%

80.00%
ratio(%)

60.00%

40.00%

20.00%

0.00%
-30 -20 -10 0 10 20 30
θ(°)

3. Numerical aperture measurement

A numerical aperture of optical fiber can be calculated from the
results of previous measurement.

Numerical aperture is sinus of nm, which is the maximum angle for

which the beam of light reflects totally at the interface between the
fiber core and cladding. During measurement, the NA is found
when the optical power falls to 5 % of original value. Fiber NA can
be evaluated as 𝑁𝐴 = √𝑛12 − 𝑛22 , where 𝑛2 is refractive index of
the core and 𝑛1 is refractive index of cladding.
Using the results of previous measurement, we calculate the
numerical aperture of the fiber. We can say that 5 % of original
value appears when the angle is about 23°. So, NA = sin23° = 0.39
When the core refraction index is 1.41 and cladding refraction
index is 1.49, the NA = 0.48. This is quite higher than our
measured value.

4. Attenuation introduced by transversal displacement

Figure 6 presents the power loss caused by transversal
displacement of fiber axes.

This measurement uses the following setup.

We kept the settings of both Tx and Rx modules and measured for

displacement in range x = −5  +5 mm with 0.5 mm step.
Unfortunately, we measured the received optical power in μW
although the task was to measure it in dBm from the Rx module
display. So, I guess, since the conversion from dBm to μW is not
valid we cannot find the exact value of dBm that the device
showed. However, I did the conversion of μW to dBm using the
formula above.
 Table 6: Received optical power

Distance x (mm) μW/nW dBm

-5 TOO LOW TOO LOW
-4.5 TOO LOW TOO LOW
-4.0 27.6nW -15.69
-3.5 51nW -12.92
-3.0 70nW -11.55
-2.5 99nW -10.04
-2.0 0.33μW -4.81
-1.5 1.09μW 0.37
-1.0 2.75μW 4.39
-0.5 3.6μW 5.56
0 4.2μW 6.23
0.5 4.1μW 6.13
1.0 3.4μW 5.31
1.5 2.53μW 4.03
2.0 1.34μW 1.27
2.5 296nW -5.29
3.0 110nW -9.59
3.5 60nW -12.22
4.0 33nW -14.81
4.5 15.8nW -18.01
5.0 TOO LOW TOO LOW

I evaluated the attenuation in dB for each displacement value

(reference value for x = 0 mm).
Table 7: Attenuation in dB for each displacement value

Distance x (mm) Attenuation (dB)

-4.0 -21.92
-3.5 -19.15
-3.0 -17.78
-2.5 -16.27
-2.0 -11.04
-1.5 -5.86
-1.0 -1.84
-0.5 -0.67
0 0
0.5 -0.6
 1.0 -0.92
1.5 -2.2
2.0 -4.96
2.5 -11.52
3.0 -15.82
3.5 -18.45
4.0 -21.04
4.5 -24.24

Graph 3: Dependence of attenuation transversal displacement.

Distance x (mm)
0
-6 -4 -2 0 2 4 6
-5
Attenuation (dB)

-10

-15

-20

-25

-30

(We didn’t measure for two axial gap sizes of 5 and 10 mm.)

5. Utilization of the AC optical signal

In all previous measurements we used the AC test signal, instead
of the DC signal. The reason is simply because with the AC signal
the offset part is deleted. So, it’s like we get the original signal.

6. Attenuation introduced by fiber bending

Another source of attenuation is fiber bending, when the conditions
for total refraction may fail and beam of light get out from the core.
This is demonstrated by the fiber wrapped around the roll of
different diameter. With decreasing diameter, the attenuation
increases. For low diameter roll and red light source the light
leakage from the cladding is visible.
Following setup is used for the measurement.

We kept the settings of both Tx and Rx modules and interconnect

them by transparent optical fiber. We coiled single loop of fiber
round the ring and measured the additional optical power
attenuation for ring diameters 3, 4 and 5 cm.
Table 8: Optical power with fiber bending

Diameter (cm) μW
3 136
4 146
5 150

7. Fiber attenuation dependence of the light wavelength

This measurement uses following setup.
We kept the settings of both Tx and Rx modules and used the
L1=1m long fiber to interconnect Tx and Rx modules. We read the
received power level in dBm for first four LED at Tx module (526
nm, 590 nm, 660 nm and 850 nm) in successive steps.
We used OUTPUTS CH1 button at Tx module to connect output
signal to the respective LED.
Table 9: Received power level using 1m long fiber

nm dBm
526 -15.41
590 -18.78
660 -4.19
850 -24.35

We repeated the same sequence for L2=50 m long fiber.

Table 10: Received power level using 50m long fiber

nm dBm
526 -26.62
590 -30.70
660 -22.2
850 TOO LOW

We also evaluated the fiber attenuation for wavelengths in dB/km

using the following formula:

(*formula 3)
Table 11: Fiber attenuation

nm Attenuation (dB/km)
526 -228.8
590 -243.29
 660 -367.58
850 -5140**

**The attenuation was so high for 850 nm that the output power
was non-measurable for 50 m long fiber. So, we used 2 m fiber
instead (we connected two 1 m fibers using ST/ST coupling).
Then, the power level for 2m long fiber at 850nm was -29.29dBm.
That’s how a=-5140 occurred.

8. Light color shift

We positioned one end of 50 m fiber towards the white light source
and checked the color at the other end. I didn’t write down the
color that appeared, so I don’t really remember what happened in
this case.

9. WDM transmission
We didn’t have time to accomplish this task. But I can just describe
the crosstalk phenomenon.
In electronics, crosstalk is any phenomenon by which a signal
transmitted on one circuit or channel of a transmission system
creates an undesired effect in another circuit or channel. Crosstalk
is usually caused by undesired capacitive, inductive or conductive
coupling from one circuit or channel to another.
Crosstalk is a significant issue in structured cabling, audio
electronics, integrated circuit design, wireless communication and
other communications systems.

10.Conlcusion
In conclusion, power of optical fibers is influenced not only by the
axial gap and axes rotation but also by transversal displacement.
Increasement of each of these three parameters decreases the
power. Fiber bending is another source of attenuation. The bigger
the diameter of bending is, the less attenuation appears.
Moreover, we can measure the numerical aperture when the
optical power falls to 5 % of original value. Finally, as the light
wavelength increases, the fiber attenuation also increases.