Physical Testing of a Multipoint Ground Flare Burner Utilizing Low BTU Flare Gas

Paper from the AFRC 2015 conference titled Physical Testing of a Multipoint Ground Flare Burner Utilizing Low BTU Flare Gas

Pressure-assisted burners are used within multipoint flares in large numbers to provide smokeless operation without the use of steam or assist air. In the United States of America flares must conform to the regulations in 40 CFR 60.18 and 40 CFR 63.11.; Previous evaluation of flares for the region of stable operation has resulted in the use of exit velocity as the means of determining a stable flame. This method cannot be applied to pressure-assisted flares. The combustion efficiency of flares has also previously been correlated with the combustion zone net heating value and the combustion zone lower flammability limit.; A review of available data reveals that neither combustion zone net heating value nor combustion zone flammability limit are good predictors of combustion efficiency for pressure-assisted flares. Statistics calculated from the data also indicate that the manufacture of the burner head is also significant to determining the combustion efficiency of the flare although all of the data available showed the burners outperformed current EPA requirements.; Based on the prior test experience of others a test was designed and carried out for the Callidus MP4U pressure-assisted flare burner utilizing low heating value vent gas with mixtures of ethylene and nitrogen. The MP4U burner demonstrated stable flame for long duration with an average vent gas heating value of 501 Btu/SCF. The burner did not experience flame instability until a vent gas mixture of 375 Btu/SCF.; It is suggested that combustion zone net heating value and combustion zone lower flammability limit might not be useful in determining both the combustion efficiency and the flame stability of pressure-assisted burners.

Physical Testing of a Multipoint Ground Flare Burner Utilizing Low BTU Flare Gas Matthew Martin, Tim Lowery, Bryan Beck, Kurt Kraus, Chris Maley Callidus Technologies LLC Tulsa, OK USA Email: [e-mail] ABSTRACT Pressure-assisted burners are used within multipoint flares in large numbers to provide smokeless operation without the use of steam or assist air. In the United States of America flares must conform to the regulations in 40 CFR 60.18 and 40 CFR 63.11. Previous evaluation of flares for the region of stable operation has resulted in the use of exit velocity as the means of determining a stable flame. This method cannot be applied to pressure-assisted flares. The combustion efficiency of flares has also previously been correlated with the combustion zone net heating value and the combustion zone lower flammability limit. A review of available data reveals that neither combustion zone net heating value nor combustion zone flammability limit are good predictors of combustion efficiency for pressure-assisted flares. Statistics calculated from the data also indicate that the manufacture of the burner head is also significant to determining the combustion efficiency of the flare although all of the data available showed the burners outperformed current EPA requirements. Based on the prior test experience of others a test was designed and carried out for the Callidus MP4U pressure-assisted flare burner utilizing low heating value vent gas with mixtures of ethylene and nitrogen. The MP4U burner demonstrated stable flame for long duration with an average vent gas heating value of 501 Btu/SCF. The burner did not experience flame instability until a vent gas mixture of 375 Btu/SCF. It is suggested that combustion zone net heating value and combustion zone lower flammability limit might not be useful in determining both the combustion efficiency and the flame stability of pressure-assisted burners. Figure 1- (Top) A typical multipoint ground flare. (Bottom) The burners inside a typical multipoint ground flare. 