WATERS CORPORATION 2424, Operator'S Manual, Page: 26 / 196

1-12 2424 ELS Detector Optics Principles
The signal of the lamp is measured, and the normalization constant is
adjusted accordingly to compensate for lamp intensity variations. This
minimizes the influence of lamp intensity changes on detected signal levels.
All settings are restored to the values present when the unit was shut down,
except for the heater setpoints and gas flow, which must be specified.
Recommendation:
Turn the power off and on once a week to compensate for
lamp aging.
Lamp energy and performance
In conventional designs of ELS detectors, the signal-to-noise performance of
the instrument is directly proportional to the lamp energy input to the
instrument. Lamp energy input to the detector can be affected by these
factors:
• Age and efficiency of the lamp
• Improperly maintained optics
• Normal degradation of optical components (including the
photomultiplier tube)
Optical components degrade slowly. In conventional ELS detectors, response
increases by incrementally increasing the gain. However, a sample’s response
varies with energy throughput. If the lamp energy is degraded, peak response
degrades accordingly. If lamp intensity diminishes, peak response decreases
and noise increases. During normal operation, lamps are usually replaced
when the reference energy falls below a user-set threshold. The useful lamp
life depends on the method’s specific requirements for noise performance.
Eventually, the detector’s performance becomes unacceptable and the lamp
replacement is necessary.
Tip:
It is good practice to inspect the detector’s general condition when lamps
are replaced.
Predicting when the detector’s performance degrades to an unacceptable level
based solely on reference energy is unsatisfactory. Each user’s analyses
require different levels of sensitivity. Determining reference energy alone to
evaluate performance assumes that lamps exhibit the same longevity and
degradation patterns. Waters therefore designed the detector to operate as
independently of lamp output as possible. Ultimately, the detector’s
performance is a function of unique application requirements.
Signal-to-noise measurements are the best way to evaluate performance and
set boundaries for acceptable operational sensitivity limits. Waters