KI 7010B-UM-11
16
ACCURACY CONSIDERATIONS
All Measurements
•
Optical connectors should be kept clean and in good condition.
•
There are limits on the repeatability of the attenuator setting. This is
typically ± 0.03 dB.
•
There may be inaccuracies due to the operational wavelength being
slightly different to the calibration wavelength. This error is normally
small, since the instrument has a flat spectral response. The expected
inaccuracy can be approximated as:
Actual attenuation = display reading - 0.00015 x n x display reading,
where n = wavelength discrepancy in nm
For example, with the attenuator set at 1310 and 60 dB, and with 1550
nm light, the actual attenuation will be approximately 57.12 dB.
•
Polarisation changes create disturbances in both the optical system and
measurement equipment. To reduce this, keep your system physically
stable. Keep patch leads neat, coiled and stable.
•
Temperature effects are too small to characterise reliably and, are
unlikely to be an issue.
•
All Ge power meters are inherently non-linear by about 0.04 dB. Ge
meters also have temperature dependent calibration drift at 1300 nm of
typically 0.03dB per 10 ºC. This will noticeably degrade the accuracy of
work done with such a meter.
•
Light source power may drift: when you have finished a test, go back to
the start position to check if your meter reading is still within acceptable
limits.
•
Cladding mode light transmission can affect some fibre types (most fiber
strip off cladding modes).
•
Measurements on multimode fibre may be unrepeatable and inaccurate
unless a mode filter is used to obtain an equilibrium modal distribution.
You may find that your power meter linearity disagrees with the attenuator at
low power levels. Many power meters have poor linearity at low power levels.
If in any doubt, introduce a step attenuator into the system (for example 20
dB, a bad connection will suffice!), and repeat similar power meter readings
using a different region of the attenuator.
For example, if during testing the power meter disagrees with the attenuator
where the meter reads below -40 dBm and the attenuator was between -35
and -50 dB to achieve these readings, try introducing an additional 20 dB loss
somewhere else. Now repeat the measurement with the power meter still
reading below -40 dBm, but to achieve this, the attenuator is reading between
-15 and -30 dB. If the same non-linearity is evident, then it is the power meter
that is none linear.
Power Meter/detector Coherence sensitivity
Fiber optic power meters and detectors typically give lower accuracy and
repeatability when used with coherent laser light. The more coherent the light,
the worse this effect. The table below gives a very approximate summary. So,
for best test repeatability, use non-coherent light if possible.
Typical Linearity / Repeatability (dB)
Detector type
Non-coherent
source
Febry-Perot
Laser
DFB
Laser
InGaAs
0.005
0.08
0.3
Si
0.005
0.08
0.3
Ge
0.04
0.08
0.3
Содержание KI 7010B Series
Страница 1: ...OPERATION MAINTENANCE GUIDE KI 7010B Series Automated Variable Optical Attenuator ...
Страница 5: ...KI 7010B UM 11 3 KI7000B SERIES AUTOMATED VARIABLE OPTICAL ATTENUATOR ...
Страница 32: ...KI 7010B UM 11 30 ...
Страница 34: ...KI 7010B UM 11 32 NOTES ...
Страница 35: ...KI 7010B UM 11 33 NOTES ...
Страница 36: ...KI 7010B UM 11 34 NOTES ...