1. INTRODUCTION Multipoint ground flares (MPGF) are depressurizing devices utilized for the safe disposal of vented gas streams. Figure 1 shows a typical multipoint ground flare. These flares are typically pressure-assisted in that a relatively high upstream pressure of vent gas is supplied to the flare burners and the resulting available air-entrainment and mixing energy is utilized to provide smokeless operation over a wide range of flow rates. This methodology is in contrast to a traditional elevated flare in which fans or steam provide the required motive force and mixing energy to the combustion air stream. Some stages of a MPGF may be steam or air assisted. There is no inherent reason that the mixing technologies cannot be combined within a single flare system. © 2015 Callidus Technologies LLC, All Rights Reserved 1 The process vent gas is distributed by a manifold to a multitude of smaller diameter pipes often referred to as ‘runners'. Each of the flare runners may be staged on or off via a controller depending on the flow rate of gas arriving at the flare. This staging results in the flow arriving at each burner head having adequate pressure to ensure smokeless operation of the flare. The burners are in turn mounted to each runner. Elevated flares are mounted well above grade, among other reasons, in order to provide protection from the significant thermal radiation that can occur from a flaring event. There also may be a sterile zone requirement around the base of the flare in order to ensure safe operation. MPGF burners are mounted 8-10 feet from grade in order to provide low visibility of the flare flames to the surrounding community. The lack of substantial elevation in turn requires another means of mitigation for the thermal radiation. The following modified from "Parameters for Properly Designed and Operated Flares" is used to calculate the NHVcz [1]: Equation 2.1.1 Where: NHVcz (Btu/SCF) = the net heating value of the combustion zone gas Qvg (SCF/hr) = the volumetric flow rate of the vent gas NHVvg (Btu/SCF) = the vent gas net heating value Qs (SCF/hr) = the total steam rate When there is no assist steam for a vent gas the equation reduces to the net heating value of the vent gas. A barrier fence ranging in height from 35' to 60' is typically used to provide protection to surrounding equipment and personnel. In addition to the fence protecting the outside environment from the flaring within it shields the burner flames from wind without. For LFLcz the equation adjusted for unit consistency from the same paper follows; the percent in the denominator is changed to mole fraction: The process vent gas composition supplied to the burners is dependent on the service to which the flare is attached. It has been long known and generally observed that the ability of the burner to maintain flame depends on the fuel gas composition supplied to a burner. The addition of inert gas into the flame zone either being premixed with the fuel or through the overuse of steam has been recognized to reduce the robust burning of the vent gas. Equation 2.1.2 Regulations within the United States of America (USA) use the heating value of the vent gas as the independent value in the design equations for assisted and unassisted flares with resulting limiting velocity criteria for flare design. The pressure supplied an MPGF burner results in choked flow and an exit velocity that is well above the regulated maximum velocity for a flare. MPGF designs therefore do not meet federal regulations based on this criteria and one must file for an Alternative Means of Emissions Limitation (AMEL). The filing of an AMEL based on the exit velocity also causes one to question the stability of the burner flames based on the heating value of the gas for these alternative designs. Where: At a minimum the flare burner must demonstrate that it provides a stable flame over the required operating range, the required destruction efficiency for the application - commonly 95% or 98% - and does not produce visible emissions except as allowed for 5 minutes during any two consecutive hours. 2. PRELIMINARIES AND REVIEW 2.1 Combustion Zone Net Heating Value and Combustion Zone Lower Flammability Limit The combustion zone net heating value (NHV cz) and combustion zone lower flammability limit (LFLcz) have been shown to correlate with combustion efficiency (CE) [1]. Interestingly the usage of these parameters appears to have been conflated with flame stability even though this may not have been the original intent. It has been shown through publicly available vendor test data by Callidus and one other manufacturer that the destruction efficiency for pressure assisted flares is greater than 99% when the flame is present but that LFLcz and NHVcz are not adequate to describe the presence of flame. LFLcz = lower flammability limit of combustion zone gas n= number of components in the combustion zone gas i= index of the individual component xi (mol