Industrial-Scale Development and Testing of an Elevated Flare Tip

Paper from the AFRC 2017 conference titled Industrial-Scale Development and Testing of an Elevated Flare Tip

A high steam efficiency flare tip aides in meeting the requirements of the United States 40 CFR 63.670 rule and provides improved plant economics. To develop a new flare tip computational fluid dynamics (CFD) was used to screen for optimal designs and physical testing was used to confirm the performance.; To the author's knowledge a unique element of this test program is a comparison to previous generation of steam flares through direct physical testing. Results of the CFD studies and test data are given.

Industrial-Scale Development and Testing of an Elevated Flare Tip Matthew Martin Honeywell UOP Callidus Technologies LLC Tulsa, OK USA Email: [e-mail] ABSTRACT A high steam efficiency flare tip aides in meeting the requirements of the United States 40 CFR 63.670 rule and provides improved plant economics. To develop a new flare tip computational fluid dynamics (CFD) was used to screen for optimal designs and physical testing was used to confirm the performance. To the author's knowledge a unique element of this test program is a comparison to previous generation of steam flares through direct physical testing. Results of the CFD studies and test data are given. Figure 1 - UOP Callidus flare test facility in Beggs, OK USA 1. INTRODUCTION In the United States of America, major source refiners must comply with new operational and monitoring standards for flares as regulated in 40 CFR 63.670 [1]. In China, the 2016 Environmental Tax established new fees that are applicable to flare emissions [2]. Worldwide, regulatory, community impact and corporate citizenship are driving a reduction in flaring and improved performance when flaring is required. The USA regulations are targeted at maximizing the destruction and removal efficiency (DRE) of flares on the hazardous air pollutants (HAP) contained in the vent gas. It has been shown that "over steaming" reduces the DRE of a flare flame [3] The visible emissions requirements of 40 CFR 63.670 are that there be no visible emissions (smoke) when the vent gas flow rate is below the smokeless capacity and regulated material is routed to the flare. If smoke occurs for 5 minutes during any 2 hours, it is considered a deviation [1]; flares that require less steam for smokeless operation reduce the probability of smoking when trying to minimize steam consumption. The heating value in the combustion zone gases must also be held above 270 Btu/SCF per the same regulation [1]. This net heating value includes any assist media (steam or air) as opposed to previous regulations. Due to changes in vent gas composition and flow, particularly during transition from standby operation with purge to a higher flow rate, the addition of costly assist fuel may be required to prevent violating this rule. A reduction in steam usage also reduces the operating cost of the flare. The operating cost for standby steam is further increased when one accounting for additional assist gas required. For these © 2017 Callidus Technologies LLC, All Rights Reserved 1 reasons - reduction in visible smoke, minimization of steam consumption, minimization of assist fuel, and maximization of DRE - a flare that requires lower steam usage, or steam-tohydrocarbon ratio (SHC) is desirable. 2. WORK PROCESS A typical work process for developing new combustion technology was adopted utilizing Computational Fluid Dynamics (CFD) modeling of baseline and candidate designs, subscale testing, and in this case industrial scale testing at the UOP Callidus flare test facility shown in Figure 1. Steady-state CFD was used calculate baseline performance of known flare tip designs and to screen new candidate designs for key performance factors. The top performing designs from the screening process were carried forward into physical testing. Industrial scale physical testing was performed for both baseline and candidate designs using both paraffinic and olefinic vent gas. Optimizations were made to the top performing designs to further lower the steam requirements for smokeless operation and improve the maximum vent gas flow rate at which the flare tip could maintain a smokeless flame. 