Maintenance 9-3
9.5 Functional Check
9.5.1 Introduction
This check should be repeated as often as appropriate, at the start of a shift, on
change of component, etc.
This procedure entails two tests: the first is to check the system integrity for leak
tightness, and the second is to ensure that the system reports a reject condition for
a known leaking component.
9.5.2 Leak Tightness Test
1. Set up the system for a normal test run. The system must be complete with all
the usual pipe work. Install a test component that is known to be leak tight.
Clamp and seal the component in the usual way. Run the system through one
test cycle and note the resultant reading.
2. Ideally, the system should be leak tight and give a near zero reading for good
components. Any large leaks should be located and repaired.
3. In practice, the technique used may indicate a leak. If so, check that the effect
is consistent from one leak tight component to another.
9.5.3 Leak Comparison Test
This test requires the use of a standard, calibrated leak, which must be fitted to the
normal pipe work of the installation.
Standard leaks should be manufactured to a specific requirement of gas type,
pressure and leak rate with a certificate of traceability to national standards.
Suitable leaks can be made to order, and are available through your supplier.
Standard leaks must be checked at least once a year either, against a suitable flow
meter, or by returning it to the supplier. A Standard leak should be treated as a
calibrated instrument.
1. Following on the Leak Tightness test, install a known, standard leak on the
normal test installation.
2. Run the system through one test cycle and note the resultant reading.
• The difference between the two readings should be the same as the standard
leak.
• If the readings do not match, the volume or temperature of the system may have
changed since the last flow rate calibration and need to be investigated. A re-
calibration of the system will be required to compensate for any changes.
• If the Reject Level is set to coincide with the standard leak used, the result of
the Comparison Test should have been signalled by the instrument as a failed
component.