frac.) = mole fraction of component i in the combustion zone Ne,H2O= Coefficient of nitrogen equivalency for H2O xH2O= mole fraction of water in the combustion zone Ne,H2O= Coefficient of nitrogen equivalency for H2O xCO2= mole fraction of water in the combustion zone Equation 2.1.3 Where: CE(%) = combustion efficiency CO2concentration = concentration of CO2 CO2concentration = concentration of CO Hydrocarbonconcentration = concentration of Hydrocarbon 2.2 Exit Velocity and Flame Stability MPGF burners achieve smokeless operation by utilizing the pressure energy to enhance mixing of the vent gas with the air. The exit velocity of a flare is limited to 400 ft/s or less for lower heating value gas per 40 CFR 60.18 and 40 CFR 63.11 [2]. At design rate a MPFG flare has choked flow and a Mach 1 exit velocity and so cannot comply with this requirement. It is noteworthy that in the review of the available literature that pressure assisted flares appear to be stable as long as the heating © 2015 Callidus Technologies LLC, All Rights Reserved 2 value of the flare gas exceeds a certain threshold even though the velocity criteria is violated [3, 4, 5]. This threshold of heating value for stability is not constant across gas composition or burner type. burner under similar conditions. Although steam capability is present in neither case is the steam utilized. One can see that two burners of similar description can perform in a substantially different manner. 2.3 Burner Operation and Flame Stability The burners that flamed-out in the prior emissions testing did so under conditions wherein the NHVcz and the LFLcz were well above previously suggested limits. According to EPA document EPA-HQ-OAR-2014-0738-0002 the flame out occurred between 2 and 17 minutes[6]. This suggests that pressure assisted burners may be able to maintain flame for some time even with a fuel gas that may ultimately make the burner unstable and that short duration tests of a minute or less are not adequate to determine flame stability. 2.4 Burner Design and Flame Stability As stated in the Federal Register document 80 FR 8023 it is apparent that "flare head design can influence the flame stability curve" [2]. This can be confirmed through physical testing of burners wherein small changes in the design can have a large effect on the performance and flame stability. The readily apparent outward characteristics of a burner such as a multitude of arms and a plurality of fuel gas ports may not be adequate to ensure equivalent operation. For example the Callidus MP4U flare head has specific features built into the burner head designed to stabilize the flame in a reliable and consistent manner. A multitude of design choices work in concert to ensure the stability of the flame such that superficially producing a burner of the same general shape does not ensure the same operation. Figure 2 shows a rendering of this burner head. 2.5 Burner Design and Smokeless Operation Even with equivalent fuel gas and pressure smokeless operation is not ensured. The burner design has a significant impact on the pressure at which smokeless operation occurs for a burner. The upper left hand image of Figure 3 shows a multipoint burner design of 1991 which has spider arms and a central hub. The upper right hand corner shows the resultant flame and smoke. The lower left hand image shows a Callidus MP4U multipoint burner design of 2014 which has spider arms and a central hub. The image in the lower right hand side shows the flame of the 2014 Figure 3 - (Top Row) Older style burners of Callidus manufacture. (Bottom) Newer Callidus MP4U burners. 2.6 Review of Prior Test Data for NHVcz and LFLcz Utilizing the publicly available information from the USA Environmental Protection Agency (EPA) docket EPA-HQ-OAR2014-0738 and the technical memorandum of EPA-HQ-OAR2014-0738-0002 in combination with the Callidus test data a review was performed for CE sensitivity to NHVcz and LFLcz. The data table used for this analysis is given in the appendix. A general linear model ANOVA was used to analyze the variance of CE with respect to NHVcz and LFLcz across all manufacturers and tests. The p value for NHVcz and LFLcz are 0.762 and 0.809 respectively, suggesting that these input values have no statistical significance on the CE response. The R2 statistic for the resulting model is 4.86% and the s statistic is 0.37. All of these measures indicate that with the given data NHVcz and LFLcz cannot be