OCR Text |
Show APPLICATION OF PREf\1JX COMBUSTION TO HIGH INPUT BURNER SYSTEM second ary fuel jets vvere scaled to provide consistent fl ue gas d i I u tion before rnixing \·"ith the lean prelllix products. To date vve have applied these scaling rules successfully on burners ranging in size frolll 2 x 106 to over 200 x 106 Btu/hr. Prototype and Field Testing The first cOlllplete prototype was tested on a 400 hp fire tube boiler (13.4 x 10h Btu/hr) in North Alllerican 's test laboratory. Both standard burners, typical of conllllercially available units, and the 421]-9 LE burner perfornlance \vere conlpared for sillliJar firing conditions. The results are shown in Figure 6. Standard burner NOx elllissions were 55-70 pPlllV at high fire and 20-25 pPlllV \vith 20% flue gas recirculation. The 4211 prototype delllonstrated significantly inlproved perfornlance, with NOx emissions below -15 pPlllV for all firing rates. ~-\dditionally no change occurred in the flue gas exit telllperature fro III the standard burner to the LE prototype. The use of FGR on the standard burner did ho\yever, increase the stack tenlperature frolll 275 F to 290 F, resulting in a corresponding loss of efficiency. CO elllissions were not detectable vvith the LE prototype. VVith the pilot and fire tube prototype vvork cOlllplete, North Anlerican 's focus changed to building a full size 62.5 x -106 Btu/hr burner for oil field steanl generator applications. The new burner was installed on a Thernlotics design stearn heater with 56 water tubes located on a 8' 11" tube circle. The design provided for direct replacenlent of the existing burner vvith a IllinilllUlll of nevv piping and no change to the blower location. The existing burner tile \-vas rernoved for installation of the larger low NOx reaction challlber. SOllle control S\ 'stelll changes vvere required for fuel/ air ratio control \vith pnnlary clnd secondar\' fuel streanlS. Initial testing began on f\1arch 3, 1994, with continuous, autolllatic operation beginning about three Illonths later. Enlission perfonnance for the stealll generator is shown in figure 7 for several different test configurations. In |