REDUCTION OF BOILER STACK PARTICULATE EMISSIONS AND SCRUBBER WATER FLOW USING LOW-SODIUM COAL

Size: px

Start display at page:

Download "REDUCTION OF BOILER STACK PARTICULATE EMISSIONS AND SCRUBBER WATER FLOW USING LOW-SODIUM COAL"

Tiffany Thompson
5 years ago
Views:

1 REDUCTION OF BOILER STACK PARTICULATE EMISSIONS AND SCRUBBER WATER FLOW USING LOW-SODIUM COAL BY: DONALD D. HILL ENGINEERlNG & MAINTENANCE MANAGER FORT MORGAN FACTORY WESTERN SUGAR COMPANY HWY. 144, FORT MORGAN, CO., AND PATRICIA R. FULLER-PRATT MANAGER OF ENVIRONMENTAL AFFAIRS WESTERN SUGAR COMPANY 1700 BROADWAY, SUITE 1600, DENVER, CO., ABSTRACT Two 1947 coal-fired stoker boilers, using a low-pressure mid-1970's design venturi scrubber, were able to demonstrate compliance at one-half the regulated limit and less than 0.1 Ib/MMBTU by converting to a low-sodium coal. The sodium content in coal impacts the particulate emissions and is the cause of higher particulate emissions and higher opacity. Analysis indicates that the sodium content must be less than 2.5 percent, or the ash to sodium ratio must be greater than 40, to be effective in reducing particulate emissions and opacity. The higher ash content provides sites to absorb or attach the sodium in the boiler without overloading the pollution control devices. The increased ash bed also insulates the grates and allows for better combustion efficiency. 305

2 INTRODUCTION Voluntary compliance tests for particulate emissions were performed in October of 1995 at the Western Sugar Company's Fort Morgan sugar beet processing plant. The boilerhouse contains two Babcock & Wilcox stoker-fired boilers. The units were placed in service in 1947 and, except for a short time using natural gas, have used Detroit Stoker stokers as the coal-firing system. Each boiler has an American Standard Series 361 fly-ash collector and an in-house (Great Western Sugar Company) designed low-pressure venturi scrubber. Exhaust gas streams from each boiler are combined in a common stack which has a chevron-type mist eliminator. BOILER STACK TESTS - 22 OCTOBER 1995 Normal intercampaign maintenance was performed on the boiler equipment and the Detroit Stoker field service representative "tuned" the boiler combustion system the week prior to the October 1995 tests. The particulate emission limit is 0.14 Ib.lMMBTU. The first day's test results were under the limit (see Table 1). The second day the limit was exceeded twice and the average for the three compliance runs was 0.16 Ib.lMMBTU. The information was presented to the State of Colorado under the State's program of voluntary disclosure and a second test was scheduled as quickly as possible. Also, several boiler consultants were contacted to observe the compliance retest. TABLE 1 OCTOBER STACK TESTS PARTICULATE EMISSIONS OCTOBER 2 AND 3; 1995 ITEM RUNNO.1 RUNNO.2 RUN NO.3 AVERAGE Boiler No.1 PLUG CONE 285 gpm 280 gpm 280 gpm 282 gpm Boiler No.2 PLUG CONE 280 gpm 260 gpm 275 gpm 275 gpm 277 gpm 253 gpm PARTICULATE EMISSIONS (lb.lmmbtu)

3 RETESTS - 15 NOVEMBER 1995 The coal supplier provided a consultant to review the coal as-delivered and as-fired, and also to observe the boiler operation. Detroit Stoker provided the same technician who was on site in October to "tune" the boilers. He also observed the boiler operation during the tests in order to advise us if any mechanical problems may have developed. A scrubber manufacturer was contacted to observe the venturi scrubber operation during the tests and to inspect it prior to the tests. This individual had been previously employed as a "stack tester" and also observed the stack testing company during the compliance runs. The November tests were run with more scrubber water flow, higher pressure drops across the scrubber, but no "hands-on" operation by any of the consultants. The average of the three compliance runs for particulates was Ib.lMMBTU (see Table 2), which is nearly twice the permitted limit. TABLE 2 NOVEMBER STACK TESTS P ARTICULATE EMISSIONS NOVEMBER 15 AND 16, 1995 ITEM RUN NO.] RUN NO. 2 RUN NO.3 AVERAGE Boiler No.1 PLUG 335 gpm 334 gpm 333 gpm 334 gpm CONE 340 gpm 338 gpm 337 gpm 338 gpm Boiler No.2 PLUG 323 gpm 321 gpm 328 gpm 324 gpm CONE 335 gpm 332 gpm 332 gpm 333 gpm PARTICULATE EMISSIONS (lb. / MMBTU) SODIUM EMISSIONS (lb. / MMBTU) SODIUM OXIDE (% in ASH) During the tests, the opacity of the stack was observed (not using Method 9) by all the consultants and a "blue haze" was evident after the vapor break of the "plume". The coal consultant suggested that the "blue haze" might be due to sodium and was a condition he had observed before with Northern Powder River Basin coal. The scrubber consultant agreed that this was possible, and 307