used as parameters to determine CE for pressure-assisted flares. Alternatively using nitrogen and LFLcz yields slightly better results with an R2 statistic of 19.2 and an s statistic of 0.34. The lack of readily available input variables to explain CE may not be significant: this observation correlates well with the previous observation by Callidus and others that if a flame is present the CE of a pressure-assisted flare is high. Interpreting the data in another way by plotting it against the previous data collated by the EPA reveals another interesting observation: it is not surprising that pressure assisted multipoint flares are operating with high CE because the condition of operation is above the previously established threshold for high CE. Figure 4 shows the pressure assisted data plotted against the data for flares conforming to 40 CFR exit velocity regulations. Figure 2 - The Callidus MP4U Flare Burner © 2015 Callidus Technologies LLC, All Rights Reserved 3 order to test for the sensitivity of flame stability to vent gas exit velocity. Each run was repeated three times resulting in a total of six runs. 3.2 Data Collection For each test run video recording was used to provide a record of flame stability. Vent gas composition was verified by a third party using an online gas chromatograph. Burner head pressure, manifold pressure, vent gas temperature, ethylene flow rate, nitrogen flow rate and ambient temperature were also recorded. All quantitative data aside from ambient temperature were collected continuously at an average of 3 second intervals. 3.3 Test Configuration Figure 4 - Pressure-assisted flare data plotted against data from [1]. The blue diamonds represent stable steam or air assisted flares. The red boxes represent unstable steam or air assisted flares. The orange circles represent stable nonCallidus pressure assisted burners. The yellow circles represent unstable non-Callidus pressure assisted burners. The light green triangles represent stable Callidus MP4U burners. 2.7 Review of Prior Data for Burner Type The publicly available pressure assisted flare data inputs are not held constant across manufacturers. However the was still analyzed for sensitivity to burner type (manufacturer). The resulting box plot of Figure 5 and the corresponding statistics demonstrate that the Callidus MP4U achieves higher destruction efficiencies when compared to other burners given the available data. Comparing the Callidus MP4U test data to that of the other flare burner using a two-sample t-test results in a rejection of the mean CE being the same with a p statistic of 0.02. The Callidus MP4U burner performance is different from the other burner. The estimated CE increase for the Callidus MP4U is 0.26%. The Callidus MP4U test data utilized only high heating value gas but did include both steam assisted and pressure-only assisted runs. The data for the other burner had a large range in vent gas heating value. Performing a two-sample t-test again against only the high heating value data from the non-Callidus burner gives similar results. With a p-value of 0.022 the differential in performance increases in favor of the Callidus MP4U to a 0.35% CE increase to 99.91% versus 99.56% for the alternative. The box plot of Figure 5 also shows the MP4U to be more consistent in performance with less variation between the data points. Ethylene was supplied by a high pressure tube trailer which was controlled by a 2 inch ball valve and 1 inch needle valve. The flow was measured across a known orifice with pressure and differential pressure readings. These pressures were in turn used to calculate the flow. Nitrogen was supplied by an onsite liquid nitrogen tank which was vaporized atmospherically. The nitrogen flow was controlled by two separate 2 inch lines, each with a 2 inch ball valve and 1 inch needle valve used for control. Each 2 inch nitrogen line utilized separate orifice plates, pressure and differential pressure meters. The vent gas was mixed in a 4 inch upstream of the flare manifold. The vent gas was burned with a single burner head having 2 square inches of exit area. The MP4U burner used a natural gas pilot for flare ignition which was subsequently disconnected during operation. 