| OCR Text |
Show MID-80's USEPA STUDY of the EFFICIENCY of INDUSTRIAL FLARESthe REAL STORY from the PRINCIPAL INVISTIGATOR AFRC/JERC PACIFIC RIM COMBUSTION SYMPOSIUM TOTeM37, INDUSTRIAL FLARES KAANAPALI BEACH, MAUI, HI SEPTEMBER 2010 JOHN H. POHL ENERGY INTERNATIONAL LAGUNA WOOD, CA, USA OUTLINE • • • • PRE 1984 WORK FLAME STRUCTURE/EMISSIONS EXPERIMENTAL TECHNIQUES DATA - FLARE TIP SIZE AND DESIGN - HEATING VALUE - GAS COMPOSITION - STEAM TO FUEL (SFR) FLARE MEASUREMENTS FLARES ARE DRIVEN MOSTLY BY BOUYANCE CE OR DE IS DETERMINED BY MASS BALANCES AIR DRAG CREATES EDDIES EDDIES BECOME OVER DILUTED THE GAS MIXTURE BECOMES OUT OF THE FLAMMABILITY RANGE, AND THE FLAME IS EXTINUISHED, PRODUCING EMISSIONS • DILUTION IS USUALLY DETERINED BY A TRACER, SO2 • • • • • ACCEPTED FLARE MEASUREMENT TECHNIQUES • POINT SAMPLES, QUESTIONABLE • HOOD, LIMITED TO SMALL SIZES~3 INCHES • WIND TUNNEL, LIMITED TO SMALL SIZES~6 INCHES • CALCULATED ENTRAINMENT, DIFFICULT • TRACER, ACCEPTED TECHNIQUE DISTANCE ABOVE FLARE HEAD (ft) COMBUSTION EFFICIENCY (PERCENT) LOCAL COMBUSTION EFFICIENCY RADIAL DISTANCE (ft) EER PILOT FLARE CONDITIONS • 1/16-12 INCH TIPS - PIPE - COMMERCIAL • • • • • • • • • • MOSTLY 56% C3H8 IN N2 UP TO 24,000 FT3/HR SO2 TRACER ~7 FT3/HR 0.2-400 FT/SEC 90,000 BTU/HR, 10,000-200,000 BTU/FT2-HR ENTRAINMENT RATE 100-2000 Re~800-2x10(5), Ri~3x10(-5)-8x10(2) KODAK 7239, 10-100 mm LENS 100-1000 FRAMES/SEC CE~98-99 % FLARE STRUCTURE EER FLAME 12 INCH COMMREIAL TIP MEASURING EMISSION FROM FLARE FLAMES AIR DRAG MXout AIR ENTRAINMENT Df=50-2000 Df=Ym-Yaf/Yb-Ym EDDY ROLL-UP FLAME ENEVLOPE EDDY~D, f~5 MS Mxin, Txin D=TIP DIAMETER FLARE FLAME STRUCTURE • BOUYANT TURBULENT DIFFUSION FLAME • FLUCATIONS - RADIAL GRIEDENT - EDDIES • TEMPORAL~5 MS • SIZE~HEAD SIZE • QUENCHING POST-1986 FLARE WORK • KUWAIT OIL FIRES - W.J. SHAFTER • LF GAS, (U. Q., BRISBANE CITY, AUS) - ENERGY INTERNTIONAL - FOR ULTRASYSTEMS • INTERNATIONAL FLARE CONSOTIUM - ENERGY INTERNATIONAL - VIRGINIA TECH Richardson Number (gd/v2) SEEBOLD SFLARE PLOT Ri vs Re PLOT Reynolds Number (pdv/u) DESTRUCTION EFFICIENCY (%) GAS DESTRUCTION EFFICIENCY HHV/(MINIMUM HHV FOR STABILITY) FLAME LENGTH • CORRELATION - HOTTEL - HAWTHORNE - COMBUSTION INSTITUE CA 1948 FLAME LENGTH + LIFT-OFF (ft) API 521 FLAME LENGTH HEAT RELEASE (BTU/HR) CONCLUSIONS FROM EER WORK • IF STABLE, COMBUSTION EFFICIENCY >98 % • STABILITY DEPENDS ON GAS (MOSTLY C3H8N2) • SCALABLE IF LARGER THAN 3 INCHES • FASTER BURNING COMPONENTS ARE MORE STABLE, OTHER GASES TESTED >99 % CE • AIR ASSIST ~0.8 >99 % CE • SINGLE PILOT IMPROVES STABILITY ASSISTED FLARES HOOD COMBUSTION EFFICIENCY (%) STEAM ON GLOBAL EFFICIENCY LB STEAM/LB PROPANE MAXIMUM STABLE VELOCITY (ft/sec) AIR-ASSISTED STABILITY 1/(1+(pv)air/(pv)gas POHL POST-1986 FLARE LIKE WORK • KUWAIT OIL FIRES - W.J. SHAFTER • LF GAS, (U. Q., BRISBANE CITY, AUS) - ENERGY INTERNTIONAL - FOR ULTRASYSTEMS • INTERNATIONAL FLARE CONSOTIUM - ENERGY INTERNATIONAL - VIRGINIA TECH KUWAIT OIL FIRES LANDFILL GAS FLARE ROAST STUDENT COMBUSTION EFFICIENCY ROGAN ROAD LANDFILL GAS FLARE 100 90 80 70 60 50 40 SAMPLES CONT 0 M CONT 1 M BAG 1 M OVERALL EFFICIENCY GREATER THAN 87.1 % 30 20 10 0 -3 -2 -1 0 1 2 3 COMPOSITIONS OF UNBURNED SPECIES IN BAGS BELOW DETECTABLE LIMIT. BAG POINTS ARE MINIMUM COMBUSTION EFFICIENCY. MAXIMUM MONOAROMATIC EMISSIONS ROGHAN ROAD LANDFILL GAS FLARE 0.1 COMPOUND BENZENE TOLUENE E BENZENE XYLENE 0.01 0.001 0.0001 MAXIMIM OVERALL TOLUENE EMISSION 0.034 ug/m3LFG -3 -2 -1 0 1 2 3 ONLY TOLUENE ABOVE QDL. OTHER BACKGROUND LEVELS TAKEN AS 0. OTHER POST 1986 WORK • • • • CMA, DAVIS, ET AL. , H2 ADDEMENT EER, WALSH, ET AL., H2 FLARES IFC, CANMETENERGY McKITTERAGE, H2 ADDITIVE TO FLARES CONCLUSION • > 3 INCH SCALABLE • COMBUSTION EFFICIENCY (CE)>98 % • DEPENDES ~ON HEATING VALUE OF THE GAS AND JET VELOCITY OF THE HEAD • FLAME LENGTH, HOTTEL & HAWTHORNE, API 521, L~Q • OTHER EMISSIONS NEEDED DATA • CE & DE • HIGHER ORDER REACTIVE HYDROCARBONS HORC'S - FOMALDEHYDE CH2O - REACTIVE HC, H2C=CH2, UNSATURATED COMPOUNDS • VERIFICATION OF REMOTE SENSING TECHNIQUES, TO IMPROVE EMISSION CONTROL IFRF CONTRIBUTION TO FLARE RESEARCH • STARTED BY HOTTLE and THRING in 40's - TO INVESTIGATE RADIATION from LARGE INDUSTRIAL FLAMES - ORIGINALLY on SITE of Royal DUTCH STEEL in IJMUDEN, HOLLAND • NEEDED DATA - EFFECT OF WIND - NEED RELABLE EMISSION TESTS on SCALABLE FLAMES - EVALUATE REMOTE TESTING of FLARES |
