130-0047 Issue 8
USED FOR CONTROL, OR THE ASSOCIATED CONTROL LOOP
SHOULD BE DISABLED BEFORE COMMENCING.
5.1 Sensor life
The N sensors have a life of 100,000 oxygen % hours or 18 months
- whichever is the sooner; and the L type a life of 16,000 oxygen %
hours or 18 months whichever is the sooner. Each sensor is date
labelled (see the Storage of the Measuring Sensor section 6.2 for
the code) when supplied. This date should be used to establish the
expiry date.
5.1.1 Sensor failure modes
All known sensor failure modes result in a loss or lowering of output.
Thus applications that look for oxygen depletion are automatically
fail safe and vice-versa
.
5.2 Calibration interval
It is recommended that the calibration is verified every month. This
frequency of verification is for typical industrial applications. When
the application is critical, the frequency of verification should be
increased in line with local safety standards.
5.3 Calibration gas level
The level of oxygen in the calibration gas depends on the particular
model and in particular the type of sensor fitted. In general, for
optimum accuracy, it best to calibrate with a gas with the same
composition and oxygen concentration as the normal sample. If you
calibrate on air, the response time for the sensor to stabilise back to
its normal working level after being exposed to air should be taken
into account.
Moreover if an air calibration is performed, it should be borne in
mind that most air taken from the atmosphere is moist. The normal
figure of 20.95% for the oxygen content of air is for dry air. Ambient
air contains nearer 20.7% because of dilution effect of the
Page 12 of 18
moisture. It is this figure that should be used when calibrating in this
way.
5.4 Calibration of a percent range instrument.
(See section 5.5 for instruments scaled in PPM)
5.4.1 Piping
The piping carrying the calibration gas must not have any leaks, and
should have an outlet pipe at least 300mm long to reduce any effect
of back diffusion. The flow rate of the gas must be controlled and
should be the same as the flow rate of the sample from the process.
Also, the pressure at the sample outlet must be the same as that
when the analyser is measuring the sample. This avoids any errors
due to the pressure at the sensor. Refer to Fig 2 for details of a
typical installation.
5.4.2 Calibration method
Minutes
Response of an N type sensor from an air saturated
condition when zero grade nitrogen is applied.
Fig. 4
10
2
3
Hours
PP
M
O
x
yge
n
4
5
6
7
8
9
10
11
12
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
Fig. 5
Response of an L type sensor from an air saturated
condition when zero grade nitrogen is applied
