2.2.1 For a meter designed for a bilge alarm, the
accuracy should be within ±5 ppm. The accuracy of bilge alarms should remain
within the above limits despite the presence of contaminants other than oil, and
the power supply varying by 10% from the design value, i. e. in respect of
electricity, compressed air. etc.
2.2.2
The sampling arrangement should be such that a representative homogeneous sample
is obtained under all conditions of operation and under all operational
proportions of oil content. The sample should be obtained from the full flow
through the meter, but when this is impracticable the sampling arrangements shown
in figure 2 in part 1 should be used. Special care should be given to this stage
of the process and the validity of the resultant findings.
2.2.3 During the various tests the response time of the
meter should be checked and it should also be noted whether alarms operate
adequately when a pre-stated threshold is exceeded.
2.2.4 A diagrammatic arrangement of a test facility for
evaluating the performance of oil content meters is given in figure 3. The
accuracy of the oil content meter will be determined by comparing its readings
against a known flow of oil injected into a known flow of water. The grab samples
taken will be analysed in a laboratory by the methods specified in part 4 of this
annex. The results of the laboratory analysis will be used for correction and to
indicate sampling and test equipment variability. The water flow rate will be
adjusted so that the entire oil-water flow passes through the oil content meter,
except the intermittent grab sample stream. Special care should be given to keep,
continuously, a constant oil content in the water that flows into the meter. The
oil and contaminant metering pumps should be adjusted to deliver a nearly
continuous quantity of oil. If oil injection becomes intermittent at low
concentrations, the oil may be premixed with water to provide continuous flow if
absolutely necessary. The oil injection point should be immediately up-stream of
the oil content meter inlet to minimize time lags.
2.2.5 The oil content meter will be calibrated
and zeroed as per the manufacturer's instructions. It will then be tested with
light distillate fuel oil of a relative density of not less than 0.83 at
15°C at the following concentrations in ppm: 0, 15, and at the full scale
of the meter. Each concentration test will last for 15 minutes. Following each
concentration test, the meter will be run on oil-free water for 15 minutes and the
reading noted. If it proves necessary to re-zero or re-calibrate the meter during
this test, this will be noted.
2.2.6
The oil content meter for the bilge alarms should undergo contaminant tests as
follows:
the oil content meter for the bilge
alarms should be run on a 10 ppm light distillate fuel oil sample, and each of the
three contaminants listed below will be added to the water tank in the
concentrations given. Any shift in the meter reading should be noted and recorded
on the Certificate.
- Fresh water (if seawater is used for the test
programme)
- Very salt water - 6% common salt
with tap water.
- Non-soluble suspended solids
- about 10 ppm air cleaner test dust to the following specifications:
2.2.7 The meter should be run on a 15
ppm light distillate fuel oil sample. The water pressure or flow rate of the
mixture should be adjusted from one-half normal, normal and twice normal. Any
effect of these changes on the meter reading should be noted and recorded on the
Certificate. This test may require modification for meters with flow or pressure
regulators or meters designed to discharge into an ambient pressure sump.
2.2.8 The meter should be run on a 15 ppm
light distillate fuel oil sample. The water and oil injection pumps should be shut
off. The meter will be left turned on with no other changes made. After eight
hours, the water and oil injection pump should be turned on and set to provide the
mixture of 15 ppm. The meter readings before and after each test and any damage to
the meter should be noted and recorded on the Certificate. This test determines
also the proper functioning of the low flow shut-off and alarm.
2.2.9 If the meter requires any utilities besides
electricity, it should be tested with these utilities at 110% and 90% of the
design figures.
2.2.10 The meter should
be calibrated and zeroed. A 15 ppm light distillate fuel oil sample will run
through the meter for eight hours and any calibration drift noted. Following this,
the meter should be run on oil-free water and any zero drift noted and recorded on
the Certificate.
2.2.11 The response
time is to be taken for the meter to give an alarm at 15 ppm oil concentration
after the supply to the meter is changed from clean water to oily water, having
more than 15 ppm oil. The colour of the water supplied to the meter is not to
affect the operation of the instrument. The meter is to be tested for colour using
2.5 ppm black ink. The meter is to show no deviation when tested for colour using
2.5 ppm black ink.
2.2.12 A
specification of the instrument concerned and a diagrammatic presentation of the
test arrangements should be provided and the following data should be reported,
using the international Metric System of Units:
- types and properties of
oils used in the tests;
- details
of contaminants used, in the form, for example, of a supplier's certificate or
laboratory test protocol; and
- results of tests and analysis of grab samples.
2.2.13 The recommendations of the manufacturer of the oil content meter
concerning the choice and application of cleansing agents used for cleaning
purposes in machinery spaces should be recorded in the appendix to the certificate
of type approval. If the manufacturer declares that the user is free in the choice
and application of cleansing agents because this not affect the performance of the
equipment, then this is to be recorded also. The manufacturer should include this
information in the instruction manual of the oil content meter.