4 suggested that a sub-micron sodium fume would pass through a low-pressure venturi scrubber. He suggested that we analyze the particulate catch for "sodium". This analysis by the stack testing company averaged lb. Sodium/MMBTU, which is nearly 90% of the permitted limit. An analysis of the coal as-fired for sodium oxide, averaged 7.58% sodium oxide in the ash. When the coal shipments were reviewed for the campaign (see Table 3), the percent sodium in the ash during the October test was around 4.0% and the highest values % to 8.0% -- occurred during the November test. The consultants agreed that the boilers were performing correctly and that the coal was the biggest contributor to the problem. TABLE 3 COAL SIDPMENT DATA FOR CAMPAIGN Date of lb/mmbtu % Ash % Sodium Calculated % Sulfur Shipment Ratio 9/17/95 9, /24/ /29/ /3/ /4/ /11 / / / /22/ /28/ /30/ /30/ / / l / / / l /2/ /7/ / / / /22/ /26/ AVERAGES _.-._---_

5 ENGINEERING TESTS - 21 DECEMBER 1995 A source for ]ow-sodium coal was found in the Southern Powder River Basin and a test burn and compliance test were arranged for December 21, These tests were not full compliance tests but were engineering-level tests only meant to test the "sodium in the coal" theory. Also, these tests were observed by a representative of the Colorado Department of PubEc Health and Environment, as were the November tests. However, no consultants were present. The scrubber water flows were returned to the levels of the October test (see Table 4) and the particulate emissions were an average of 0.09 IbIMMBTU, which is ]ess than 65% of the permit limit of 0.14 Ib.lMMBTU. Unfortunately, we were ten days away from the end of campaign and a full compliance test could not be scheduled until the start of the campaign. TABLE 4 DECEMBER STACK TESTS ("ENGINEERING TESTS") PARTICULATE EMISSIONS DECEMBER 21, 1995 ITEM RUN NO.1 RUN NO.2 AVERAGE Boiler No.1 PLUG 200 gpm 200 gpm 200 gpm CONE 200 gpm 200 gpm 200 gpm Boiler No.2 PLUG 195 gpm 200 gpm 198 gpm CONE 200 gpm 200 gpm 200 gpm PARTICULATE EMISSION S (lb.lmmbtu) CALCULATION MISCONCEPTION Research at another Western Sugar factory regarding opacity suggested that an Ash to Sodium-inthe-Ash ratio of 48 or better at that factory would produce an opacity of about 10%. The consultant suggested a b ~ end of "high" and "low" sodium coals to reduce opacity at that factory. However, that consultant felt that sodium wou1d not contribute much to particulate emissions (which proved incorrect in our testing at Ft. Morgan). The consultant's report attempted to establish the Ash/Sodium ratio by using lb. AshlMMBTU and lb. Sodium/MMBTU. Basically, this is mathematically incorrect. In order to calculate each of these numbers and then take the ratio you have essentially canceled out the units and the ratio is really one (1) divided by the Sodium in the Ash percentage expressed as a number, i.e. 2.5% Sodium in the Ash is equal to Therefore, 1/0.025 = 40. For the Ft. Morgan factory, we developed a 309