3. COMPUTATION FLUID DYNAMICS RESULTS Currently, CFD modeling of flare flames for smokeless operation is qualitative in nature due in part to the fact that correlating the visual occlusion of the smoke with the CFD predicted soot concentration is difficult. Nonetheless the soot models available in Ansys Fluent were used as an indicative screening parameter for candidate designs. The CFD results did in fact trend with actual smokeless operation as shown in Figure 2 wherein the first candidate flare design labelled "Early Development Flare" underperformed when compared to the Callidus IS3 flare tip. 4. PHYSICAL TEST RESULTS Figure 3 shows images of a typical test run designed to find the minimum SHC for a given flare tip. The upper picture shows the flare at a rate where no smoke is produced while the bottom picture shows a ratio of steam to hydrocarbon at which smoking operation has fully developed. Some of these tests were performed with vent gas flow rates far greater than 100,000 lb/hr. The flare tip designs in general were found be sensitive to mechanical detail; two flares that appear substantially similar can have markedly different SHC performance. Figure 4 shows a ranked plot of SHC required for smokeless operation for a commercially available non-Callidus flare tip, the Callidus IS3 flare tip, and various iterations of research and development flare tips designated R0-R7. Again, the wide variation in performance showed that specific design elements of flare tips have a large impact on smokeless SHC requirements. Additionally, the wide variation in SHC performance suggests that physical testing adds rigor to flare performance assessment that cannot be achieved solely through calculation methods or indicative rules. The effect of the application of combustion expertise, physical testing resources and CFD modeling is evident in the performance improvement. The R7 case exhibited a SHC ratio that is 49% lower than that of current fielded commercially available flare tips. 5. CONCLUSIONS While properly executed CFD modeling is indicative of smokeless performance, physical testing or field experience is likely still required to quantify performance, at least in the base case. The integrated soot production from the CFD simulated cases trended correctly with the actual physical test performance for smokeless rate, but the CFD model overpredicted the worst case smokeless operation when compared to the physical test of the same configuration. Calculated or Measured Vaue / Callidus IS3 Value Normalized CFD Soot Prediction and Normalized Observed SHC Ratio for Smokeless Performance 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1 0.9 Callidus IS3 Published Competitive Flare Soot Production, CFD Early Development Flare Observed SHC Ratio Figure 2 - Normalized CFD Predicted Soot Production and Observed SHC Ratios from Physical Testing. Higher numbers represent worse smoke production from the flare. © 2017 Callidus Technologies LLC, All Rights Reserved 2 The SHC performance of the currently commercially available Callidus IS3 compares favorably with other technology based on publicly available test data from tests of similar scale; The IS3 exhibits an 11% reduction in steam consumption. It is possible to reduce steam requirements for smokeless operation significantly from that required by current commercially available flare tips. The best performing flare tip during this test program had a 49% reduction in steam consumption compared to widely commercially available technology. Future work includes further optimization of the improved steam flare technology as well as optimization of other key performance criteria. 6. REFERENCES [1] ‘2016 Code of Federal Regulations, Title 40, Volume 12, Section 63.670'. https://www.gpo.gov/fdsys/pkg/CFR-2016title40-vol12/pdf/CFR-2016-title40-vol12-sec63-670.pdf. Originally July 1, 2016. Referenced July 2017. [2] ‘Environmental Protection Tax Law of the People's Republic of China'. English Translation of: http://www.npc.gov.cn/npc/xinwen/201612/25/content_2004993.htm. Originally December 25, 2016. Reference July 2017. [3] Author not listed, ‘Parameters for Properly Designed and Operated Flares', Report for Flare Review Panel, Prepared by U.S. EPA Office of Air Quality Planning and Standard, 2012 Figure 3 - Tests for minimum steam usage. (Left) shows the flare flame in a non-smoking state. (Right) shows the flare flame in a smoking state. Mass Flow of Steam / Mass Flow of Hydrocarbon Steam to Hydrocabon Ratio Required for Smokeless Performance vs Flare Tip Type 0.5 0.4 0.3 0.2 0.1 0 R0 R2 R1 R3 Non-Callidus Tip R4 Callidus IS3 R5 R6 R7 Flare Tip Designation Figure 4 - Ranked plot of steam to hydrocarbon ratio requirements for smokeless performance with like vent gas compositions. © 2017 Callidus Technologies LLC, All Rights Reserved 3