OCR Text |
Show 6 Furnace air in-leakage can be characterized by comparing measurements of the O2 concentration at the furnace exit and economizer exit as shown in Figure 2. The O2 measurements made at the furnace exit plane of a 125 MW face-fired unit at the Whiting Station of Consumers Power Company are shown in the lower half of the figure. These measurements were made with a water cooled high velocity thermocouple (HVT) probe, and a computer graphics interpolation program was used to generate the O2 contours. The O2 contours at the economizer exit (upper half of the figure) were generated using the same software based on O2 measurements made with a multi-point O2 analyzer at the 12-point sample grid. The air in-leakage was estimated based on the O2 increase along a hypothetical stream line between the two locations as shown in Figure 3. The high O2 region along the north wall of the economizer duct corresponds to an in-leakage approaching 180/0, whereas the in-leakage along the opposite south wall is only 2 to 60/0. The higher O2 region along the back (west) wall of the economizer exit is most likely due to air in-leakage from the furnace roof region. The air in-leakage profile from Whiting Unit 3 can be used to predict the furnace exit O2 levels based on economizer readings rather than performing additional labor intensive HVT measurements. The "02 correction factor" could also be used to "adjust" the plant O2 analyzer readings to more property reflect furnace combustion conditions. Diagnostic testing to evaluate O2 instrumentation placement and combustion control performance is strongly recommended at the start of any NOx emissions baseline or postretrofit warranty testing. Even in cases where air infiltration is not an issue, the O2 analyzer reading may not be representative of the duct average. This is a more common problem with tilting tangentially-fired units where a large O2 gradient or "split" between furnace halves can develop for certain burner tilt positions. A large O2 gradient can result not only in above average NOx emissions, as discussed previously, but it can lead to operational problems if not rectified. At one test site, a major slagging episode was traced to operation for an extended period (three days) with the burner tilts in a position that created a very low O2 level on the south side of the furnace. In this case, failure to display the low O2 side (as well as the average) did not alert the operator to the problem since the average O2 was within an acceptable band of the recommended O2 level. Many utilities are installing multiple O2 analyzers in each duct to obtain a more representative average. Combustion diagnostic testing during a NOx test program can aid in defining the best location for these new analyzers. Another problem experienced during NOx testing is non-representative O2 measurements caused by instrumentation equipment problems. Recalibration of the plant O2 analyzers is strongly recommended before conducting any NOx emissions testing. Although a gas sample grid may be installed at the economizer exit specifically for the NOx test program, the operation of the unit after the test program is completed will depend on the existing plant O2 instrumentation. Therefore, the performance of the plant O2 instrumentation should be completely evaluated during the NOx test program. |