6 conservative estimate that a coal with an Ash to Sodium in the Ash ratio of greater than 40, or a Sodium in the Ash content of less than 2.5 %, would achieve compliance. COMPLIANCE AND ENGINEERING TESTS - 22 OCTOBER 1996 A low-sodium Southern Powder Ri ver Basin coal was contracted for the beet campaign. A compliance test was conducted on October 22, 1996 (see Tabl 5). The results averaged Ib.lMMBTU, which is well below the 0.14 Ib.lMMBTU particulate emission limit. TABLES OCTOBER 1996 COMPLIANCE DEMONSTRATION STACK TEST PARTICULATE EMISSIONS OCTOBER 22,1996 ITEM RUN NO.1 RUN NO.2 RUN NO.3 AVERAGE Boiler No.1 PLUG 210 gpm 211 gpm 210 gpm 210 gpm CONE 208 gpm 209 gpm 207 gpm 208 gpm Scrubber Diff. Press. 5.8 in. H 2 O 5.6 in H 2 O 5.9 in H 2 O 5.8 in H 2 O Boiler No.2 PLUG 199 gpm 214 gpm 211 gpm 208 gpm CONE 199 gpm 204 gpm 214 gpm 206 gpm Scrubber Diff. Press 5.6 in. H 2 O 5.6 in H 2 O 5.6 in H 2 O 5.6 in H 2 O PARTICULATE EMISSIONS (lb.lmmbtu) After the compliance tests were completed, the water flows to the scrubber plug and cone and the venturi plug placement were varied in order to establish parameters which could be tested in order to prove compliance using parmetric monitoring for the Title V Operating Permit. A total of eight engineering tests were performed at four different water flows and lor pressure differentials (two tests at each level). ~ The water flowrates for the plug and cone of the venturi were varied from 200 gpm to 150 gpm to 100 gpm and the pressure drop was changed from 5.0, 4.0, to 3.0 in. H 2 0. The idea was to vary the water flow and hold the pressure drop constant to simulate a condition of loss/reduction of water flow (pump belt slippage, pat1ial blockage or some other reason). The pressure drop was varied while holding the water flow constant in order to simulate a loss in air flow (a mechanical problem with the fans or their steam-turbine drivers, a damper problem or some other reason). The variations all worked, except the 3.0 in. H 2 0 pressure drop setting, which required decreasing the water flow by 50 gpm in order to get the pressure to drop even though the venturi plug had been lowered to its lowest point. As the results show (see Table 6), the water flow variation tests had 310

7 particulate emissions of 0.072, and lb.lmmbtu and the pressure drop particulate emission rates were 0.072, 0.09 and lb.lmmbtu. TABLE 6 OCTOBER 1996 "ENGINEERING" STACK TEST PARTICULATE EMISSIONS VARYING WATER IiLOW & PRESSURE DIFFERENTIAL OCTOBER 23 & 24, 1996 VARYING WATER FLOW VARYING PRESSURE ITEM RUN NO.1 RUN NO.2 RUN NO.3 RUN NO.4 Scrubber Water Flow PLUG & CONE #1 307 gpm 214 gpm 414 gpm 300 gpm * PLUG & CONE #2 309 gpm 211 gpm 412 gpm 296 gpm * Scrubber Differential Pressure BOILER N O in. H in H in H in H 2 0 BOILER N O in. H in H in H in H 2 0 Particulate Emissions (lb.lmmbtu) These engineering tests showed that there is a wide operating window for the scrubbers when using a low-sodium coal. It appears that low water flowrate and a small pressure drop will still allow the boilers to operate below their particulate permit limit of 0.14 Ib.lMMBTU. CONCLUSIONS AND BENEFITS The Fort Morgan factory is in compliance with boiler particulate limits when it uses low-sodium coal. The Fort Morgan factory can successfully demonstrate on-going compliance by monitoring the following parameters: COAL SUPPLY Maintain a record of the coal analysis of each shipment of coal to the factory and verify that it is less than 2.5% Sodium in the Ash. (Ash/Sodium ratio of greater than 40). SCRUBBER WATER FLOW Keep the water flow to each venturi plug above 100 gpm and 100 gpm to each venturi cone. These are recorded continuously on a strip chart recorder. 311

8 SCRUBBER PRESSURE DIFFERENTIAL Set the venturi plug at a point to maintain at least 2.5 in. H 2 0 or more of pressure drop for each venturi scrubber. The pressure drops are also continuously recorded on strip chart recorders in the boiler control room. BENEFITS 1. The original scrubber water flowrates had a "design" need for 800 gpm for each scrubber system or 1600 gpm for both, which all ended up in the ash pond. This flow has now been reduced to 200 gpm to for each system or 400 gpm to 500 gpm, a total savings of gpm. 2. By having a thicker ash bed, the combustion air entry points can be redirected to have the undergrate air decreased as the grates are better insulated and the overfire air increased to "lower" the flame and improve 02 addition in the suspended- burning zone. 3. With better air flow distribution, even though the BTU content of the low-sodium coal is less than that of the previous higher-sodium coal, the combustion efficiency has improved and coal consumption is less (about 30 + tons per day). 4. Post campaign boiler cleanup is easier and quicker, as the lower sodium coal does not "slag" as much and the slag that does form appears softer and easier to remove. 312

Part 70 Operating Permit Amendment

Part 70 Operating Permit Amendment Effective Date: February 18, 2016 Facility Name: Scherer Steam-Electric Generating Plant Facility Address 10986 Highway 87 Juliette, Georgia 31046, Monroe County Mailing