Читать книгу Continuous Emission Monitoring - James A. Jahnke - Страница 102

In today's applications for CEM systems, new demands have been made for dilution systems and CEM systems in general. For example, in the acid rain program, relative accuracy specifications became more stringent at 10%, daily calibration error (drift) is limited to 2.5% of span, and a 5% linearity criterion has been established. Although relative accuracy specifications are not difficult to meet, it is found that a clear understanding of the operating characteristics of dilution systems is necessary to achieve the optimum performance necessary for meeting the drift and linearity test specifications.

Figure 3‐24 STI external dilution system design.

In dilution systems, the flue gas concentration is calculated from the dilution ratio as follows:

(3‐3)

where

 c = the calculated source‐level concentration of the gas (ppm)

 cmeas = the analyzer response (ppm) to the diluted sample

 Do = the dilution ratio at the time of calibration

This equation assumes that the dilution ratio remains the same while the reading c_meas is obtained as when the dilution system was calibrated initially. This is not always the case. If the absolute pressure changes from P_o to P, the stack temperature changes from T_o to T, or the molecular weight of the sampled gas changes from M_o to M, the dilution ratio will change also. Pressure, temperature, and molecular weight all affect gas density, which affects the sonic flow of the gas through the orifice. A change in gas density will therefore affect the dilution ratio.

Problems arise when the original conditions at which the dilution system was first calibrated change for any subsequent measurements. Figure 3‐25 illustrates the principal factors that may cause a change in the dilution ratio.Consider the following scenarios illustrated in the figure:

Figure 3‐25 Principal factors causing changes in dilution systems.

 Scenario 1 A dilution system is calibrated initially when the total stack pressure PT = Pbar + ps, the unit is on, the flue gas temperature = To, and the calibration gas has an average molecular weight = Mo. Typically, any variation from the nominal dilution ratio is adjusted out by setting the analyzers to the certified calibration gas concentration during calibration.

 Scenario 2 A dilution system is calibrated at a stack absolute pressure of Po. A weather front moves in and the atmospheric pressure is reduced. A calibration check will indicate an upward drift in reading.

 Scenario 3 A dilution system is installed on a cycling unit and is calibrated when the unit is off, at a temperature To. After the unit is turned on and the stack temperature increases, a calibration check will indicate a downward drift in reading.

 Scenario 4 A dilution system is calibrated with an SO2 in N2 calibration gas. An auditor conducts a linearity test using a gas mixture containing SO2 and 20% CO2. The test gives a reading lower than expected.

These are all scenarios that may cause discrepancies in dilution system measurements. However, these discrepancies can be corrected (i) empirically or (ii) theoretically. Empirical corrections, based on experimental data from installed systems, have been used successfully in specific applications. Lacking experimental data, theoretical expressions can be used and have also been successful in correcting for the effect of gas density variation on the dilution ratio.