Analytical Industries GPR-2000 ATEX Portable Oxygen Analyzer - Manual (Page 11)

Advanced Instruments Inc.
11
Span Calibration
Span Calibration involves adjusting the transmitter electronics to the sensor’s signal output at a given oxygen standard.
Maximum drift from calibration temperature is approximately 0.11% of reading per °C. The frequency of calibration varies with
the application conditions, the degree of accuracy required by the application and the quality requirements of the user.
However, the interval between span calibrations should not exceed three (3) months.
Note: Regardless of the oxygen concentration of the standard used, the span calibration process takes approximately 10
minutes, however, the time required to bring the analyzer back on-line can vary depending on a combination of factors and
assume exposure to a zero/purge/sample gas** with an oxygen content below the stated thresholds immediately after span
calibration:
* Refer to analyzer specifications for comparable data on the Pico-Ion UHP and MS oxygen sensors.
Recommendations General:
¾ The interval between span calibrations should not exceed three (3) months.
¾ Initiate the DEFAULT ZERO and DEFAULT SPAN procedures before performing either a ZERO or SPAN CALIBRATION.
¾ Caution: Prematurely initiating the SPAN CALIBRATION function before the analyzer reading has stabilized can result in
erroneous readings. This is especially true when installing a new sensor that must adjust to the difference in oxygen
concentrations. It should take about 2 minutes for the sensor to equilibrate in ambient air from storage packaging.
¾ Always calibrate at the same temperature and pressure of the sample gas stream.
¾ For 'optimum calibration accuracy' calibrate with a span gas approximating 80% of the full scale range one or a higher
range than the full scale range of interest (normal use) to achieve the effect of “narrowing the error” by moving downscale
as illustrated by Graph A in the Accuracy & Calibration section.
¾ Calibrating with a span gas approximating 5-10% of the full scale range near the expected oxygen concentration of the
sample gas is acceptable but less accurate than ‘optimum calibration accuracy’ method recommended – the method usually
depends on the gas available.
¾ Calibrating at the same 5-10% of the full scale range for measurements at the higher end of the range (example:
calibrating an Oxygen Purity Analyzer in air at 20.9% oxygen with the intention of measuring oxygen levels of 50-100%)
results in the effect of “expanding the error” by moving upscale as illustrated by Graph A and Example 1 in the Accuracy &
Calibration section above and is not recommended. Of course the user can always elect at his discretion to accept an
accuracy error of +2-3% of full scale range if no other span gas is available.
Recommendations Air Calibration:
¾ Do not calibrate an analyzer employing the Pico-Ion UHP or MS sensor, or, an oxygen purity sensor with air.
¾ The inherent linearity of the galvanic fuel cell type oxygen sensor enables the user to calibrate any analyzer with ambient
air (20.9% oxygen) and operate the analyzer within the stated accuracy spec on the lowest most sensitive range available
with the analyzer – it is not necessary to recalibrate the analyzer with span gas containing a lower oxygen concentration.
¾ When installing or replacing a ppm or percent oxygen sensor.
¾ To verify the oxygen content of a certified span gas.
¾ When certified span gas is not available to calibrate a ppm analyzer (immediately following air calibration reintroduce a gas
with a low oxygen concentration to expedite the return to ppm level measurements as described above **).