4. RESULTS 4.1 Stability Testing at 500 Btu/SCF NHVcz Stable operation of the burner flame was demonstrated on each of the six runs with an average vent gas NHVcz of 501 Btu/SCF. The resulting summary data is given in Table 1. The nitrogen to ethylene ratio was then intentionally increased to reduce the heat content until the flame became unstable and was extinguished. For both the high pressure and low pressure conditions the flame became unstable and was extinguished at approximately 375 Btu/SCF but did not exhibit instability prior to achieving said mixture. Pictures of the burner in operation are given in Appendix II. 3. APPROACH AND TEST PLAN 3.1 Objective The objective of for the physical testing of the Callidus MP4U flare burner was to establish stable operation utilizing a vent gas with less than 800 Btu/SCF NHVcz. A series of tests were performed utilizing a target vent gas heat content of 500 Btu/SCF utilizing a mixture of ethylene and nitrogen. Due to previous observation of flame-out for other manufacturers flare tips after a significant time with apparently stable operation a trial duration of 20 minutes was selected for each test run. The vent gas pressure upstream of the burner head was altered between 15 and 5 Psig in Figure 5 - Comparison of CE for two different burner designs. © 2015 Callidus Technologies LLC, All Rights Reserved 4 Table 1 - Results from the Callidus MP4U low heating value burner test. enough data to perform a rigorous analysis for differences because not enough of the operating conditions overlap. Figures 6 and 7 show Zabetakis plots from Bureau of Mines Publication 627 with the stoichiometric vent gas compositions of the relevant flare test operating conditions overlaid [7]. Figure 6 shows the Callidus MP4U stable point with a green dot and the unstable point with a red dot. Figure 7 shows two of the unstable points for the other burner. One noticeable difference is that the nose of the Zabetakis plot extends to 50% added inert for ethylene mixtures and only 43% added inert for propylene mixtures; there is no clear scaling rule that can be derived from this observation that can be applied between the point of flame-out for each burner. Table 3 - Comparison of stability limiting vent gas compositions for different burner styles and tests. 4.2 Further Analysis of Results Table 2 shows a comparison of flame stability for the MP4U versus another type of burner. Utilizing NHVcz and LFLcz one cannot discriminate whether the flame will be stable; the Callidus MP4U is consistently stable with a lower NHVcz and LFLcz than the other type of burner. The Callidus MP4U was tested using mixtures of ethylene and nitrogen while the other burner was tested utilizing mixtures of propylene and nitrogen. This suggests that the determining factor may be dependent on the flammable constituents of the vent gas and not the burner type. However, the lower flammability limit of ethylene in air is 37.5% higher than that of propylene suggesting that the other style of burner should have been advantaged from using the alternate fuel mixture as opposed to the MP4U. Table 2 - Comparison of stable operating conditions for different burner styles. Burner Type MP4U MP4U MP4U MP4U MP4U MP4U Other Other Other Other Stable Flame? TRUE TRUE TRUE TRUE TRUE TRUE FALSE FALSE FALSE FALSE NHVcz LFLcz 481 8.44 511 7.95 508 7.99 499 8.14 497 8.16 509 7.97 595 7.6 589 7.8 746 6.6 650 7.6 Table 3 shows a comparison of the vent gas expressed in terms of calculated composition instead of NHVcz or LFLcz. Here the data seems to be more consistent. The hydrocarbon to inert ratio expressed in terms of volume percent is much closer in composition than in immediately evident from partitioning performance by burner type or NHVcz. The Callidus MP4U experienced stability issues with a vent gas hydrocarbon content of 25% while the other burner experienced stability issues with an average vent gas hydrocarbon content of 28%. There is not Comparison of Vent Gas at Burner Flammability Limit Callidus MP4U Other Compound Volume % Compound Volume % C2H4 25.00% C3H6 28.32% N2 75.00% N2 71.68% LHV (Btu/SCF) LFLcz NHVcz (Btu/SCF) LFL MW NHVlfl-vg NHVratio 375 LHV (Btu/SCF) 11.00% LFLcz 375 NHVcz (Btu/SCF) 2.75% LFL 28.03 MW 10.31 NHVlfl-vg 9.09 NHVratio 618 7.06% 618 2.00% 32.00 12.36 14.16 5. CONCLUSIONS Examination of previous test data suggests the following prerequisites for rigorous testing and evaluation of a pressureassisted flare burner: 1) 2) 3) 4) 5) The test runs must be of sufficient duration to detect flame-out. Previous experience suggests this might be in excess of 17 minutes. NHVcz and LFLcz are not adequate markers of flame stability