Analytical Industries GPR-1900 Series Trace PPB Oxygen Analyzer - Manual (Page 8)

Advanced Instruments Inc.
8
Accuracy: In light of the above parameters, the overall accuracy of an analyzer is affected by two types of errors: 1) those
producing 'percent of reading errors', illustrated by Graph A below, such as +5% temperature compensation circuit, tolerances
of range resistors and the 'play' in the potentiometer used to make span adjustments and 2) those producing 'percent of full
scale errors', illustrated by Graph B, such as +1-2% linearity errors in readout devices, which are really minimal due to today's
technology and the fact that other errors are 'spanned out' during calibration.
Graph C illustrates these 'worse case' specifications that are typically used to develop an analyzer's overall accuracy statement
of < 1% of full scale at constant temperature or < 5% over the operating temperature range. QC testing is typically < 0.5%
prior to shipment.
Example 1: As illustrated by Graph A any error, play in the multi-turn span pot or the temperature compensation circuit,
during a span adjustment at 20.9% (air) of full scale range would be multiplied by a factor of 4.78 (100/20.9) if used for
measurements of 95-100% oxygen concentrations. Conversely, an error during a span adjustment at 100% of full scale range is
reduced proportionately for measurements of lower oxygen concentrations.
Zero Calibration: In theory, the electrochemical galvanic fuel cell type oxygen has an absolute zero meaning it produces no
signal output when exposed to an oxygen free sample gas. In reality, expect the analyzer to generate an oxygen reading when
sampling oxygen free sample gas due to contamination or quality of the zero gas; minor leakage in the sample line connections;
residual oxygen dissolved in the sensor’s electrolyte; and, tolerances of the electronic components. The Zero Offset capability of
the analyzer is limited to 50% of lowest most sensitive range available with the analyzer.
Recommendation 1: Zero calibration, see Determining True Zero Offset below, is recommended only for online analyzers
performing continuous analysis below 5% of the lowest most sensitive range available with a ppm analyzer, e.g. analysis below
0.5 ppm on the 10 ppm range, or below 0.1% (1000 ppm) with a percent analyzer. Note 1: Once the zero offset adjustment is
made, zero calibration is not required again until the sample system connections are modified, or, when installing a new oxygen
sensor. As a result, zero calibration is not practical and therefore not recommended for higher ranges or portable analyzers.
Determining True Zero Offset: Allow the analyzer approximately 24 hours to stabilize with flowing zero gas as evidenced by a
stable reading or horizontal trend on an external recording device. Note 2: 24 hours is required to assure the sensor has
consumed the oxygen that has dissolved into the electrolyte inside the sensor while exposed to air or percentage levels of
oxygen. For optimum accuracy, utilize as much of the actual sample system as possible.
Span Calibration: Involves adjusting the analyzer electronics to the sensor’s signal output at a given oxygen standard.
Regardless of the oxygen concentration of the oxygen standard used, a typical span calibration takes approximately 10 minutes.
Note 3: The amount time required to get the analyzer back on line for normal use is influenced by a.) the level of oxygen
analysis anticipated during normal operation (also determines the initial analyzer selection), and, b.) whether the sensor is new
or has been in service for at least two weeks.
General guidelines for analyzers to come online following span calibration and exposure to a zero/purge/sample gas with an
oxygen content below the stated thresholds:
¾ measurements above 1000 ppm or 0.1% require less than 3 minutes
¾ measurements above 100 ppm (parts-per-million analyzer) require less than 10 minutes
¾ measurements below 10 ppm (part-per-million analyzer) require 20 minutes if the sensor has been in service at ppm
levels for at least two weeks, and, 60 minutes if the sensor is new assuming the zero/purge/sample gas has an oxygen
concentration below 1 ppm
Recommendation 2: For 'optimum calibration accuracy' calibrate with a span gas approximating 80% of the full scale range
one or two ranges higher than the full scale range of interest (normal use) to achieve the effect illustrated on Graph A and
Example 1. Always calibrate at the same temperature and pressure of the sample gas stream.