for pressure-assisted flares; different burners will lose flame stability at different values of these parameters. Setting an unduly high NHVcz and LFLcz in order to ensure stable combustion may not ensure stable combustion for an untested burner and may have undesirable process consequences such as the need for assist fuel. Flare burners of different but similar design have different flame stability points and different combustion efficiencies. Physical testing is recommended to disambiguate the performance characteristics of different burners. The Callidus MP4U MPGF burner has proven stable operation for mixtures of 500 Btu/SCF at 5 and 15 Psig vent gas supply pressure for durations of 20 minutes each over multiple trials. The minimum stable operation was just above 375 Btu/SCF vent gas heating value. © 2015 Callidus Technologies LLC, All Rights Reserved 5 6. REFERENCES [1] 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 [2] ‘Receipt of Approval Requests for the Operation of PressureAssisted Multi-Point Ground Flare Technology'. Federal Register / Vol. 80, No. 30, 2015 [3] ‘Performance Test of Steam-Assisted and Pressure Assisted Ground Flare Burners with Passive FTIR - Garyville', EPA Document EPA-QA-OAR-2014-0738-0003, Originally 2013 referenced in 2015 [4] ‘Dow Chemical Test Report', EPA Document EPA-HQOAR-2014-0738-0008, Originally 2013 referenced in 2015 [5] ‘Comments submitted for Receipt of Approval Requests for the Operation of Pressure-Assisted Multi-Point Ground Flare Technology, Docket ID No. EPA-HQ-OAR-2014-0738 at 80 Fed. Reg. 8023', EPA Document EPA-HQ-OAR-2014-0738003, 2015 Figure 6 - Zabetakis plot showing stoichiometric mixtures of for the Callidus MP4U burner. The green dot represents stable flame while the red dot represents unstable flame. [6] Bouchard, Andrew, ‘Review of Available Test Data on Multipoint Ground Flares', EPA Document EPA-HQ-OAR2014-0738-0002, 2015 [7] Zabetakis, Michael, ‘Flammability Characteristics of Combustible Gases and Vapors', Bureau of Mines Bulletin 62, 1965 Figure 7 - Zabetakis plot showing unstable operation for another style of burner. Red dots represent unstable flame. © 2015 Callidus Technologies LLC, All Rights Reserved 6 7. Appendix I - Data Table Test Condition PA1 PA1 PA1 PA1 PA1 PA1 PA1 PA2 PA2 PA2 PA2 PA2 PA2 P1H1 P1H2 P1H3 P1L1 P1L2 P1L3 P2H1 P2H1B P2H2 P2H3 P2L1 P2L1b P2L2 P2L3 P3H1 P3H2 P3H3 P3L1b P3L2 P3L3 C1 C2 C3 C4 C5 C6 C7 Stable Flame? YES YES YES YES NO YES NO YES YES YES YES YES YES YES YES YES YES YES YES NO YES YES YES NO YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES NHV 906 915 917 591 595 607 589 920 906 903 611 600 600 2170 2117 2149 2133 2134 2132 746 771 788 770 650 699 717 718 720 712 715 706 705 721 1499 1190 1499 1499 1499 1098 1499 LFL 5 4.9 4.9 7.7 7.6 7.6 7.8 4.9 4.9 5 7.4 7.8 7.8 2.4 2.4 2.4 2.4 2.4 2.4 6.6 6.3 6.3 6.3 7.6 6.9 7 7 6.9 6.9 6.9 7 7 6.9 2.75 3.46 2.75 2.75 2.75 3.76 2.75 CE 98.7 100 100 98.5 99.6 99.8 100 98.9 98.8 100 99.8 99.4 99.3 100 99.8 99.8 99.7 99.9 99.5 99.98 99.9 99.8 99.8 99.74 99.7 99.8 99.6 99.8 99.8 99.6 99.9 99.8 99.7 99.96 99.94 99.95 99.97 99.72 99.82 99.99 N2 0.016 0.012 0.013 0.365 0.354 0.356 0.372 0.016 0.016 0.016 0.337 0.376 0.377 0 0 0 0 0 0 0.634 0.62 0.621 0.621 0.684 0.65 0.655 0.658 0.524 0.526 0.523 0.531 0.533 0.524 0 0 0 0 0 0 0 CO2 0.006 0.011 0.008 0.003 0.006 0.004 0.004 0.006 0.006 0.007 0.005 0.004 0.005 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 CH4 0.937 0.931 0.935 0.605 0.617 0.612 0.595 0.935 0.933 0.935 0.633 0.592 0.591 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.243 0.246 0.249 0.241 0.239 0.242 0 0 0 0 0 0 0 C2H4 © 2015 Callidus Technologies LLC, All Rights Reserved 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 C2H6 0.038 0.042 0.039 0.025 0.021 0.026 0.027 0.039 0.04 0.037 0.022 0.026 0.026 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 C3H8 0.003 0.004 0.004 0.002 0.002 0.002 0.002 0.004 0.004 0.004 0.003 0.002 0.002 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 C4H6 0 0 0.001 0 0 0 0 0.001 0.001 0.001 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 C3H6 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 0.366 0.38 0.379 0.379 0.316 0.35 0.345 0.342 0.233 0.228 0.229 0.227 0.228 0.234 0 0 0 0 0 0 0 7 8. Appendix II - Pictures of a Callidus MP4U in Operation From Left to Right: Less than 500 Btu/SCF, 500 Btu/SCF, and 1500 Btu/SCF © 2015 Callidus Technologies LLC, All Rights Reserved 8